Generator Cycling due to High
Penetrations of Wind Power
by
Niamh Troy
A thesis submitted to the
School of Graduate Studies
in fulfilment of the
requirements for the degree of
Doctor of Philosophy
in the
School of Electrical, Electronic and Communications Engineering
University College Dublin, Ireland
Supervisor of Research: Prof. Mark O’Malley
Co-Supervisor of Research: Dr. Damian Flynn
Nominating Professor: Prof. Mark O’Malley
August 2011
Abstract
Power systems have changed considerably in recent years. The introduction of deregulation has brought about competitive electricity markets, forcing generators to operate
in a more flexible manner. Coupled with this, the rapid integration of wind power
world-wide has introduced increased levels of variability and uncertainty into power
system operation. This has led to generators being started up and shut down, ramped
and operated at part-load levels more frequently, in order to meet an increasingly
variable net load (load minus wind generation) and respond to unexpected net load
changes. As base-load units are designed to achieve maximum fuel efficiency they tend
to have limited operational flexibility and consequently this type of cycling operation
results in serious degradation of plant equipment through various mechanisms such
as thermal fatigue, erosion, corrosion, etc. leading to more frequent forced outages
and reduced plant lifetime. Increased costs for base-load generators will also result
from cycling operation, the most apparent being increased operations and maintenance
(O&M) and capital costs resulting from deterioration of the components. However,
fuel costs, environmental penalties and income losses will also arise. Quantifying these
costs is challenging given the vast array of components affected and the time delay that
is typical between cycling operation occurring and the damage manifesting itself. The
uncertainty surrounding cycling costs can lead to these costs being under-estimated by
generators, which in turn can lead to increased cycling.
I
This thesis examines how the operation of base-load units, coal and combined-cycle
gas turbines (CCGTs) in particular, are impacted with increasing penetrations of wind
generation on a system. The technical characteristics of these units, such as their startup times and contribution to system reserve requirements, are shown to influence the
type and level of cycling that will be experienced. Despite collective agreement that
more flexible generation is needed to support the variability and uncertainty of wind
generation, it is shown here that paradoxically it is the most inflexible generation (i.e
coal plants) that are the most rewarded as wind generation increases.
Having identified that CCGT units are severely impacted by increasing wind penetrations and in many cases are forced into mid-merit operation, a novel operating
strategy is investigated for these units. Many CCGTs include bypass stacks allowing
them to vent exhaust gas directly into the atmosphere and bypass the steam section
of the plant entirely. Running in this open-cycle manner, CCGTs will have reduced
efficiency but can start-up quickly. This thesis examines if a system with increasing
wind penetration can benefit from increased flexibility when CCGT units are allowed
to operate in a multi-mode regime. It is shown that such operation can improve system
reliability by increasing the sources of replacement reserve and that production from
peaking capacity is displaced, reducing the need for such units to be built.
Other options which are commonly cited as improving the flexibility of power systems
include pumped storage, compressed air energy storage, interconnection and demand
side management. Each of these can assist in balancing net load variability and so are
considered beneficial to the integration of wind power, however typically their impact
on the operation of base-load units has not been examined. This thesis investigates how
various forms of flexibility can alleviate or aggravate cycling of base-load generation.
It is found that many of these options will in fact be in competition with base-load
generation to provide energy and/or reserve to the system and so can actually increase
plant cycling.
On the premise that penetrations of variable renewables will continue to increase
for the foreseeable future, and that cycling operation will be a growing concern for
generators, this thesis presents a novel formulation for cycling related costs to be represented in a unit commitment algorithm. Incremental cycling costs related to start-ups
II
or ramping can be represented using the new formulation and depending on the level of
knowledge that is available, the resulting cost function can be linear, piece-wise linear
or step shaped. This new approach to modelling cycling costs has applications for both
long-term planning studies and real-world scheduling models. A case study on a 20
unit system was carried out and the inclusion of the new cycling cost formulation was
shown to reduce cycling operation, distribute the burden of cycling more evenly across
the units and reduce overall system costs relative to the case where cycling costs were
not modelled.
III
Contents
Abstract
I
Publications Arising from Thesis
VII
Acknowledgements
VIII
Acronyms and Symbols
X
Nomenclature
XII
1 Introduction
1.1 Evolving Power Systems and the Rise of Wind Power .
1.2 Impact of Wind Power on System Operation . . . . .
1.3 Wind Power on the Irish Power System . . . . . . . .
1.4 Thesis Objectives . . . . . . . . . . . . . . . . . . . . .
1.5 Summary of Thesis Contributions . . . . . . . . . . . .
1.6 Thesis Overview . . . . . . . . . . . . . . . . . . . . .
2 Cycling of Thermal Plant
2.1 Introduction . . . . . . . . . . . . . . . .
2.2 Damage to Power Plants Due to Cycling
2.3 Cycling Costs . . . . . . . . . . . . . . .
2.4 Next Generation Thermal Plant . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3 Unit Commitment with High Wind Power Penetrations
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.2 The Wilmar Planning Tool . . . . . . . . . . . . . . . . . .
3.2.1 The Scenario Tree Tool . . . . . . . . . . . . . . . .
3.2.2 The Scheduling Model . . . . . . . . . . . . . . . . .
3.3 Other Unit Commitment Models . . . . . . . . . . . . . . .
3.4 The Irish 2020 Test System . . . . . . . . . . . . . . . . . .
IV
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
4
7
10
11
12
.
.
.
.
14
14
15
18
21
.
.
.
.
.
.
22
22
23
23
24
28
29
4 Cycling of Base-load Plant on the Irish Power System
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2 Scenarios Examined . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3.1 Increasing Wind Penetration and the Operation of Base-Load
4.3.2 Sensitivity Analysis . . . . . . . . . . . . . . . . . . . . . . .
4.3.3 Effect of Modelling Assumptions . . . . . . . . . . . . . . . .
4.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5 Multi-mode Operation of Combined-Cycle Gas Turbines
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3 Test System . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.4.1 Utilization of the Multi-mode Function . . . . . . .
5.4.2 Benefits Arising from Multi-mode Operation . . . .
5.4.3 Sensitivity Studies . . . . . . . . . . . . . . . . . . .
5.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
6 Power System Flexibility and the Impact on Plant Cycling
6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.2 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.3 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
6.3.1 Impact on the Operation of Base-load Units . . . . . . .
6.3.2 Impact on Wind Curtailment and CO2 Emissions . . . .
6.4 Summary of Results . . . . . . . . . . . . . . . . . . . . . . . .
6.5 Other Flexibility Options . . . . . . . . . . . . . . . . . . . . .
6.5.1 Battery Electric Vehicles . . . . . . . . . . . . . . . . . .
6.5.2 Maintenance Scheduling . . . . . . . . . . . . . . . . . .
6.5.3 Control of Wind Power Output . . . . . . . . . . . . . .
6.5.4 Market Options . . . . . . . . . . . . . . . . . . . . . . .
7 Unit Commitment with Dynamic Cycling Costs
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . .
7.2 Formulation of Dynamic Cycling Costs . . . . . . . .
7.2.1 Cycling Costs Related to Start-ups . . . . . .
7.2.2 Cycling Costs Related to Ramping . . . . . .
7.3 Model and Test System . . . . . . . . . . . . . . . .
7.4 Results . . . . . . . . . . . . . . . . . . . . . . . . . .
7.4.1 Start-up Related Cycling Costs Results . . .
7.4.2 Ramping Related Cycling Costs Results . . .
7.4.3 Start-up and Ramping Cycling Costs Results
7.5 Summary . . . . . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
. . 33
. . 34
. . 37
Units 37
. . 43
. . 49
. . 50
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
53
53
56
60
61
62
66
71
77
.
.
.
.
.
.
.
.
.
.
.
79
79
81
84
84
91
92
93
93
94
96
96
.
.
.
.
.
.
.
.
.
.
98
98
99
100
104
108
113
113
117
118
120
8 Conclusions
122
8.1 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
V
References
127
Appendix A. Probability distribution of net load ramps
136
Appendix B. Cycling data for CCGT and coal units
138
Appendix C. Base-load cycling with/without storage/interconnection 141
Appendix D. Fuel Cost Curves
143
Appendix E. Publications
145
VI
Publications Arising from Thesis
Journal Publications:
1. Troy, N., Flynn, D., Milligan M. and O’Malley, M. “Unit Commitment with
Dynamic Cycling Costs”, IEEE Transactions on Power Systems, in review.
2. Troy, N., Flynn, D. and O’Malley, M. “Multi-mode Operation of Combined-Cycle
Gas Turbines with Increasing Wind Penetration”, Accepted to IEEE Transactions
on Power Systems
3. Troy, N., Denny, E. and O’Malley, M. “Base-load cycling on a system with significant wind penetration”, IEEE Transactions on Power Systems, vol. 25, issue
2, pp. 1088 - 1097, 2010.
Conference Publications:
1. Troy, N. and O’Malley, M. “Multi-mode Operation of Combined-Cycle Gas Turbines with Increasing Wind Penetration”, in Proceedings of the IEEE Power &
Energy Society General Meeting, Minnesota, USA, July 2010.
2. Troy, N. and Twohig, S. “Wind as a Price-Maker and Ancillary Services Provider
in Competitive Electricity Markets”, in Proceedings of the IEEE Power & Energy
Society General Meeting, Minnesota, USA, July 2010.
3. Troy, N., Denny, E. and O’Malley, M. “Evaluating which forms of flexibility most
effectively reduce base-load cycling at large wind penetrations”, in Proceedings
of the 8th International Workshop on Large-Scale Integration of Wind Power,
Bremen, Germany, October 2009.
4. Tuohy, A., Troy, N., Gubina, A. and O’Malley, M. “Managing wind uncertainty
and variability in the Irish power system”, in Proceedings of the IEEE Power &
Energy Society General Meeting, Calgary, USA, July 2009.
5. Troy, N., Denny, E. and O’Malley, M. “The relationship between base-load generation, start-up costs and generator cycling”, in Proceedings of the 14th Annual North American Conference of the International Association of Energy Economics, Louisiana, USA, December 2008.
VII
Acknowledgements
I would like to thank everybody whose help and support contributed to this thesis, but
in particular the following:
My supervisor Professor Mark O’Malley for his guidance and encouragement over the
past four years. Through his hard work and efforts I have benefited from many wonderful opportunities for which I am extremely grateful and consider myself lucky to have
found such a dynamic mentor.
My co-supervisor Dr Damian Flynn for the enthusiasm and time he invested in my
work. His attention to the finest detail is exemplary and I am very thankful for the
effort he put into my thesis.
Dr. Eleanor Denny for her support, insights and advice in the earlier stage of my
PhD.
Dr. Aidan Tuohy, to whom I am indebted for getting me up to speed with the Wilmar
model and answering my many annoying questions even when busy writing up his own
thesis.
Dr. Michael Milligan for hosting me at NREL and indeed all the other members of the
Grid Integration Group. My time at NREL was both insightful and enjoyable and I
am very grateful for the opportunity.
Dr. Jonathan O’Sullivan and Sonya Twohig for hosting me at EirGrid and for many
interesting discussions during that time.
Ms Magdalena Szczepanska for all her help over the past four years and for keeping things running smoothly.
All the students at the ERC who have been good fun and great friends. I look forward
to many more adventures together!
VIII
My parents, grandparents, extended family and friends, for their support and for providing relief from my academic pursuits.
But most especially I’d like to thank Shane for always being kind, supportive and
patient. I could not have done it without you.
IX
Acronyms and Symbols
ADGT Aero-derivative gas turbine
AIGS All Island Grid Study
ARMA Auto-regressive moving average
BNE Best new entrant
CAISO California Independent System Operator
CCGT Combined-cycle gas turbine
CEMS Continuous Emissions Monitoring Scheme
CHP Combined heat and power
CO Carbon monoxide
CO2 Carbon dioxide
DOE Department of Energy (US)
DSM Demand side management
EDUD Expected duration of unmet demand
ELCC Effective load carrying capability
EPRI Electric Power Research Institute
ERCOT Electric Reliability Council of Texas
EU European Union
EV Electric vehicle
EWEA European Wind Energy Association
X
GADS Generating Availability Data System
GAMS Generic Algebraic Modeling System
GE General Electric
HRSG Heat recovery steam generator
IRRE Insufficient ramping resource expectation
LOLE Loss of load expectation
NEPOOL New England Power Pool
NERC North American Electric Reliability Council
OCGT Open-cycle gas turbine
O&M Operations and maintenance
PHEV Plug-in hybrid electric vehicle
REFIT Renewable energy feed-in tariff
ROC Renewable Obligation Certificate
ROCOF Rate of change of frequency
RPS Renewable Portfolio Standards
SEM Single Electricity Market
SONI System Operator Northern Ireland
SPP Southwest Power Pool
STT Scenario Tree Tool
TR1 Tertiary operating reserve (Ireland)
UK United Kingdom
US United States
V2G Vehicle-to-Grid
XI
Nomenclature
Chapter 3 & 5
Indices
ccgt CCGT units
ccgtopen CCGT units in open-cycle mode
g Units
i Interval of the start-up process
s Scenarios
t Time
Parameters
Pgmin Minimum power output for unit ’g’ (MW)
Pgmax Maximum power output for unit ’g’ (MW)
PU (g, i) Power output for unit ’g’ at interval ’i’ of the start-up process (MW)
Startf uelg Start-up fuel required by unit ’g’ (GJ)
U Dg Duration of start-up process for unit ’g’ (h)
Binary Variables
Online 0/1 variable equal to 1 if unit ’g’ is online in scenario ’s’, at time ’t’
Vs,t,g
Start 0/1 variable equal to 1 if unit ’g’ is started in scenario ’s’, at time ’t’
Vs,t,g
Shut 0/1 variable equal to 1 if unit ’g’ is started in scenario ’s’, at time ’t’
Vs,t,g
XII
p(s,t,g) power output for unit ’g’, in scenario ’s’, at time ’t’ (MW)
Positive Variables
Start Start-up fuel used if unit ’g’ is started in scenario ’s’, at time ’t’ (GJ)
F uels,t,g
PgOf f Offline contribution to replacement reserve from unit ’g’ in scenario ’s’, at time
’t’ (MW)
Chapter 7
Indices/Sets
t, T Time step, set of time steps
g, G Units, set of units
i, I Interval of cycling cost function, set of intervals of cycling cost function
j, J Level of ramp, set of all ramp levels
l, L Segment of the piecewise linearisation of the variable cost function, set of all
segments of the piecewise linearisation of the variable cost function
Constants
costSg Cycling cost increment for each additional start
ThSg (i) ith threshold corresponding to cumulative start-ups
costSg (i) Cycling cost increment for each additional start-up, while NS (t,i) < ThS (i+1)
Rg production change (MW) over time period ‘t’ deemed damaging for unit ‘g’
Rg (j) jth production change (MW) over time period ‘t’ deemed damaging for unit ‘g’
costR
g Cycling cost increment for each additional ramp > R
ThR
g (i) ith threshold corresponding to cumulative ramps
costR
g (i) Cycling cost increment for each additional ramp, while NR (t,i) < ThR (i+1)
costX
g Cycling cost increment for each additional bi-directional ramp
ThX
g (i) ith threshold corresponding to cumulative bi-directional ramps
XIII
costX
g (i) Cycling cost increment for each additional bi-directional ramp, while NX (t,i) <
ThX (i+1)
Ig Total number of intervals in cycling cost function for unit ‘g’
j̄g Number of ramp levels defined for unit ‘g’
P̄g Maximum capacity for unit ‘g’
P g Minimum capacity for unit ‘g’
Ag Fixed cost for unit ‘g’($/h)
ag , bg , cg Coefficients of the quadratic production cost function of unit ‘g’
NLg Number of segments in piecewise linearization of the variable cost function for
unit ‘g’
Fl g Slope of segment l of the variable cost function for unit ‘g’
Tl g Upper limit of segment ‘l’of the variable cost function of unit ‘g’(MW)
U Tg Minimum up time for unit ‘g’
DTg Minimum down time for unit ‘g’
T̄ Number of hours in the planning period
Tgcold Number of hours unit must be offline for, beyond its minimum downtime, before
it is considered to be in a cold state
ccg Start up cost for cold start for unit ‘g’
hcg Start up cost for hot start for unit ‘g’
hup Number of hours unit ‘g’has been online for at start of planning period (h)
hdown Number of hours unit ‘g’has been offline for at start of planning period (h)
M Large number
α, β, γ Scaling factors
Binary Variables
sg (t) equal to 1 when a unit starts up at time t,
zg (t) equal to 1 when a unit shuts down at time t,
vg (t) equal to 1 when a unit is online at time t,
stepSg (t, i) equal to 1 when NS (t,1) ≥ ThS (i) at time t,
rg (t) equal to 1 when a unit undergoes ramp > Rg between time t and t-1,
XIV
rg (t, j) equal to 1 when a unit undergoes ramp > Rg (j) between time t and t-1,
stepR
g (t, i) equal to 1 when NS (t,1) ≥ ThR (i) at time t,
upg (t) equal to 1 when production at time t > production at t-1,
downg (t) equal to 1 when production at time t < production at t-1,
xg (t) equal to 1 when ramping switches direction between consecutive periods,
stepX (t, i) equal to 1 when NX (t,1) ≥ ThX (i) at time t.
Positive Variables
NSg (t) Cumulative start-ups,
NSg (t,i) Cumulative start-ups beyond threshold ThS (i),
CSg (t) Total cycling cost attributed to start-ups,
NR
g (t) Cumulative ramps > Rg ,
NR
g (t,i) Cumulative ramps > Rg beyond threshold ThR (i),
CR
g (t) Total cycling cost attributed to ramping,
NX
g (t) Cumulative incidents of bi-directional ramping,
NX
g (t,i) Cumulative incidents of bi-directional ramping beyond threshold ThX (i),
CX
g (t) Total cycling cost attributed to bi-directional ramping,
cpg (t) Production cost for unit ‘g’ at time ‘t’,
csg (t) Start-up fuel cost for unit ‘g’ at time ‘t’,
pg (t) Output (MW) for unit ‘g’ at time ‘t’,
D(t) System demand (MW) at time ‘t’,
δl (g,t) Variable used in the linearization of the variable cost function of unit ‘g’ at time
‘t’, represents the lth segment (MW).
XV
CHAPTER
1
Introduction
1.1
R
Evolving Power Systems and the Rise of Wind Power
ECENT years have seen the power generation sector undergo significant changes.
Traditionally electricity systems were operated by vertically integrated monopo-
lies whose main aim was to meet the demand as opposed to minimising cost (Narula et al.,
2002). However, by the 1980s deregulation and unbundling of utilities was seen as a
means of improving economic performance. In 1982 Chile kick-started electricity deregulation by passing a law which allowed large consumers of electricity to choose their
retailer and negotiate their prices freely. In 1990 the United Kingdom (UK) government
privatised the electricity supply industry, which led to other Commonwealth countries,
notably New Zealand and Australia, also pursuing deregulation. The European Union
(EU) directive 96/92 introduced in 1996 required Member States to create competitive
electricity markets, whilst by the late 1990s many states in the United States (US) were
also moving towards deregulation (Al-Sunaidy and Green, 2006).
1
Chapter 1. Introduction
2
Figure 1.1: Increased cycling due to introduction of electricity market in Ontario
(APPrO, 2006)
In the resulting competitive and volatile marketplaces that were created, generators
which had previously operated as base-load plant were often forced into flexible operation (Kitto Jr et al., 1996; Narula et al., 2002). Figure 1.1 which illustrates increased
plant start-ups following the introduction of a competitive electricity market in Ontario
provides an example of how greater flexibility is required in competitive markets. In a
competitive marketplace, energy traders or suppliers, seeking to maximise profitability,
will offer generation into power, exchange and ancillary service markets requiring units
to have short start-up times and good cycling capabilities. The ability to operate flexibly can bring considerable economic advantage for generators, as they have increased
opportunities to earn revenue from the market, such as through hourly and seasonal
market arbitrage or peak shaving for example (Balling and Hofmann, 2007). However,
the financial pressure to reduce capital costs in a competitive market can often lead to
power generating companies purchasing plants with cheaper and consequently poorer
performing components, which are more susceptible to cycling related wear and tear.
As such, older coal plants have been found to be more rugged and cost effective to cycle
compared to newer combined-cycle units (Lefton and Besuner, 2006).
Meanwhile, the acceptance that anthropogenic greenhouse gas emissions are resulting in climate change has led to the introduction of energy policies seeking to reduce the
Chapter 1. Introduction
3
environmental impact of electricity generation. Supporting renewable energy sources
and energy efficiency measures has been identified as vital to achieving emission reductions. Coupled with this, rising fossil fuel prices and instability in countries where fossil
fuels are sourced has led to widespread backing of renewables as a means of improving
security of supply and reducing exposure to fossil fuel price volatility. In 2008 the EU
imposed demanding climate and energy targets known as the ‘20-20-20’ targets which
are to be met by 2020. These aim to reduce EU greenhouse gas emissions by 20% below 1990 levels, supply 20% of energy consumption from renewable energy sources and
reduce primary energy consumption by 20% through energy efficiency measures (EU,
2008). Although the US has no comprehensive long-term energy policy, initiatives such
as Renewable Portfolio Standards (RPSs) and Renewable Energy Certificates (RECs)
have been taken at a state level to increase the use of renewable energy (Black & Veatch,
2011). Thus 28 out of 50 US states have set compulsory targets seeking renewable energy penetrations up to 40%, with a further 5 states having voluntary targets (DOE,
2009).
Wind power, now a proven and mature technology which offers near-zero emissions
and operating costs, represents a feasible means of meeting emissions and renewable
energy targets and consequently has experienced rapid growth over the past decade.
The cumulative installed wind power capacity in the U.S stood at 41.4 GW in the first
quarter of 2011 (AWEA, 2011b), just behind China which is set to reach 58 GW by
the end of 2011 (Castano, 2011). In Europe the total installed wind capacity exceeds
84 GW, with countries such as Germany, Spain and Denmark representing the largest
shares. In terms of energy penetration however, Denmark, Portugal, Spain and Ireland
lead the way, as seen for 2009 in Figure 1.2 (IEA, 2010).
In spite of the unprecedented economic downturn, the annual growth rate for wind
power has remained high with the installed wind power capacity in the EU increasing
by 12.4% in 2010, compared with 15% in the US (AWEA, 2011a; EWEA, 2011c). This
rapid pace of wind power installation is set to continue through the coming years, as
with the majority of hydro resources already exploited, wind power (and solar power in
some countries) represents the most scalable and competitive means of achieving 2020
Chapter 1. Introduction
4
Figure 1.2: Top 10 highest wind penetrations, as % of electricity consumption, in EU
countries (IEA, 2010)
targets. Up to now wind power has been supported by some form of subsidy such a
feed-in tariff or renewable certification scheme. However, with growing sales and larger
and more efficient turbines the cost of wind power is, in some countries (Brazil, Sweden,
Mexico and US) similar to the cost per MWh of coal generation and consequently it
may be possible to phase out subsidies over the coming years without hampering wind
power development (Bloomberg, 2011). EWEA (European Wind Energy Association)
predicts between 230 and 265 GW of installed wind power in Europe for the year 2020
(40 GW of which is assumed to be offshore wind) which would supply between between
14.4% and 16.7% of the total electricity demand (EWEA, 2011b).
1.2
Impact of Wind Power on System Operation
As higher penetrations of wind power are achieved, power system operation becomes
increasingly complex due to the variable and unpredictable nature of wind power. Traditionally system demand has been largely predictable as demand profiles follow daily,
weekly and seasonal patterns, allowing generation to be efficiently committed. Wind
power however introduces another element of uncertainty and thus systems with significant levels of wind need to utilise wind forecasts when committing and dispatching
Chapter 1. Introduction
5
generation. Approaches to wind forecasting can be categorised as physical or statistical, with modern forecasting systems employing a combination of the two. Physical
approaches, namely weather prediction models, which are typically used for horizons
of 6 to 72 hours, utilise data such as land and sea surface temperatures to physically
model atmospheric dynamics. Statistical approaches transform meteorological predictions into wind generation (often using artificial-intelligence based models) and are
found to give better accuracy for horizons up to 6 hours (Monteiro et al., 2009). The
desire of system operators for information regarding the reliability of forecast has led to
ensemble or probabilistic forecasts becoming popular. Ensemble forecasting produces
multiple forecasts, by varying the input parameters or by using multiple weather prediction models, to generate a probability density function of the most likely outcome
(Möhrlen et al., 2007). Wind power forecast error however, increases with the forecast
horizon and even when these state-of-the-art methods of forecasting are employed, the
day-ahead wind forecast error (root mean square error) for a region can be 8-12% of the
total wind capacity as reported in Siebert (2008), which can result in thermal units being over- and under-committed (Ummels et al., 2007). Thus power systems with large
wind power capacities will need to re-evaluate commitment decisions on a continual
basis as more up-to-date wind forecasts become available. The unpredictable nature
of wind power also requires conventional plant to carry additional reserves in order to
maintain system reliability, should an unexpected drop in wind power occur. Many approaches have been proposed to determine how much wind power increases the reserve
requirement on a given system and it has often been found that the increased reserve
requirements represents only a small percentage of the wind power output (Dany, 2001;
Doherty and O’Malley, 2005; Holttinen et al., 2008; Holttinen, 2005)
The variable nature of wind power will increase variability in net load (load minus
wind generation), which must be met by conventional generation on the system, resulting in a greater demand for operational flexibility from these units (Ummels et al., 2007;
Holttinen, 2005). Expected or unexpected reductions in net load, which can arise due
to declining wind power output, will force conventional plant to ramp up their output,
or if sufficient ramping capability is not available, fast-starting units will need to come
Chapter 1. Introduction
6
online. Periods of low demand coinciding with high wind power output can lead to
conventional plant being shut down, a problem which has been exacerbated of late due
to a reduction in demand as a result of widespread economic recession (Axford, 2009).
The culmination of adding more variability and unpredictability to a power system is
that thermal units will undergo increased start-ups, ramping and periods of operation
at low load levels, collectively termed “cycling” (Braun, 2004; Göransson and Johnsson,
2009; Holttinen and Pedersen, 2003; Meibom et al., 2009). Furthermore, in some systems wind is allowed to self-dispatch, so the forecast output from wind farms is not
included in the day-ahead schedule. This can lead to increased transmission constraints
which will further intensify plant cycling (GE, 2005).
Many systems are currently experiencing increased plant cycling as a result of wind
power and wind integration studies are predicting this problem to worsen. The Southwest Power Pool (SPP) wind integration study noted that in order to accommodate
higher wind penetration levels more operational flexibility (i.e. more start-ups and
cycling of units) would be required and this would increase as the forecast error increases (Charles River Associates, 2010). The Californian Independent System Operator’s (CAISOs) renewables integration study had similar findings, but with combinedcycle gas turbine (CCGT) units specifically identified as undergoing increased cycling.
Relative to a 2012 reference case, CCGT plant start-ups increased by 35% with 20% renewables on the system. Both the ‘NYISO 2010 Wind Generation Study’ and the ‘New
England Wind Integration Study’ also found that the operation of CCGTs was significantly impacted by an increased penetration of wind power (NYISO, 2010; GE, 2010).
The Nova Scotia wind integration study predicted that start-ups for large thermal units
would be significantly increased as wind penetrations increased and acknowledged that
the cost impact of this was not fully understood (Hatch, 2008), while Xcel Energy are
currently experiencing cycling of their coal fleet due to high wind penetrations in Colorado. In Göransson and Johnsson (2009), which studied the power system of Western
Denmark, the capacity factor for units with low start-up and turn-down performance
and high minimum load levels (i.e. base-load units) were found to be the most significantly impacted by wind power, while Oswald et al. (2008) found that more ramping
Chapter 1. Introduction
7
will be required from fossil fuel plants on the British system to maintain the power
balance.
Wind power will also impact a system’s dynamic performance. As wind power will
tend to displace conventional generation, it will also displace the inertial response provided by these units, which is vital to maintain system security when faults or outages
occur. In addition, wind turbines supply asynchronous power to the system which
can impact the system’s voltage stability. However, it is possible to implement control
features to emulate inertial response and mitigate the impact on voltage stability and
these will be necessary in order to increase the upper limit to the maximum penetration
of wind power on a system.
1.3
Wind Power on the Irish Power System
Situated on the western edge of Europe, Ireland is well positioned to benefit from strong
Atlantic winds and consequently has one of the best wind resources in the world (SEAI,
2010), as seen in Figure 1.3. The current installed wind capacity in Ireland stands at
over 1.8 GW, which provides in excess of 10% of the electrical energy demand and
another 4 GW of proposed wind capacity is in various stages of planning. The growth
of wind power in Ireland is also supported by ambitious Government targets (40% of
all electricity consumption to come from renewables by 2020) and competitive feedin tariffs (REFIT in Republic of Ireland and ROCs in Northern Ireland). REFIT
(Renewable Energy Feed-In Tariff) is guaranteed for up to 15 years (but not to extend
beyond 2024) and is paid to suppliers to encourage them to enter into power purchase
agreements with wind generators. The REFIT is linked to the Best New Entrant (BNE)
generator and has averaged at e57/MWh for large scale wind and e59/MWh for small
scale wind in previous years. The ROCs (Renewable Obligation Certificate) scheme
in Northern Ireland is somewhat different in that an obligation to purchase renewable
generation is mandated on suppliers.
The Irish system is small and relatively electrically isolated: a 500 MW interconnec-
Chapter 1. Introduction
8
Figure 1.3: Wind resource in Europe (Risø National Laboratory, 1989)
tor is in place linking Northern Ireland and Scotland, however, trading arrangements
limit exports from Ireland to Great Britain to a maximum of 70 MW and with power
exchanges set one month ahead of time, Ireland infrequently exports power. Therefore, variations in power output and high penetrations of wind generation are managed
domestically by conventional generation rather than by exchanges to Great Britain.
As such, record high instantaneous wind penetrations, in excess of 50%, have been
experienced on the Irish system, as seen in Table 1.1. This is resulting in increased
cycling of conventional generation on the Irish system as seen in Table 1.2, which compares the annual number of plant start-ups for three CCGT units in 2008 and 2010.
Consequently the Market Monitoring Unit (MMU) within the Energy Regulatory Authorities has identified power plant cycling as one of the foremost concerns of thermal
power generators operating in SEM (Single Electricity Market), the electricity market
of the Republic and Northern Ireland (MMU, 2010). However, MMU (2010) attributes
the intense plant cycling that some plants are experiencing to the introduction of SEM
Chapter 1. Introduction
9
and the subsequent increase in competition rather than the increasing wind power
penetration.
Table 1.1: Wind penetration on the Republic of Ireland and Northern Ireland systems
Installed Wind (MW)
Maximum Output (MW)
Maximum Energy Penetration (%)
Maximum Daily Energy Penetration (%)
Republic of Ireland
1455
1323
52.3
37
Northern Ireland
355
320
50
29
Table 1.2: Annual plant start-ups
Unit
Huntstown 1
Tynagh
Dublin Bay Power
2008
23
27
7
2010
63
45
37
In anticipation of the challenges facing power system operation with such high wind
penetrations, the Irish Governments commissioned a study entitled the ‘All Island
Grid Study’ (AIGS), published in 2008, to examine the ability of the 2020 Irish power
system to handle various amounts of electricity from renewable sources. Various levels
of installed wind capacity ranging from 2000 MW to 8000 MW were examined in this
study, with an assumed peak demand of 9.6 GW assumed. The study was divided
into several workstreams, the most relevant to this work being ‘Workstream 2B’, which
utilized a stochastic unit commitment and economic dispatch model to examine system
operation under the various renewable scenarios (AIGS, 2008). The key results of this
workstream, which are particularly relevant to the work of this thesis are as follows:
“With increasing wind power capacity installed, extreme values and the
standard deviation of the variation of the net load (load minus wind
power production in the actual hour) increases as well. Hence, the
power plant portfolio has to show enough flexible units (for example
with sufficient ramp up and down rates as well as low start-up times)
to be able to follow the net load.”
“Generally, the bigger part of the electricity production in the All
Chapter 1. Introduction
10
Island power system from conventional power plants is borne by coal
fired plants and newer CCGTs. With increasing wind power capacity
installed, the production and capacity factors of these units tends to
be decreased.... Coal fired units and newer CCGTs have a relative low
number of start-ups and high number of online hours. The number of
start-ups of these units tends to be increased with increasing wind power
capacity installed.”
Following on from the All Island Grid Study, the tranmsission system operators of
Northern Ireland (SONI) and the Republic of Ireland (EirGrid) conducted a comprehensive study to better understand the technical and operational implications associated with high shares of renewable energy called the ‘All Island TSO Facilitation of
Renewbles Studies’, which identified two key limitations to wind power penetration,
namely (i) frequency stability after a loss of generation and (ii) frequency and transient stability after severe network faults. This study suggested that the maximum
amount of ‘inertialess power’ (wind power and interconnector imports) that the system
could cope with lies between 60% and 80%, but could be as low as 50% unless ROCOF
relays on distribution connected wind farms were disabled. Nonetheless, the study
found that the limitations for instantaneous wind penetration did not fundamentally
conflict with the 2020 policy targets aiming at 40% electricity from renewables by 2020
(EirGrid and SONI, 2010a).
1.4
Thesis Objectives
The main objective of this research has been to investigate how the operation of thermal
plant will be impacted by high penetrations of wind generation on a power system.
Base-load coal and CCGT units in particular are examined, as these units, having been
designed for maximum fuel efficiency, tend to have limited operational flexibility. As
such, when subjected to cycling operation these units can accrue large levels of damage
to plant components, leading to increased maintenance requirements and forced outage
rates. In examining the operational impacts of high wind power penetrations on CCGT
Chapter 1. Introduction
11
units, a novel operating strategy for these units was identified, which involved allowing
CCGTs to switch between combined- and open-cycle mode when economically optimal.
The potential benefits and impacts of this new multi-mode strategy are investigated in
this research.
In an effort to improve power system flexibility and support integration of variable
renewable generation, various flexibility options such as storage, interconnection and
demand side management are commonly put forward in the literature. Analysis of these
options is typically concerned with their profitability in a system or their impact on
system production costs, wind curtailment or emissions, while the impact on base-load
generation is typically over-looked. This research investigates how incorporating such
flexibility options (and others) into a power system will impact cycling of base-load
units in a high wind power scenario.
Finally, having identified that the operation of base-load plant will be significantly
impacted as wind power penetrations increase, this research develops a unit commitment formulation to allow cycling related costs to be modelled in a dynamic manner.
The impact of accounting for cycling costs in a dynamic manner on plant dispatch is
evaluated.
1.5
Summary of Thesis Contributions
The novel contributions emanating from this thesis can be categorised as (i) the identification and investigation of a new operating strategy for CCGT units in a high wind
power scenario, (ii) the investigation of how various power system sources of flexibility
will impact the operation of base-load plant and (iii) the development of a new unit
commitment formulation to allow cycling costs to be modelled dynamically, such that
they accumulate over time based on plant operation, reflecting increased wear to plant
components and reduced plant life-time.
Examining the potential for running CCGT units in open-cycle mode, as well as
combined-cycle mode, revealed that a system can benefit from the additional fast-
Chapter 1. Introduction
12
starting capacity. The increased replacement (non-spinning) reserve availability from
CCGT units in open-cycle mode also results in increased system security. Furthermore,
open-cycle operation of CCGT units will displace production from conventional peaking
units, reducing the need for such units to be built and thus indicating a societal benefit.
Sensitivity studies revealed how the usage of this multi-mode function will be dependent
on the underlying level of flexibility present in the system. Optimizing the system
stochastically or allowing intra-day trading on interconnectors reduced the need for
flexibility to be extracted from generators and consequently resulted in less frequent
deployment of the multi-mode function.
The impact of various sources of power system flexibility, such as storage, interconnection or demand side management, on the operation of base-load plant has been
examined in this thesis. A side-by-side comparison reveals which are effective at reducing plant cycling, or alternatively which will aggravate plant cycling, in a high wind
power context. The results are somewhat surprising as it was found that many of
these options will in fact be in competition with base-load generation to provide energy
and/or reserve to the system and so actually increase plant cycling.
A novel unit commitment formulation was developed which utilises binary variables
to incur a dynamic incremental cost when cycling operation occurs. The types of operation which elicit a cycling related cost can be plant start-ups or ramping. The cycling
cost accumulates in tandem with plant operation such that it influences the dispatch
decisions. This formulation has particular applications for long term studies, such as
wind integration studies, as it can reflect the depreciation of a plant and potentially
show how the merit order of generation can be altered over time. A case study in
which this new formulation was implemented revealed that by modelling cycling costs
dynamically, the burden of cycling operation will, over time, be distributed more evenly
across the fleet of generators.
1.6
Thesis Overview
The remainder of this thesis is organised as follows:
Chapter 1. Introduction
13
• Chapter 2 describes the effects of cycling operation on plant equipment and the
damage mechanisms involved. It describes the cost components which make up
the total cost of cycling a unit and the difficulties in calculating these costs.
Various approaches which have been used to approximate these costs are also
described.
• Chapter 3 describes the stochastic unit commitment and economic dispatch modelling tool that was used in this thesis. A detailed description of the test system
is also provided.
• Chapter 4 examines how the operation of base-load units, coal and CCGT units
specifically, will be impacted with an increasing wind penetration. Sensitivities
are also conducted to examine the level of cycling these units would undergo in
the absence of pumped storage or interconnection on the system.
• Chapter 5 examines the potential for multi-mode operation of CCGT units under
various wind scenarios to determine if this new mode of operation can deliver benefits to the power system, via increased flexibility, or the generators themselves,
via increased generation opportunities.
• Chapter 6 examines how various flexibility options, namely pumped storage, interconnection, demand side management (DSM), multi-mode operation of CCGTs
and reduced minimum generating levels impact the operation of base-load units.
A side-by-side comparison of these options reveals which are the most effective
at reducing cycling of base-load plant.
• Chapter 7 presents a novel formulation for modelling the cycling costs in a dynamic manner within a Mixed Integer Programming (MIP) unit commitment
model. The formulation can be used to implement incrementing cycling costs for
starts or ramps for linear, piece-wise linear or step cycling cost functions.
• The thesis is concluded in Chapter 8.
CHAPTER
2
Cycling of Thermal Plant
2.1
Introduction
A
S discussed in Chapter 1, the increased variability and uncertainty that arises
when wind power is integrated into a power system can lead to more flexible
operation or ‘cycling’ being demanded from conventional plants. In addition to wind
power, the competitive markets in which these units operate are also a significant driver
of plant cycling as generators are forced into more market-orientated, flexible operation
to increase profits, while at the same time maintenance intervals are often lengthened
in order to minimize downtime and costs. An overcapacity of generation on a system
can also exacerbate plant cycling as less efficient plant may be prematurely forced down
the merit order.
Thermal plant can be broadly categorised as base-load, mid-merit or peaking. Midmerit units follow the daily demand profile and shut down nightly whilst peaking units
14
Chapter 2. Cycling of Thermal Plant
15
are used to meet the extreme peaks in demand. Base-load thermal units, typically coal,
Combined-Cycle Gas Turbine (CCGT) or nuclear, are those units which traditionally
run on a continuous basis, at maximum efficiency, to supply the base electricity demand
and therefore tend to have minimal operational flexibility. As such, the rapid changes in
temperatures and pressures that occur during cycling operation will result in accelerated
deterioration of these units’ components through various degeneration mechanisms such
as fatigue, erosion, corrosion, etc. This in turn will lead to more frequent forced outages,
reduced plant lifetime and significant costs for these units. As illustrated in Figure 2.1,
the damage incurred from cycling operation is related to the temperature transients
in the plant’s components, with online ramping being the least damaging and cold
start-ups the most damaging.
Figure 2.1: Cycling damage increases as plant temperature decreases (Lefton, 2004)
This chapter discusses some of the common wear-and-tear effects that plants will
experience when undertaking cycling operation. The various cost implications of cycling
operation are identified and the approaches used to quantify these costs are examined.
2.2
Damage to Power Plants Due to Cycling
Fatigue damage is the most common problem for cycling units (EPRI, 2001b). Fatigue
is caused by repeated exposure to large temperature and pressure transients, typical
of cycling operation (Lefton et al., 1997), and manifests as cracking or mechanical failure of structures (EPRI, 2001b). Traditionally, base-load units ran uninterrupted at
full production and as such were designed to operate under creep conditions (constant
Chapter 2. Cycling of Thermal Plant
16
stress), with older design codes neglecting to consider fatigue (fluctuating stress) as a
damage mechanism (EPRI, 2001b). Creep and fatigue can interact in a synergistic
manner in that creep will reduce fatigue life and likewise fatigue reduces creep life, as
depicted in Figure 2.2 (EPRI, 2001b). Therefore when a base-load unit which has been
operating under creep conditions, switches to cycling operation, the creep-fatigue interaction renders the unit highly susceptible to component failure (Lefton et al., 1995,
1997). Creep-fatigue interaction is a particular concern for components such as superheater, reheater and economizer headers (MMU, 2010).
Figure 2.2: Creep-fatigue interaction (EPRI, 2001b)
Thick-walled components such as boilers, which are necessary to withstand the
extreme temperature and pressure associated with base-load operation, can develop
through-wall temperature differences during cyclic operation. This results in differential thermal expansion and ultimately places the component under high stress,
causing cracks to initiate and grow (EPRI, 2001b). An example of this is shown in
Figure 2.3. Rapid temperature transients will also cause differential thermal expansion
and fatigue issues in components such as header ligaments or boiler tube ties (EPRI,
2001b). Peak stresses typically occur in regions of discontinuity (Brown, 1994), and
therefore welded joints are highly stressed locations (King, 1996).
Expansion related issues can also arise due to cycling operation. Thin-walled com-
Chapter 2. Cycling of Thermal Plant
17
Figure 2.3: Cracking seen from inside economizer header (King, 1996)
ponents, heat recovery steam generator (HRSG) ducts for example, will heat up rapidly
during plant start-up, whilst the supporting steelwork remains cold, resulting in differential thermal expansion and consequently high stress (Brown, 1994). Likewise, on
start-up, a typical large boiler will expand downward from its roof support by 250 mm
which must be supported by the boiler support framework. If start-ups are occurring
on a regular basis it can lead to failure of the boiler support framweork (MMU, 2010).
Mechanical fatigue is also common during turbine run-up, when the rotor passes
through a series of critical speeds where vibration levels are increased significantly. Repeated start-ups can subject components such as turbine blades to high cycle fatigue
levels (MMU, 2010).
Thermal shocking of economizer headers occurs when cold feedwater is introduced
to warm headers when a unit is re-starting following an overnight shut-down, for example (King, 1996), or alternatively when hot steam is admitted to cold superheater
headers (EPRI, 2001b). If this is occurring on a regular basis it will lead to internal
fatigue cracking (King, 1996). This is irreparable and must be monitored constantly
for propagation. Start-ups and shut-downs can also cause oxide scales that have accumulated in steam-side equipment to spall due to the differences in the coefficients of
thermal expansion between the oxide and the metal. The hard oxide particles become
entrained in the steam and are carried through to the turbine causing erosion of the
turbine blades (French, 1993).
Increased frequency of shutdowns can contribute to infiltration of dissolved oxy-
Chapter 2. Cycling of Thermal Plant
18
gen and other non-condensible gases, which will also lead to higher levels of erosion
and corrosion. This can occur in cycling units when the condenser vacuum is not
maintained sufficiently during offline periods. In addition as a plant goes through various modes of operation and cycles, contaminant can be disturbed and disseminated
throughout the steam-condensate cycle. Thus cycling units will need to employ continuous water chemistry monitoring (Energy-Tech, 2004).
Fatigue stresses during start-up and shutdown can also result in cracking of electrical equipment such as copper turns, as shown in Figure 2.4, and the resulting arcing
and burning can cause short-circuits (Moore, 2006). Coils with shorted turns operate at lower temperatures than regular coils and the resulting temperature difference
can give rise to rotor bowing. This will cause unbalanced magnetic forces giving rise
to rotor vibration. If the problem becomes severe enough forced outages can occur
(Albright et al., 1999).
Figure 2.4: Cracked copper turn and rotor bowing (Moore, 2006)
2.3
Cycling Costs
Any power generating company seeking to maintain profitable operation desires to
know the cost impact of cycling operation for their fleet of generators. However, quantifying, or even estimating, the magnitude of these cycling costs is challenging given
the extensive range of components affected by cycling, as discussed in Section 2.2. In
addition, the damage caused by cycling may not be immediately apparent and often it
can be several years before it manifests itself. Studies by Aptech Engineering Inc. (now
Intertek Aptech) suggest that it can take from 1 to 7 years for an increase in the failure
rate to become evident after switching from base-load to cycling operation Lefton et al.
Chapter 2. Cycling of Thermal Plant
19
(1998). The challenge of attributing costs to cycling operation is complicated further
by the fact that normal base-load operation also results in some degree of damage to
a units components and identifying the damage due to cycling from that associated
with normal operation is also problematic. Considering these difficulties Aptech have
concluded that utilities typically underestimate cycling costs by a factor of 3 to 30
Lefton et al. (1998).
Research related to the cost of generation cycling has been led by EPRI (Electric
Power Research Institute) and Aptech and the approaches employed can be categorized
as top-down (statistical analysis) or bottom-up (component modelling). EPRI carried
out a top-down study as part of its ‘Cycling Impacts Program’ which utilized multivariate regression models to analyze the operating regimes of 158 units from NERC (North
American Electric Reliability Corporation) GADS (Generating Availability Data System) and CEMS (Continuous Emission Monitoring) data, in an attempt to identify
patterns relating operation to capital expenditure. However, the inconsistency in accounting practices between individual units complicated the modelling process and no
correlation was found (EPRI, 2001a, 2002). Aptech employ a combination of top-down
models based on historical operations, forced outage and cost data, as well as bottomup methods which calculate operational stresses and the life expenditure of critical
components using physical models fine-tuned with real plant data, in order to determine cycling costs for individual generating units (Lefton, 2004). Aptech have analyzed
cycling costs for over 300 generating units and found that the cost of cycling a conventional fossil-fired power plant can range from $2,500-500,000 per start/stop cycle
depending on unit age, operating history and design features, and are often grossly
underestimated by utilities (Lefton, 2004; Lefton et al., 1998). Babcock Energy Ltd.
also developed a methodology for determining the long-term damage that arises from
two-shift operation in order to optimize operating procedures and minimize damage.
This involved identifying components most susceptible to creep-fatigue damage using
data from thermocouples and modelling these components using finite elements so that
operational events could be related to induced stresses (Brown, 1994).
The factors which contribute to the total cost of cycling are: (i) increased fuel con-
Chapter 2. Cycling of Thermal Plant
20
Figure 2.5: Impact of cycling on forced outage rate (Lefton, 2011)
sumption due to increased plant start-ups and operation at part-load levels (and therefore reduced efficiency), (ii) increased fuel consumption due to loss of plant efficiency
arising from increased wear to components, (iii) increased operations and maintenance
(O&M) costs due to increased wear-and-tear to plant components, (iv) increased capital costs resulting from component failures, (v) increased environmental costs resulting
from increased emissions, and (vi) loss of income due to longer and more frequent forced
outages. Figure 2.5, provides an example of how the forced outage rate of a plant can
increase as a result of cycling operation, however, capital expenditure on plant upgrades
can help combat this. Of the studies undertaken to date, the magnitude of these cycling
costs has been significant. For example, a recent study by Aptech on Excel Energy’s
Harrington coal plant suggested that for each additional hot start the unit performed,
the maintenance related cycling costs the unit would incur were $87k, more than 5
times greater than the cost of the start-up fuel consumed (Xcel Energy, 2010). This
would indicate the importance of having a good understanding of cycling costs in order
to maintain profitable operation in the long term.
However, in reality generators will often under-value these costs in order to keep
their short-run costs down in a competitive marketplace, the consequence of which is
that the generator will subsequently be scheduled to cycle more often. Or in some
situations generators will take advantage of the uncertainty surrounding these cycling
Chapter 2. Cycling of Thermal Plant
21
costs in order to exercise market power. For example, a generator may increase its startup costs excessively in order to avoid shut-down, although this strategy may result in
them being left offline following a trip or scheduled shut-down because of their excessive
start-up cost. In any case in most markets at present it is unclear how these costs should
be represented in a generator’s bid. Generators in SEM, the Irish electricity market,
are directed to include cycling costs in their start-up costs, however cycling costs could
also be included in shut-down, no-load or energy costs, or even defined as a new market
product such as ramping costs (Flynn et al., 2000).
2.4
Next Generation Thermal Plant
With increasing penetrations of variable renewables and competitive electricity markets
becoming the norm worldwide, power plant manufacturers are recognising the need for
greater operational flexibility (Probert, 2011). Siemens, for example, have outlined
areas where CCGT plant can be upgraded with new features such as a stress and
fatigue monitoring system for the HRSG, a piping warm-up system, attemperators
in the steam lines to maintain required temperatures in order to make them more
capable of frequent cycling (Siemens, 2008b). General Electric (GE) meanwhile have
launched their ‘FlexEfficiency CCGT’ which offers faster ramp rates, shorter start-up
times, lower turndown and fuel flexibility whilst achieving an efficiency of 61%. Next
generation thermal plant can also avail of new materials which have been developed such
as the high strength P91 steel which allows for high-pressure components to be made
thinner (EPRI, 2001b). Thinner components will reach thermal equilibrium quicker
and therefore are less susceptible to cracking. Improvements in instrumentation will
also allow for easier start-up and shut-downs and part-load operation (Energy-Tech,
2004). Online monitoring systems can help to protect critical components from thermal
stresses. However, although next generation thermal plant may be more suited to
cycling operation, current generation will still be in operation for decades more. Thus
cycling poses serious difficulties for generators seeking to remain in profitable operation
and system operators who must maintain a stable system in spite of increasing forced
outages.
CHAPTER
3
Unit Commitment with High Wind Power Penetrations
3.1
Introduction
P
RIOR to the large-scale deployment of renewables, uncertainty in power systems
was limited to load forecast error and the unplanned outages of generators or
transmission lines. In order to maintain a secure system, adequate levels of spinning
and non-spinning reserve were maintained to cover this error. Incorporating variable
renewable generation adds an additional source of uncertainty given the unpredictable
nature of renewable power sources. With low levels of renewables on power systems,
additional reserve is needed to cover the additional uncertainty associated with renewables. However, as the penetration of renewables grows, it becomes increasingly inefficient to rely on reserves alone to cover the uncertainty related to renewables. Rather
more robust schedules are required through stochastic scheduling, which considers multiple scenarios corresponding to multiple values of the stochastic variable, in this case
the power output from the renewable generation (Monteiro et al., 2009). In addition,
22
Chapter 3. Unit Commitment with High Wind Penetration
23
to make the most efficient use of the renewable generation, forecasts need to be utilized. As the accuracy of these forecasts increases as the forecast horizon decreases,
it is important that updated forecasts are used to update the commitment decisions
through a rolling unit commitment mechanism (Kiviluoma and Meibom, 2011). This
can in turn lead to a reduced reserve requirement (AIGS, 2008).
3.2
The Wilmar Planning Tool
The Wilmar Planning Tool is an output of a collaborative research effort supported by
the European Commission to develop a tool to analyse the integration of wind power in
large liberalised electricity systems (Meibom, 2006). The original model was developed
for two power pools: NordPool and the European Power Exchange, (i.e. Germany,
Denmark, Norway, Sweden and Finland). It was later adapted to the Irish system
as part of the All Island Grid Study (AIGS, 2008; Meibom et al., 2011; Tuohy et al.,
2009). Wilmar is an advanced stochastic, mixed integer unit commitment and economic
dispatch model, the main functionality of which is embedded in the Scenario Tree Tool
and the Scheduling Model.
3.2.1
The Scenario Tree Tool
The Scenario Tree Tool (STT) generates scenarios trees which feed into the Scheduling
Model. Each branch of the scenario tree represents a realistic forecast scenario of load,
wind power output and demand for replacement reserve (activation time > 5 minutes).
The STT also produces a forced outage time series for each generating unit.
The STT utilizes knowledge of historical wind speed forecast accuracy and knowledge
of the correlation between wind speed forecast errors in neighbouring areas, as well as
historical load data and load forecasts, to identify an Auto Regressive Moving Average
(ARMA) series, based on the methods described in (Söder, 2004). The parameters of
the ARMA series are determined by minimizing the difference between the standard
Chapter 3. Unit Commitment with High Wind Penetration
24
deviation of the historical forecast error and the standard deviation of the forecast error
produced by the ARMA series for each hour. The ARMA series is used to simulate load
and wind speed forecast errors for various time horizons. These simulated load and wind
speed forecast errors are paired in a random way before a scenario reduction technique,
following the approach of (Dupacova et al., 2003), is applied. The resulting load and
wind speed forecast error scenarios are combined with scaled-up load and wind speed
time series to produce load and wind speed forecast scenarios. Finally, the wind speed
forecast scenarios are transformed to wind power forecast scenarios using an aggregated
wind power curve following the approach of (Norgaard and Holttinen, 2004). For each
scenario the demand for replacement reserve (activation time >5 minutes) is calculated
based on a comparison of the hourly power balance considering perfect forecasts and no
forced outages with the power balance considering scenarios of wind and load forecast
errors as well as forced outages. A percentile of the deviation between the compared
power balances must be covered by replacement reserves; in this case the 90th percentile
is chosen based on current practice (Meibom et al., 2011). A forced outage time series
for each unit is also generated by the STT using a semi-Markov process based on
historical plant data of forced outage rates, mean time to repair and scheduled outages.
3.2.2
The Scheduling Model
The Scheduling Model minimizes the expected costs for all scenarios, subject to system
constraints for reserve and minimum number of units online (in this case 6 units must
be online in all time periods in the Republic of Ireland and 2 units in Northern Ireland).
A minimum number of online units are maintained to ensure a sufficient level of system
inertia. These costs include fuel, carbon and start-up fuel costs (always assumed to
be hot starts). In addition to replacement reserve, one category of spinning reserve,
namely tertiary operating reserve (TR1), is modelled, which has a response time of 90
seconds to 5 minutes and can only be supplied by online units. Wind generators, when
curtailed, are assumed to be capable of contributing to spinning reserve requirements.
Sufficient spinning reserve must be available to cover an outage of the largest online
unit occurring concurrently with a fast decrease in wind power production over the
Chapter 3. Unit Commitment with High Wind Penetration
25
TR1 time frame, as described in (Doherty and O’Malley, 2005). Generator constraints
such as minimum down times (the minimum time a unit must remain offline following
shut-down), synchronization times (time taken to come online), minimum operating
times (minimum time a unit must spend online once synchronized) and ramp rates
must also be obeyed.
Rolling planning is employed to re-optimize the system as new wind generation and
load information become available. Starting at noon each day, the system is scheduled
over 36 hours until the end of the next day. The model steps forward with a three hour
time step and in each planning period a three-stage, stochastic optimisation problem
is solved. This involves a deterministic first-stage covering three hours, a stochastic
second stage with three scenarios covering three hours and a stochastic third stage with
six scenarios covering a variable number of hours, depending on the planning period
in question, as seen in Figure 3.1 (AIGS, 2008). The structure of the scenario tree
assumes perfect knowledge of load and wind power output in the first three hours and
uncertainty in subsequent hours, and the opportunity to revise the planned commitment
every three hours based on information from new forecasts. The model produces a yearlong dispatch at an hourly time resolution for each individual generating unit.
The model can also be run in deterministic and perfect foresight modes whereby
only one wind generation and load scenario are planned for. In deterministic mode, this
scenario is the expected value of wind and load. The expected value of wind is found by
summing, for all (post-reduction) scenarios, the product of the wind power forecasts and
their probability of occurrence. The expected value of load and replacement reserve is
found similarly (Tuohy et al., 2009). Consequently, the scenario planned for will differ
from the realized scenario. This mode is typical of the scheduling process currently
practiced by most system operators, i.e. only one scenario is planned for and it will
contain some level of forecast error. Perfect foresight mode contains no forecast error
for wind generation or load but forced outages still occur, as with all other modes.
Further detail on the model and formulation of the unit commitment problem can
be found in (Meibom et al., 2011). The Generic Algebraic Modeling System (GAMS)
Chapter 3. Unit Commitment with High Wind Penetration
26
Figure 3.1: Illustration of rolling planning and decision structure in Wilmar
is used to solve the unit commitment problem using the mixed integer feature of the
CPLEX solver (version 12). For all simulations in this study the model was run with a
duality gap of 0.5%. A year-long simulation takes > 3 hours when run in deterministic
mode or > 24 hours in stochastic mode, on an Intel core quad 3 GHz processor with 4
GB of RAM.
3.2.2.1
Modelling DSM
Later versions of the Wilmar planning tool included add-ons to model demand side
management (DSM), which is utilised in Chapter 6.
Chapter 3. Unit Commitment with High Wind Penetration
27
DSM units can be either peak clipping or peak shifting units. Peak clipping units
allow demand to be reduced at a cost to the system without increasing demand at
another time. They are modelled as flexible gas turbines with a variable operating
cost and no fuel or start-up costs. Peak shifting units allow demand to be reduced
and reallocated in time at a cost to the system, without reducing the overall energy
demand. They are modelled as storage units with 100% efficiency. The constraints
implemented in Wilmar to model DSM ensure that (i) the DSM units are scheduled
day-ahead and their dispatch cannot be revised intra-day, (ii) the DSM units cannot
provide non-spinning reserve, (iii) all demand shifted over a day must must be added
to demand at another point in that day, and (iv) the amount of demand clipped by a
peak clipping unit cannot exceed a defined energy limit.
3.2.2.2
Improved Modelling of Plant Start-ups
More detailed modelling of plant start-ups was implemented to improve the validity
of results. In the original version of the Wilmar Planning Tool, units remained at
zero production over the course of their start-up period. Here, units are block loaded
from zero to minimum output over the course of the start-up process, following the
formulation given in (Arroyo and Conejo, 2004).
The start-up and shut-down binary variables are set appropriately by Equation 3.1.
Power output levels, PU (i), are defined for each interval, i, of the units’ start-up process.
Equation 3.2 sets the minimum allowable power output for a unit equal to PU (g,i)
when the unit is in the ith interval of the start-up process, or equal to its minimum
stable operating level when the unit is online and not in its start-up process. Likewise,
Equation 3.3 sets the maximum allowable power output for a unit equal to PU (g,i)
when the unit is in the ith interval of the start-up process, or equal to its maximum
operating level when the unit is online and not in its start-up process. Equation 3.4 is
needed for the commitment logic (Arroyo and Conejo, 2004).
Start
Shut
Online
Online
Vs,t,g
− Vs,t,g
= Vs,t,g
− Vs,t−1,g
(3.1)
Chapter 3. Unit Commitment with High Wind Penetration
U Dg
Online
p(s, t, g) ≥ Pgmin [Vs,t,g
−
X
U Dg
Start
Vs,t−i+1,g
] +
i=1
X
Start
PU (g, i)Vs,t−i+1,g
X
(3.2)
i=1
U Dg
p(s, t, g) ≤
28
U Dg
Start
Online
PU (g, i)Vs,t−i+1,g
+ Pgmax [Vs,t,g
−
i=1
X
Start
Vs,t−i+1,g
]
(3.3)
i=1
U Dg
Online
Vs,t,g
≥
X
Start
Vs,t−i+1,g
(3.4)
i=1
3.3
Other Unit Commitment Models
Many approaches are available for solving the unit commitment problem, as discussed in
(Padhy, 2004; Salam, 2007; Sen and Kothari, 1998), ranging from heuristic approaches
such as priority list to mathematical programming approaches such as dynamic programming, Lagrangian relaxation or mixed integer programming. Dynamic programming was the first optimization based method to be applied to the unit commitment
problem and is used worldwide (Padhy, 2004), however it suffers from the curse of dimensionality as it evaluates the complete decision tree, and thus for larger systems the
solution time can become impractical (Sen and Kothari, 1998). Simplifications such
as truncation or fixed priority ordering have been implemented to reduce the search
space but this can lead to suboptimal schedules (Salam, 2007). Lagrangian relaxation,
which involves decomposing the primal problem into sub-problems which are linked by
Lagrangian multipliers, is one of the most commonly used unit commitment formulations in electricity markets worldwide. However, it is well understood that given the
nature of how it works it will generally produce sub-optimal solutions and not produce
a global optimal solution (EirGrid and SONI, 2010b).
One of the key advantages to using MIP models (such as the Wilmar model), in addition to global optimality is the ease of adding constraints. As noted in Streiffert et al.
(2005), MIP models do not require complex algorithmic development to implement sim-
Chapter 3. Unit Commitment with High Wind Penetration
29
ple constraints unlike Lagrangian relaxation models for example which would require
the addition of new Lagrangian multipliers. (Streiffert et al., 2005) also notes that
more accurate modelling of combined-cycle plant is more challenging for a Lagrangian
relaxation model compared to a MIP model; a topic that is dealt with in this thesis.
MIP models have also benefited in recent years from improvements in the solution
methods. Traditionally MIP models were solved using the branch and bound technique, however, more recently other techniques such as node pre-solve, heuristics and
cutting planes have been implemented to improve the solution and the optimization
time. The commercial solver CPLEX (which was used in this work to solve the Wilmar
model) employs these techniques to reduce the upper (heuristics and node presolve) and
lower (cutting planes and node presolve) bounds of the objective function (Bixby et al.,
2000). Implementing a combination of solution techniques has been found to yield a
dramatic reduction in optimization time. Branch and bound algorithms have an additional advantage of being suitable for parallel processing (Streiffert et al., 2005). Many
systems such as CAISO, PJM and the Irish system are now using or testing MIP unit
commitment models (EirGrid and SONI, 2010b).
3.4
The Irish 2020 Test System
The test system used in the following chapters is the Irish 2020 system, based on
portfolio 5 from the All Island Grid Study (AIGS, 2008; CER, 2010). Table 3.1 shows
the number of units, installed capacity and average operating cost (fuel) by generation
type for this test system. The peak demand from AIGS (2008) was 9.6 GW peak
and the total demand was 54 TWh. More recent long term forecasts (EirGrid, 2009)
however, have indicated a considerably lower peak demand for 2020, resulting from
the current economic depression. Thus a revised test system has also been studied, in
which the demand profile is scaled down to a 7.55 GW peak and a total demand of
42 TWh. In this revised system four 103.5 MW OCGT units were also removed from
the original grid study portfolio (which contained 8 OCGT units as seen in Table 3.1),
as recent generation adequacy reports would indicate they are unlikely to be built by
Chapter 3. Unit Commitment with High Wind Penetration
30
2020 (EirGrid, 2009).
Table 3.1: Generation Mix of Test System
Generation Type
Wind power
CCGT
Coal
OCGT
Gasoil
Other renewables
Peat
Pumped storage
Hydro
Legacy CCGT
CHP
ADGT
Tidal
Capacity
(MW)
No. Units
2000/4000/6000
4012
1324
828
383
360
343
292
216
215
166
111
72
10
5
8
8
3
4
15
2
2
1
Avg. Operating
Cost (e/MWh)
0
39.79
18.45
61.16
121.26
10
36.32
0
0
47.97
37.94
47.85
0
For both of the test systems, three different levels of installed wind power were examined: 2000, 4000 and 6000 MW, which supply 11%, 23% and 34% or 15%, 29% and 43%
of the total energy demand, on the 9.6 GW peak and 7.55 GW peak systems respectively. The wind power data used to generate the scenario trees, used in AIGS (2008)
and this thesis, was 2004 data from 11 onshore regions across Ireland and Northern
Ireland and 10 offshore regions. The wind power time series collected from each region
were smoothed to account for wind correlation effects. To simulate wind speed forecast
errors, required for generating the scenario trees, wind speed forecasts for 6 locations
were used. However, forecast results were only available for time horizons greater than
5 hours so in order to generate forecast errors for the first 5 hours persistence forecasts
were assumed for the 6 locations. Figure 3.2 shows the day-ahead wind power forecast
error probability function (mean absolute error is 9.6%). It is evident that wind power
is more frequently over-forecast on the test system but the largest forecast errors were
under-forecasts. The additional amount of spinning reserve that must be carried to
cover wind power uncertainty was determined in Doherty and O’Malley (2005) for the
Irish 2020 test system and is shown in Table 3.2.
Chapter 3. Unit Commitment with High Wind Penetration
31
Figure 3.2: Day-ahead wind forecast error
Table 3.2: Additional spinning reserve requirement due to wind generation
Wind Generation (MW)
0-1000
1000-2000
2000-3000
3000-4000
4000-5000
5000-6000
TR1 (MW)
5
18
37
63
94
131
The pumped storage units, with a round-trip efficiency of 75% and a maximum
pumping capacity of 70 MW each, are large providers of spinning reserve to the system,
however at least 50% of the spinning reserve target has to be provided by conventional
units (excluding pumped storage and wind generation). The 2 CHP units have ‘mustrun’ status as they provide heat for industrial purposes. The outputs for hydro and
tidal units are inputted to the scheduling model as a time series and these units are
also not dispatchable. Sewage gas, landfill gas, biogas and biomass generation make up
the ‘other renewables’ category. Fuel prices are as given in Table 3.3. Base-load gas
generators (i.e. CCGTs and CHP) are assumed to have long-term fuel contracts and
therefore pay a cheaper fuel price compared to mid-merit gas generators (i.e. OCGTs,
Chapter 3. Unit Commitment with High Wind Penetration
32
ADGTs and legacy CCGTs). Differences in the fuel price for coal and gasoil in the
Republic of Ireland and Northern Ireland reflect varying delivery costs.
Table 3.3: Fuel Prices by Fuel Type
Fuel
Renewables
Coal - Republic of Ireland
Coal - Northern Ireland
Peat
Base-load gas
Mid-merit gas
Gasoil - Northern Ireland
Gasoil - Republic of Ireland
Fuel Price (e/GJ)
0
1.75
2.11
3.71
5.91
6.12
8.33
9.64
The test system assumes that there is 1000 MW of HVDC interconnection in place
between Ireland and Great Britain and it is scheduled on an intra-day basis, i.e. it can
be rescheduled in every 3 hour rolling planning period. It is assumed that the total
1000 MW can be exported from Ireland to Britain, however, when Ireland is importing
from Britain 100 MW of capacity is maintained to provide spinning reserve. In addition
another 50 MW of spinning reserve is assumed to be available from interruptible load.
A simplified model of the British power system is included, with aggregated units, no
integer variables for generators and where wind generation and load are assumed to be
perfectly forecast. The total demand in Britain is assumed to be 370 TWh with a peak
of 63 GW and the installed wind capacity is assumed to be 14 GW. A carbon price of
e30/ton was assumed.
The 2020 Irish system serves as an interesting test system to study issues arising
from large-scale wind power. Being a small island system, with limited interconnection
to Great Britain integration issues arise and become more obvious at lower levels of
wind power and can indicate future issues for other power systems pursuing large-scale
wind power. The large proportion of base-load units on the Irish system, most of which
are CCGTs, combined with the high wind penetration deem it useful for studying plant
cycling and investigating means of limiting the extent of this cycling operation. Thus,
the findings in this thesis bear relevance to other gas and wind-dominated systems, for
example the ERCOT system.
CHAPTER
4
Cycling of Base-load Plant on the Irish Power System
4.1
C
Introduction
ERTAIN developments in the electricity sector may result in suboptimal operation of base-load generating units in countries worldwide. Despite the fact that
they were not designed to operate in a flexible manner, increasing penetration of variable power sources, such as wind generation, coupled with increased competition in
the electricity sector can lead to these base-load units being shut down, ramped or
operated at part-load levels more often. An overcapacity of generation on a system can
also exacerbate plant cycling as less efficient plant may be prematurely forced down
the merit order.
Although all conventional units will be impacted to some degree by the integration
of wind generation, it is cycling of base-load units that is particularly concerning for
system operators and plant owners. As these units are designed for maximum efficiency,
33
Chapter 4. Cycling of Base-load Plant
34
they typically have limited operational flexibility, and as such cycling these units will
result in accelerated deterioration of plant components through various degeneration
mechanisms such as fatigue, erosion, corrosion, etc. This will lead to more frequent
forced outages and loss of income, as discussed in Chapter 2. Start/stop operation and
varying load levels result in thermal transients being set up in thick-walled components
placing them under stress and causing them to crack. Cycling interrupts plant operation
which can in turn disrupt the plant chemistry resulting in higher amounts of oxygen and
other ionic species being present, and therefore leading to corrosion and fouling issues.
Thus, excessive cycling of base-load units can potentially leave these units permanently
out of operation prior to their expected lifetimes.
The severity of plant cycling, will be dependent on the generation mix and the
physical characteristics of the power system. It is widely reported that the availability
of interconnection and storage can assist the integration of wind on a power system
(IEA, 2008; EWEA, 2011a). Interconnection can allow imbalances from predicted wind
power output or variations in net load to be compensated via imports/exports, whilst
some form of energy storage can allow excess wind to be more easily absorbed by
charging (and thereby increasing demand) during these periods. This should relieve
cycling duty on thermal units as the onus on them to balance fluctuations is relieved.
This chapter examines the effect that an increasing penetration of wind power will
have on the operation of base-load units. The role that interconnection and storage
play in alleviating or aggravating the cycling of base-load units is also investigated
across different wind penetration scenarios.
4.2
Scenarios Examined
The 2020 Irish system, as described in Chapter 3, was chosen as a test case for this study
because its unique features make it suitable for investigating base-load cycling. It is a
small island system, with limited interconnection to Great Britain, a large portion of
base-load plant and significant wind penetration. Thus, potential issues with cycling of
Chapter 4. Cycling of Base-load Plant
35
base-load units may arise on this system at a lower wind energy penetration, compared
to a larger, more interconnected or more flexible system. Two versions of the 2020
Irish system are discussed in Chapter 3, one with a 7.55 GW peak demand and the
other with a 9.6 GW peak demand. Both versions are examined in this chapter. For
each of the demand scenarios, three levels of installed wind generation, namely 2000,
4000 and 6000 MW, were examined. As seen in Chapter 3, the remaining generation
is primarily thermal generation, with a small portion of inflexible hydro capacity while
the base-load is composed of coal and combined-cycle gas turbine (CCGT) generation.
The characteristics of a typical base-load CCGT and coal unit on the test systems are
shown in Table 4.1.
Table 4.1: Characteristics of a Typical CCGT and Coal Unit on the Test System
Characteristic
CCGT
Coal
Maximum Power (MW)
400
260
Minimum Power (MW)
200
105
Maximum Efficiency (%)
57.6
36.9
Hot Start-up Cost (e)
13,280
5,320
Full Load Cost (e/hour)
15,900
4,880
(% of Max Power)
9
13
Minimum Down Time (Hour)
1
5
Start-up time (Hour)
2
5
Maximum Spinning Reserve Contribution
The Wilmar model was run deterministically (i.e. the expected value of wind and
load is planned for), for one year, for each of the three wind cases, and for both levels
of peak demand in order to examine the effect that increasing wind power penetration
will have on the operation of base-load units. These are the units with the most limited
operational flexibility, and as such, will suffer the greatest deterioration from increased
cycling. A sensitivity analysis was conducted to investigate the role that storage and
interconnection play in altering the impact of increasing wind penetration on base-load
operation. This involved running the model deterministically for one year, for each of
the three wind cases, first, without any pumped storage on the system, and second
Chapter 4. Cycling of Base-load Plant
36
without any interconnection on the system. In order to fairly compare systems without
storage/interconnection to the systems with storage/interconnection, the systems must
maintain the same level of reliability. Thus it was necessary to replace the pumped
storage units and interconnection with conventional plant. The 292 MW of pumped
storage was replaced with three 97.3 MW open cycle gas turbine (OCGT) units while
the 1000 MW of interconnection was replaced with nine 100 MW OCGT units (as 100
MW is always used as spinning reserve, the maximum import capacity is 900 MW). The
characteristics of these substitute units were set such that they could deliver the same
amount of generation over the same time period as the interconnection/storage units
that they replaced. The OCGT units which replaced the storage units were capable of
delivering the same amount of spinning reserve (132 MW in total). The OCGT units
that replaced the interconnection did not contribute to spinning reserve but instead
100 MW was subtracted from the demand for spinning reserve in each hour. This is
the assumption used when the interconnector is in place.
The cost per MWh from the OCGT units is generally greater than the cost of
imports or production from the storage units, thus the production previously provided
from storage/interconnection is not shifted directly to these OCGT units. This is
advantageous in this type of study, as the operation of other units on the system
without storage/interconnection can be observed, whilst the system adequacy is not
undermined by the reduced capacity, thus facilitating the sensitivity analysis. For
example, had CCGT capacity been used to replace the interconnector, it would likely
provide the energy that had been previously delivered by the interconnector, but this
would not allow examination of how the existing units on the system are affected
in the absence of interconnection. The results from the systems without storage and
interconnection were compared to the base case (i.e. with storage and interconnection).
To examine the results, the base-load units were categorized as coal or CCGT. The
results for the individual units in each group were normalized by their capacity to
obtain the result per MW for each unit. The average result per MW was then obtained
and this was multiplied by the capacity of a typical coal or CCGT unit (chosen to be
260 MW and 400 MW respectively) to give the result for a typical coal or CCGT unit
Chapter 4. Cycling of Base-load Plant
37
as shown below:
Pn
i=1 (xi /ci )
n
∗ T ypical U nit Size
(4.1)
where xi is the result for the ith unit, ci is the capacity of the ith unit and n is the
number of units
4.3
4.3.1
Results
Increasing Wind Penetration and the Operation of Base-Load
Units
As the penetration of wind generation on a power system is increased, large fluctuations
in the net load (load minus wind generation) will occur more frequently, as seen in Table
4.2 and Table 4.3, which shows the annual number of hourly net load ramps which
exceed 1000 MW on the 7.55 and 9.6 GW peak demand test systems. (The probability
distribution for net load ramps can also be found in Appendix A and shows larger net
load ramps occur more frequently on the 9.6 GW peak demand system relative to the
7.55 GW peak demand system.)
Table 4.2: No. hours when net load changes by >1000 MW from previous hour on the
7.55 GW peak demand system
Wind energy penetration
7.55 GW peak system
15%
29%
43%
90
135
211
Table 4.3: No. hours when net load changes by >1000 MW from previous hour on the
9.6 GW peak demand system
Wind energy penetration
11%
23%
34%
9.6 GW peak system
277
342
454
In addition, as wind generation is modelled as having zero operating costs, produc-
Chapter 4. Cycling of Base-load Plant
38
Table 4.4: Number of thermal units online with increasing wind penetration (averaged
at each hour shown over the year)
Time (Hour)
00
03
06
09
12
15
18
21
15% wind energy penetration
12.9
11.1
12.3
16.1
16.5
16.1
17.2
15.5
29% wind energy penetration
12.3
10.8
11.6
14.6
15.2
14.9
15.7
14.6
43% wind energy penetration
11.6
10.5
11.1
13.9
14.3
13.8
14.6
13.6
tion from thermal units is increasingly displaced, thus the number of units online will
decrease. This is shown for the 7.55 GW peak demand system, in Table 4.4. Therefore,
with less units online to manage growing fluctuations in net load, the onus on thermal
units becomes more demanding with increasing wind penetration.
Figure 4.1 and Figure 4.2 show the annual number of start-ups and capacity factor
for an average sized CCGT (400 MW) and coal unit (260 MW), as wind penetration
increases on the 7.55 and 9.6 GW peak demand systems respectively. The capacity
factor is defined as the ratio of actual generation to maximum possible generation in
a given time period (in this case over the test year). As the wind energy penetration
grows and the variability and unpredictability involved in system operation is increased,
the operation of a base-load CCGT unit is severely impacted. Moving from 15% to 43%
wind energy penetration the annual start-ups for a typical base-load CCGT unit rise
from 67 to 107, an increase of 60%. On the 9.6 GW peak demand system, annual CCGT
start-ups increase from 31 to 86, a 177% increase, as wind energy penetration increases
from 11% to 34%. This increase in CCGT start-ups corresponds to a plummeting
capacity factor for the units as seen in Figure 4.1 and Figure 4.2, as increasing levels of
wind power will displace production from CCGT units and force them closer to midmerit type operation. The start-ups are higher and the capacity factor is lower for a
typical CCGT unit on the 7.55 GW peak demand system relative to the 9.6 GW peak
demand system, as CCGTs will more frequently be the marginal units on the system
with less demand. (Not shown in Figures 4.1 and 4.2 are those CCGT units on the
system, originally built for base-load operation, but having over time been displaced
into mid-merit operation. With increasing penetration of wind power, such units also
Chapter 4. Cycling of Base-load Plant
39
Figure 4.1: Annual number of start-ups and capacity factor for an average CCGT and
coal unit with increasing wind penetration on the 7.55 GW peak demand system
tend to have a decreasing capacity factor, however their annual number of start-ups,
which are much larger than a typical base-load CCGT shown in Figure 4.1 and 4.2,
actually reduce as they are forced from mid-merit into peaking operation.)
As wind generation on the system increases, the timing and predictability of when
CCGT units will be started is also impacted, as seen in Table 4.5, which shows the
percentage of total CCGT starts that occur during each two hour interval over the
year, on the 7.55 GW peak demand system. With just 2000 MW installed wind power
(15% energy penetration) CCGT start-ups are seen to be concentrated around 6-7am.
However moving to 6000 MW installed wind power CCGT start-ups are now more
widely distributed throughout the day. This will have repercussions for plant personnel
who are responsible for plant start-ups and would indicate that stringent start-up procedures will need to be put in place for these units, to minimize the risk of difficulties
arising during the start-up process. In the UK market, for example, a generating unit
must be synchronized within a +/- five-minute window when delivering power to the
grid. If late, the grid operator may not accept the power at all, regardless of the fuel
and production costs already incurred (OSIsoft, 2007).
Unlike a CCGT unit, the annual number of start-ups for a typical coal unit on the
Chapter 4. Cycling of Base-load Plant
40
Figure 4.2: Annual number of start-ups and capacity factor for an average CCGT and
coal unit with increasing wind penetration on the 9.6 GW peak demand system
Table 4.5: Percentage of total start-ups occurring during each two-hour interval over
the year
Time (Hour)
00-01
02-03
04-05
06-07
08-09
10-11
2000 MW wind power
1.09
0.82
1.63
46.74
22.01
11.68
6000 MW wind power
1.03
2.07
7.76
28.79
23.45
9.66
Time (Hour)
12-13
14-15
16-17
18-19
20-21
22-23
2000 MW wind power
2.17
2.45
8.15
2.44
0.27
0.54
6000 MW wind power
3.10
6.38
14.14
1.21
1.72
0.69
7.55 GW peak demand system decreases somewhat as the wind energy penetration
increases, as seen in Figure 4.1. On the 9.6 GW peak demand system start-ups for a
coal unit increase with wind energy penetration up to 23% (albeit not as drastically as
a CCGT unit). However, at wind energy penetrations greater than 23%, this pattern
diverges and the start-ups for a coal unit begin to decrease, as seen in Figure 4.2. As
wind energy penetration grows, the demand for spinning reserve will increase. Due to
high part-load efficiencies, coal units are the main thermal providers of spinning reserve on this system. As CCGT units are taken offline more frequently with increasing
wind penetration, the requirement on coal units to provide reserve to the system is
driven even higher. Coal units also have lengthy start-up times; once taken offline it
Chapter 4. Cycling of Base-load Plant
41
is a minimum of ten hours (minimum down time plus synchronization time as seen
in Table 4.1) before the unit can be online and generating again. Thus on a system
with a high wind energy penetration, coal units are even less likely to be cycled offline
to avoid shortfalls in spinning reserve. This would indicate that the units with the
most limited operational flexibility may actually be rewarded at high levels of wind for
their inflexibility and suggests that some form of incentive may be needed to secure
investment in flexible plants (for example OCGTs), which are commonly reported as
being beneficial to system operation with large amounts of wind (Kirby and Milligan,
2008; Strbac et al., 2007). Coal units do, however, have low minimum outputs so at
times of high wind power penetration more coal units can remain online to meet the
minimum units online constraint, thus minimizing wind curtailment. CCGT units, on
the other hand, are typically restricted by high minimum outputs because of emissions restrictions as opposed to physical limitations. When running base-load, CCGTs
achieve high firing temperatures which allows CO (carbon monoxide) to be oxidized
into CO2 . However, at part load levels, when the firing temperature is lower, the CO
to CO2 oxidation reaction is quenched by cool regions near the walls of the combustion
liner resulting in increased levels of CO (Siemens, 2008a).
It would appear from examination of capacity factors in Figure 4.1 and Figure 4.2
that a crossover point exists when coal units become the most base-loaded plant on
the system. It is clear from Table 4.1, that coal generation is cheaper than generation
from CCGT units and so these units are in fact the most base-loaded plant at all wind
energy penetrations examined, however they are modelled as having more frequent
outages compared to the CCGT plant, thus yielding relatively lower capacity factors
(at some wind energy penetrations) compared with the CCGT units.
Figures 4.3 and 4.4 show the utilization factor for an average base-load coal and
CCGT unit, and the number of hours they perform severe ramping as wind penetration
increases. The utilization factor is the ratio of actual generation to maximum possible
generation during hours of operation in a given period. Severe ramping is defined here
as a change in output greater than half the difference between a unit’s maximum and
minimum output over one hour. Periods when the unit was starting up or shutting down
Chapter 4. Cycling of Base-load Plant
42
Figure 4.3: Utilization factor and annual number of hours where severe ramping is
performed for an average CCGT and coal unit with increasing wind penetration on the
7.55 GW peak demand system
were not included. Although coal units, as the most base-loaded thermal generation,
will avoid heavy start-stop cycling as wind levels grow they do experience increased
part-load operation. This is indicated by a drop in utilization factor from 0.90 to 0.81,
or 0.92 to 0.88, as wind energy penetration increase from 15% to 43%, or 11% to 34%,
on the 7.55 and 9.6 GW peak systems respectively, as seen in Figures 4.3 and 4.4. The
utilization factor for a CCGT unit is also seen to decrease with increasing levels of wind
power, however, it remains high in comparison with a coal unit, indicating the relatively
smaller contribution to spinning reserve it provides to the system and correspondingly
the infrequent periods of part-load operation. As seen in both Figures 4.3 and 4.4,
both types of unit experience a dramatic increase in periods where severe ramping
is required as wind energy penetration increases. For both test systems CCGT units
experience more ramping as they are more frequently the marginal units on the system,
however the rate of increase in ramping is higher for the coal units as the number of
operating hours exceeds that for a CCGT as the wind energy penetration increases.
Such increases in part-load operation and ramping can lead to cycling damage such
as fatigue damage, boiler corrosion or cracking of headers, as discussed in Chapter 2.
A recent study of the impacts of ramping on three of Xcel Energy’s coal plants, for
Chapter 4. Cycling of Base-load Plant
43
Figure 4.4: Utilization factor and annual number of hours where severe ramping is
performed for an average CCGT and coal unit with increasing wind penetration on the
9.6 GW peak demand system
example, predicted a 200%-500% increase in variable O&M costs and capital expenses
(Danneman and Beuning, 2011).
The results reported here are for “average sized” CCGT and coal units. In order to
show how these results correspond to the actual results for the real units modelled, the
maximum, minimum, average and standard deviation of the number of start-ups and
capacity factor for the modelled CCGT and coal units are given in Appendix B.
4.3.2
Sensitivity Analysis
The previous section showed the serious impact increasing levels of wind power will
have on the operation of base-load units. The extent of this impact will be determined
by the generation portfolio and the characteristics of the system. This section provides
a sensitivity analysis of the effect of the portfolio on the results, by examining the
operation of the base-load units with increasing levels of wind power when storage and
interconnection are removed from the system.
Chapter 4. Cycling of Base-load Plant
4.3.2.1
44
No Storage Case
Figure 4.5 shows the number of hours online for an average CCGT and coal unit,
for increasing wind energy penetrations on the 7.55 GW peak demand system, with
and without pumped storage. Although storage will typically charge overnight when
prices are low, thus raising the base-load, Figure 4.5 reveals that base-load units in
fact spend more hours online on the system without pumped storage, compared to the
system with storage. Pumped storage units can provide spinning reserve to the system
when pumping and when generating, and as such they are large providers of spinning
reserve to the system. Their typical mode of operation is to charge at night, although
this is typically seen to be concentrated over a small number of hours, and generate
at minimum load, providing the maximum amount of spinning reserve possible to the
system (a maximum of 50% of the total spinning reserve demand can come from storage
units), throughout the day. On the system without pumped storage, this spinning
reserve must now be provided by conventional plant. The increased requirement on
base-load units to provide spinning reserve in the absence of storage is evident in Table
4.6, which shows the total amount of spinning reserve provided by a CCGT or coal
unit on the 7.55 GW peak demand system, with and without pumped storage. Thus
on occasions when CCGT or coal units may have been cycled offline on the system with
pumped storage, they will now be more likely to be kept online on the system without
pumped storage.
Table 4.6: Total contribution to spinning reserve (MWh) from typical CCGT and coal
unit on the 7.55 GW peak demand system
Installed wind capacity (MW)
With storage
Without storage
2000
4000
6000
CCGT
150,001
156,557
151,965
coal
166,069
167,295
173,303
CCGT
238,609
238,473
217,015
coal
223,884
228,462
225,772
As such, Figure 4.6, which shows the number of start-ups for a typical base-load
CCGT and coal unit on a system with and without pumped storage as wind penetration
Chapter 4. Cycling of Base-load Plant
45
Figure 4.5: Number of hours online for an average CCGT and coal unit with/without
storage and an increasing wind penetration on the 7.55 GW peak demand system
increases on the 7.55 GW peak demand test system reveals that without storage on the
system, both CCGT and coal units have reduced start-ups (although the difference is
small). However, although base-load units may benefit from less start-ups and more
hours online on the system without storage, the increase in reserve provision from these
units implies increased part-load operation, which has been shown to cause component
degradation in base-load plant. The HRSGs in CCGTs in particular can be affected by
flow instability, which is associated with part load operation (Wambeke, 2006). (Similar
results were obtained for the 9.6 GW test system and have been included in Appendix
C.)
4.3.2.2
No Interconnection Case
Figure 4.7 compares the number of hours spent online by a typical CCGT and coal
unit on the 7.55 GW peak demand system with and without interconnection, as wind
energy penetration is increased. The base-load units are seen to spend significantly
more hours online on the system without interconnection compared to the system with
interconnection. (A similar result was found for the 9.6 GW test system and this has
been included in Appendix C.) Due to a large portion of base-load nuclear plant and
Chapter 4. Cycling of Base-load Plant
46
Figure 4.6: Number of start-ups for an average CCGT and coal unit with/without
storage and an increasing wind penetration on the 7.55 GW peak demand system
cheaper gas prices compared with Ireland, the market price for electricity tends to be
cheaper in Great Britain. As a consequence Ireland tends to be a net importer of
electricity from Great Britain and as such will often favour importing electricity before
turning on domestic units. Thus interconnection to Great Britain displaces conventional
generation on the Irish system, forcing units down the merit order and exacerbating
plant cycling. Without the option to import electricity, as shown in Figure 4.7, all
demand must be met by domestic units, requiring more units to be online generating
more often. Thus, a typical CCGT and coal unit are seen in Figure 4.7 to spend more
time online without interconnection, particularly the CCGT unit which is closer to
being the marginal unit and therefore its production is displaced ahead of production
from a coal unit. Likewise interconnection is seen to displace more base-load production
on the 7.55 GW peak demand system compared to the 9.6 GW peak demand system,
and at higher wind energy penetrations compared to lower wind energy penetrations,
as with less demand less conventional generation will be online and therefore base-load
units are closer to being the marginal units on the system.
On the 9.6 GW peak demand system, removing interconnection is seen in Figure 4.9
to also reduce plant cycling as domestic units were required to stay online. However,
Chapter 4. Cycling of Base-load Plant
47
Figure 4.7: Number of hours online for an average CCGT and coal unit with/without
interconnection and an increasing wind penetration on the 7.55 GW peak demand
system
as the wind energy penetration is increased, the electricity price in Ireland undercuts
British prices more often making exports economically viable and eventually a crossover
point is reached when the system with interconnection can deal with large fluctuations
in the wind power output via imports/exports more favourably and avoid plant shutdowns relative to the system without interconnection. Coal units, being the most
base-loaded units on the system, are first to benefit from increased exports on the
system and thus a crossover point can be observed for the coal units in Figure 4.9 at
34% wind energy penetration.
Likewise, for the 7.55 GW peak demand test system electricity prices in Ireland
tend to be lower than British prices so exports to Britain are up to four times greater
than on the 9.6 GW peak demand system. Thus coal units are seen in Figure 4.8 to
benefit from reduced start-ups relative to the case without interconnection. However,
similar to the 9.6 GW peak demand system, at lower wind energy penetrations the
CCGT units experience less cycling on the system without interconnection, until again
a crossover point occurs, this time at 35% wind energy penetration, beyond which the
system with interconnection benefits from reduced CCGT cycling.
Chapter 4. Cycling of Base-load Plant
48
Figure 4.8: Number of start-ups for an average CCGT and coal unit with/without
interconnection and an increasing wind penetration on the 7.55 GW peak demand
system
A further sensitivity was conducted to examine base-load cycling when CCGT generation were the most base-loaded plant on the system. This was conducted for the
7.55 GW peak demand system with 6000 MW installed wind capacity and with no interconnection, so that changes to plant operation could be directly attributable to the
change in the merit order rather than a change in the operation of the interconnector.
In this sensitivity analysis the cost of coal generation was increased such that it was
more expensive than generation from base-load CCGT units (but still less expensive
than mid-merit CCGTs). Such a scenario is not unrealistic given the current trend for
low gas prices and the rising cost of coal generation due to environmental restrictions
(Carrino and Jones, 2011). As seen in Table 4.7, the results showed drastic increases in
cycling, not only for coal plant, but CCGT plant also. During periods of very low net
load (i.e. high wind generation) it becomes difficult to meet the minimum number of
units online constraint while also avoiding curtailment of wind generation. As CCGT
units have high minimum loads relative to coal units, they cannot reduce their output
sufficiently to accommodate the wind power, forcing them to be cycled off-line and
requiring other units, in this case the coal units, to be started. The result is greatly
increased cycling for both types of units.
Chapter 4. Cycling of Base-load Plant
49
Figure 4.9: Number of start-ups for an average CCGT and coal unit with/without
interconnection and an increasing wind penetration on the 9.6 GW peak demand system
Table 4.7: Annual start-ups for a typical CCGT and coal unit on 7.55 GW peak demand
system with 6000 MW installed wind power and no interconnection
4.3.3
Coal most
CCGT most
base-load generation
base-load generation
CCGT starts
130
204
Coal starts
47
214
Total
117
418
Effect of Modelling Assumptions
The results in this chapter were produced by running the Wilmar model in deterministic
mode, i.e. the model planned for the expected values of load, wind generation and
demand for replacement reserve. It is considered that this mode is most representative
of current practice. However, in the future as higher wind energy penetrations are
reached, stochastic scheduling is likely to be implemented to provide schedules that are
more robust, thus maintaining reliable system operation when large forecast errors may
occur. To determine the impact that stochastic scheduling will have on the operation
of base-load plant further simulations were run using the Wilmar model in stochastic
mode, for the 7.55 GW and 9.6 GW peak demand systems, each with 2000, 4000 and
Chapter 4. Cycling of Base-load Plant
50
6000 MW installed wind power. Figure 4.10 shows the difference in annual start-ups
that was found for a typical CCGT and coal unit, at each of the wind penetrations,
when optimized deterministically and stochastically on the 7.55 GW peak demand
system.
As can be seen there is relatively little difference between the two optimization
methods at the lower wind energy penetrations, but at 43% wind energy penetration
(6000 MW installed) the system optimized stochastically has slightly increased starts
(+11 for CCGT, +2 for coal). When optimized stochastically, the model must find a
solution that satisfies multiple scenarios, covering high and low net loads. Units with
long start-up times, i.e. base-load units, are therefore more often committed when the
system is optimized stochastically, because if they are required for any of the scenarios
they will have to be committed in advance. No decision has to be made in advance for
fast start units, on the other hand, as these can be started in a given hour, when the
load and wind generation are known. At lower wind penetrations, as compared to high
wind penetrations, it is more likely that the committed base-load generation will be
required as the net demand will simply be higher. However, with a high wind energy
penetration, base-load generation that has been committed to meet a high net demand
scenario, may in fact not be needed, if the net demand that is realised is low. This may
lead to these units being shut-down more frequently, as seen in Figure 4.10, although
the difference is relatively small.
4.4
Summary
Increasing penetration of wind generation on a power system will lead to changes in the
operation of the thermal units on that system, but most worryingly to the base-load
units. The base-load units are impacted differently by increasing levels of wind, depending on their characteristics. CCGT units see significant increases in start-stop cycling,
plummeting capacity factors and are essentially displaced into mid-merit operation. On
the test systems examined coal units are the main thermal providers of spinning reserve
to the system and also are highly inflexible and as a result avoid start-stop cycling but
Chapter 4. Cycling of Base-load Plant
51
Figure 4.10: Annual start-ups on the 7.55 GW peak demand system, optimized stochastically and deterministically
see increased part-load operation and ramping. This increase in cycling operation can
over time lead to increased forced outages and plant depreciation.
Certain power system assets are widely reported to assist the integration of wind
power. This chapter examined if pumped storage and interconnection reduced cycling
of base-load units by comparing a system with storage and interconnection to a system
without either, across a range of wind penetrations. It was found that in the absence
of storage there was a greater requirement to keep base-load units online to meet the
system’s spinning reserve requirement. Thus, base-load units were seen to be cycled
less on a system without pumped storage, compared to a system with pumped storage.
For a system with a high electricity price relative to its neighbours, interconnection was
found to displace generation from domestic units. As such, base-load units were also
seen to be cycled less on a system without interconnection compared to a system with
interconnection.
In the long term if power systems are to include large portions of variable wind
power, a flexible plant portfolio will be needed. As shown in this chapter, a unit
that is highly inflexible but provides a large portion of spinning reserve to the system
will benefit from its inflexibility by being kept online more. It is also possible that
generators that are repeatedly cycled would alter the technical characteristics of the
plant which are bid into the market, such as minimum down time or ramp rates,
Chapter 4. Cycling of Base-load Plant
52
in an attempt to avoid or minimise cycling. This would indicate that in order to
incentivise new plant to be flexible the revenue streams available to a unit may need
to be adjusted to reflect the value of flexibility. Some markets include a capacity
payment in order to incentivise generators to be available as much as possible. As
power systems evolve to include greater penetrations of wind, these payments could be
restructured in order to incentivise generator performance, such that new plant is more
adequately designed to deal with cycling. New ancillary services could also be defined
and increasing the ancillary services fund could also incentivise operational flexibility.
For example, ramping payments to generators for providing ramping service has been
proposed for the Ontario power system (APPrO, 2006).
CHAPTER
5
Multi-mode Operation of Combined-Cycle Gas Turbines
5.1
C
Introduction
OMBINED -cycle gas turbines (CCGTs) are a type of power generating unit
that achieve high efficiencies (up to 61%) by capturing the waste heat from
a gas turbine in a heat recovery steam generator (HRSG) and using it to produce
superheated steam to drive a steam turbine (Kehlhofer et al., 2009). The high efficiencies achieved, combined with their ease of installation, short-build times and relatively low gas prices have made the CCGT a popular technology choice (Watson, 1996;
Colpier and Cornland, 2002). In the Republic of Ireland, for example, 43% of the installed thermal capacity is CCGT technology, whilst in the markets of Texas (ERCOT)
and New England (NEPOOL) CCGTs represent 37% of the total installed capacity.
The operational flexibility of a CCGT unit is limited by the steam cycle, which
contains many thick-walled components, necessary to withstand extreme temperatures
53
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
54
Figure 5.1: Schematic of CCGT in open- and combined-cycle mode (Eskom, 2007)
and pressures (Shibli and Starr, 2007; Starr, 2003). To avoid differential thermal expansion across these components and the subsequent risk of cracking, these components
must be brought up to temperature slowly, resulting in slower start-up times and ramp
rates for the unit overall (Anderson and van Ballegooyen, 2003). Although, as CCGT
units were traditionally base-loaded, this was not a major concern for plant operators.
However, by incorporating a bypass stack upstream of the HRSG at the design stage,
as shown in Figure 5.1, a CCGT unit has the option to bypass the steam cycle and
run in open-cycle mode, whereby exhaust heat from the gas turbine is ejected directly
into the atmosphere via the bypass stack (Anderson and van Ballegooyen, 2003). This
reduces the power output and efficiency of the plant but offers greater operational flexibility. Running in open-cycle mode, the gas turbine has a short start-up time of 15
to 30 minutes and is capable of changing load quickly. However, bypass stacks are not
always incorporated because they can potentially lead to leakage losses, thus reducing
plant efficiency, while also introducing additional capital costs (Kehlhofer et al., 2009).
As discussed in Chapter 1, international energy policy is driving ever greater penetrations of renewable energy and thus wind power is set to represent a larger portion
of the future generation mix (Bird et al., 2005). This is driving a greater demand for
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
55
flexibility within power systems in order to deal with high penetrations of variable and
difficult to predict energy sources (IEA, 2008; Van Hulle and Gardner, 2008). Storage,
interconnection and responsive demand are commonly cited as flexible options for dealing with variability issues (Brown et al., 2008; Göransson, 2008; Hamidi and Robinson,
2008) however these options have considerable costs associated with them. Facilitating
open-cycle operation of CCGT units that have the technical capability to run in opencycle mode (i.e. those with a bypass stack) can also deliver much needed flexibility to a
system with a high wind penetration. This resource is often technically available, but
inaccessible due to market arrangements.
For example, in SEM (Single Electricity Market), the electricity market of Northern
Ireland and the Republic of Ireland, generators submit technical (operating characteristics) and commercial (cost characteristics) data day-ahead and the cheapest generators
are dispatched on the trading day until the demand is met (EirGrid and SONI, 2010b).
The current market rules do not facilitate multiple bids from CCGT units which are
capable of open-cycle operation. Instead these units can bid into the market day-ahead
either their combined-cycle or open-cycle characteristics, but not both at the same
time.
In order to derive the greatest benefits from a CCGT unit that can run in open-cycle
mode, it is necessary for the scheduling algorithm to explicitly consider both modes
of operation for the unit, i.e. open-cycle and combined-cycle (Lu and Shahidehpour,
2004). These will have greatly different technical and cost characteristics and so need
to be declared individually. Currently most markets do not facilitate CCGT units to
submit multiple bids representing different modes of operation, thus presently opencycle operation of a CCGT unit is typically limited to periods when the steam section
is undergoing maintenance. However, some US systems have begun addressing this
issue to varying degrees, with ERCOT and CAISO seeking to implement configuration
based modelling of CCGTs (Blevins, 2007; CAISO, 2010b).
The option to run in open-cycle mode could also provide benefits for the generators.
Renewable integration studies have shown that CCGT units will experience signifi-
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
56
cant decreases in running hours and thus will receive less revenue from the market
as they are displaced by greater levels of wind generation which has an almost zero
marginal cost (CAISO, 2010a; Göransson and Johnsson, 2009; NREL, 2010; NYISO,
2010; Troy et al., 2010). Due to their high minimum loads CCGTs are shut down
frequently with high wind penetrations as they cannot reduce output sufficiently to
accommodate the wind power output (Troy et al., 2010). By facilitating CCGT units
to operate in open-cycle mode, these units may have a new opportunity to capture
revenue from increased operation during periods when they might otherwise be offline.
For example, if a CCGT unit has been forced offline by high wind generation on the
system, it may have the opportunity to run as a peaking unit.
Multi-mode operation may also lead to a reduction in plant cycling. Online CCGT
units which have bypass stacks can instantaneously switch to open-cycle operation,
while remaining online, by opening the bypass damper to release exhaust gases through
the bypass stack. This could allow the gas turbine to remain online during periods when
the CCGT would otherwise be shut-down for minimum load reasons, thereby reducing
start-ups for the gas turbine. Likewise, offline CCGT units with bypass stacks can
start-up in open-cycle mode and the steam unit can be warmed slowly to be brought
into operation at a later point.
This chapter examines if a power system with a high wind penetration can benefit
from the additional flexibility introduced, or if the CCGT units themselves benefit,
when they are facilitated to operate in open-cycle mode when technically feasible and
economically suitable. As discussed in Chapter 3, the all-island Irish 2020 system
(AIGS, 2008) is expected to contain both a large share of wind power and CCGT units
(50% of which include a bypass stack) and thus provides an appropriate test system.
5.2
Methodology
In order to examine the potential for multi-mode operation of CCGT units some changes
were made to the Wilmar model. A set, ‘ccgt’, of all CCGT units capable of prolonged
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
57
open-cycle operation, i.e. those with bypass stacks, was defined. The set ‘ccgtopen
’ cora
responds to these CCGT units when run in open-cycle mode. CCGT units comprised
of two or more gas turbines will have multiple ‘ccgtopen
’ units, as indicated by index
a
‘a’. The relation ‘multi-mode’ is defined to pair each member of ‘ccgt’ with the corresponding member(s) of ‘ccgtopen
’. To ensure the mutually exclusive operation of these
a
‘ccgt’ units and the corresponding ‘ccgtopen
’ units, the constraint shown in (5.1) was
a
added to the model, where VOnline is the state binary variable which describes the online status of the unit. This allows the model to dispatch, when economically optimal,
either the ‘ccgt’ (combined-cycle mode) or any/all of the corresponding ‘ccgtopen
’ units
a
(open-cycle mode), for all scenarios ‘s’ and time steps ‘t’, but not both simultaneously
as they are in reality the same unit.
Online
Online
open ≤ 1,
Vs,t,ccgt
+ Vs,t,ccgt
a
∀ s, t, multi − mode(ccgt, ccgtopen
)
a
(5.1)
Equation (5.2), taken from (Arroyo and Conejo, 2004), sets the state binary variShut
ables VStart
s,t,i or Vs,t,i equal to 1 for all units ‘i’, when a unit is started up or shut down
respectively.
Start
Shut
Online
Online
Vs,t,i
− Vs,t,i
= Vs,t,i
− Vs,t−1,i
(5.2)
When modelling multi-mode operation of CCGT units two new circumstances arise
when calculating the start-up fuel consumption, FuelStart
s,t,i , which must be explicitly
represented. Firstly, when a ‘ccgt’ unit transitions from conventional combined-cycle
operation into open-cycle operation no start-up fuel is consumed by the ‘ccgtopen ’ unit as
represented by inequality (5.3), where Startfueli is the start-up energy used by each unit
(measured in MWh). When the ‘ccgtopen ’ unit starts from zero production (VStart
s,t,ccgtopen
a
= 1 and
VShut
s,t,ccgt
= 0), the first term on the right hand side of inequality (5.3) determines
the fuel used by the unit whilst the second term equals zero. Alternatively, when the
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
58
unit switches from combined-cycle to open-cycle operation (VStart
= 1 and VShut
s,t,ccgt
s,t,ccgtopen
a
= 1) the second term causes the right hand side of (5.3) to equal zero. Setting FuelStart
s,t,i
as a positive variable and using an inequality condition ensures that when a ‘ccgt’ unit
is shutting down and the corresponding ‘ccgtopen ’ unit is not starting up FuelStart
s,t,ccgtopen
a
will be 0.
Start
Start
open ≥ (Startf uelccgtopen ∗ V
)
F uels,t,ccgt
s,t,ccgtopen
a
a
a
Shut
− (Startf uelccgtopen
∗ Vs,t,ccgt
)
a
(5.3)
The second circumstance relates to the unit transitioning from open-cycle to combinedcycle operation. In this case the start-up fuel consumed is less than the start-up fuel
used in bringing the CCGT online from zero production, as some of this start-up fuel
has already been used to bring the unit online in open-cycle mode and the gas section
of the plant is in a hot state. As an approximation, the start-up fuel used to bring the
unit into combined-cycle operation from open-cycle operation is the difference between
the start-up fuel for the ‘ccgt’ and a fraction, α, of the start-up fuel for the ‘ccgtopen ’,
as seen in (5.4). Based on the operating experience of generators, α was chosen to
be 0.5 here. When the ‘ccgt’ unit is started from zero production (VStart
s,t,ccgta = 1 and
VShut
= 0), the first term on the right hand side of (5.4) provides the start-up
s,t,ccgtopen
a
fuel consumed whilst the second term equals zero. When the unit switches from opencycle to combined-cycle operation the second term is included, thus approximating the
start-up fuel consumed in this situation.
Start
Start
F uels,t,ccgt
≥ (Startf uelccgt ∗ Vs,t,ccgt
)
Shut
open ∗ α)
− (Startf uelccgtopen
∗ Vs,t,ccgt
a
a
(5.4)
In the Wilmar model any unit can contribute to the target for replacement (nonspinning) reserve, provided that an offline unit can come online in time to provide
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
59
reserve for the hour in question and the reserve available from an online unit is not
needed to meet spinning reserve targets. In Wilmar, the contribution from online and
f
offline units to the replacement reserve target, POf
s,t,i (MW), are calculated individually.
In this case the ‘ccgt’ units cannot provide offline replacement reserve as they have
long start-up times, but the corresponding ‘ccgtopen ’ units can, given their fast start-up
times. The constraints shown in (5.5) and (5.6), where Pmax
is a unit’s maximum
i
capacity (MW), ensure that if either the ‘ccgt’ unit or the ‘ccgtopen ’ unit is online,
then the ‘ccgtopen ’ unit cannot contribute to the portion of replacement reserve that is
provided from offline units. This is necessary to avoid the situation where a ‘ccgt’ unit
is online and the model allows the corresponding ‘ccgtopen ’ unit to contribute to offline
replacement reserve.
Of f
Online
max
∗ (1 − Vs,t,ccgt
)
Ps,t,ccgt
open ≤ P
a
ccgtopen
a
(5.5)
Of f
max
Online
open )
Ps,t,ccgt
∗ (1 − Vs,t,ccgt
open ≤ P
ccgtopen
a
a
(5.6)
a
a
When the bypass stack is utilized to switch from combined-cycle to open-cycle operation, the transition is automatic and occurs without shutting down the gas turbine or
reducing its power output. However, the transition from open-cycle to combined-cycle
operation is dependent on the temperature state of the boiler. Therefore, if the CCGT
unit has been operating for a period of time in open-cycle mode and is then scheduled
to switch to combined-cycle mode, its output must adjust in order to achieve the correct
HRSG inlet temperature, as depicted in Figure 5.2. This was implemented by setting
the allowable power output (PU (i) from (Arroyo and Conejo, 2004)) for each interval
of the CCGT’s start-up process, which begins at hour 0 in Figure 5.2, such that the
appropriate soak time is achieved.
Scheduled outages for each unit, determined from historical experience (AIGS, 2008),
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
60
Figure 5.2: CCGT start-up from open-cycle mode
are inputted in time-series format to the Wilmar model. In this case, CCGT units with
the capability to operate in open-cycle mode are considered to be available to run in
open-cycle mode for a portion of their scheduled outage. Given that gas turbine equipment is more accessible and compact in comparison with the steam turbine equipment,
it was assumed that one third of the maintenance period was sufficient for the gas
turbine.
5.3
Test System
The test system used was the 7.55 GW peak demand test system as set out in Chapter 3. Five (of the ten) CCGT units on the Irish system include bypass stacks and
therefore can run in open-cycle mode. Each of these units is currently installed and
operational. The characteristics of these units in combined-cycle mode are given in
Table 5.1. Limited data was available for these units in open-cycle mode so each was
given characteristics similar to a typical open-cycle gas turbine (OCGT) unit, as shown
in Table 5.1. As CCGT 2 and CCGT 5 are comprised of two gas turbines connected to
one steam turbine (2+1 configuration), these units were modelled as having two iden-
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
61
tical open-cycle units available for dispatch when the CCGT is operated in open-cycle
mode. CCGTs 2 and 3, located in Northern Ireland and CCGTs 1, 4 and 5, located
in the Republic of Ireland contribute to the minimum units online constraint which is
modelled in Wilmar (as discussed in Chapter 3), for their respective regions.
Table 5.1: Characteristics of CCGT units (capable of multi-mode operation) in
combined- and open-cycle modes
CCGT
Configuration
Max output (MW)
Min output (MW)
Max efficiency (%)
Min up time (Hours)
Min down time (Hours)
Start-up time (Hours)
Hot start-up fuel (GJ)
Max spinning reserve
contribution (MW)
Efficiency at max
spinning reserve (%)
Max output (MW)
Max efficiency (%)
Min up time (Hours)
Min down time (Hours)
Start-up time (Hours)
Hot start-up fuel (GJ)
Max spinning reserve
contribution (MW)
Efficiency at max
spinning reserve (%)
5.4
1
2
3
4
5
1+1 2+1 1+1 1+1
2+1
Characteristics in combined-cycle mode
445
480
404
343
480
240
232
260
220
280
57.6 58.9 53.9 52.9
52.3
4
4
6
4
4
1
2
4
4
2
2
1
1
2
4
2600 2000 1080 1732
2000
42
37
40
57.4 58.1 52.8
Characteristics in
280
160
256
39.5
38
39.3
0
0
0
0
0
0
0
0
0
14
8
13
25
52.2
open-cycle
265
39.3
0
0
0
13
40
51.3
mode
160
38
0
0
0
8
20
20
20
20
20
39.3
37.5
39.1
39.2
37.5
Results
A number of model runs were conducted to investigate the potential for multi-mode
operation of CCGT units. The Wilmar model was run in deterministic mode as this is
more representative of current scheduling practice. A year long dispatch was produced
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
62
for each of the three wind power penetrations outlined in Section III, when (i) multimode operation of CCGT units is not allowed and (ii) when multi-mode operation of
CCGT units is allowed.
5.4.1
Utilization of the Multi-mode Function
The average number of times a CCGT unit with multi-mode capability was run in
open-cycle mode and the average production from a CCGT in open-cycle mode over
the year, at each of the wind penetrations examined, is shown in Figure 5.3. Despite
increasing wind penetration being correlated with an increased demand for flexibility,
be it fast starting or ramping, Figure 5.3 shows the multi-mode function is used less
frequently as wind penetration on the system increases.
As more wind power, with an almost zero marginal cost, is added to a system,
the production from thermal plant is increasingly displaced and as such there is an
increased likelihood of generators operating at part-load. To illustrate this, Table 5.2
gives the annual utilization factor (ratio of actual generation to maximum possible
generation during hours of operation) averaged for the coal, CCGT and peat units on
the system with 2000, 4000 and 6000 MW wind power. Therefore, as wind penetration
increases, online part-loaded units are more often available to ramp up their output to
meet unexpected shortfalls in production, avoiding the need to switch on fast-starting
units, such as the CCGTs in open-cycle mode.
Table 5.2: Average utilization factors with increasing wind penetration
Installed Wind
Coal
CCGT
Peat
2000 MW
0.90
0.83
0.75
4000 MW
0.87
0.79
0.55
6000 MW
0.82
0.80
0.51
The trend seen in Figure 5.3 is consistent with the production from peaking plants
as wind penetration increases. Table 5.3 shows the drop in production from the most
utilized OCGT unit, with increasing wind penetration when multi-mode operation
is, and is not, allowed. Reduced production from peaking plants due to increased
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
63
Figure 5.3: Average production from a CCGT in open-cycle mode (line) and average
number of instances generators utilized open-cycle operation (grey column), shown for
various levels of installed wind capacity
wind penetration has also been observed in other wind integration studies, such as
NYISO(2010), however, it is also likely that systems with base-load units that have
slower ramp rates than those examined in this study will rely on fast-starting units
(such as CCGTs in open-cycle mode) more often as wind penetration increases. (All
units on the test system are assumed to be capable of ramping from minimum to
maximum output in one hour or less.) The average production from the CCGT units
in open-cycle mode, as seen in Figure 5.3, is comparable with average production levels
from dedicated OCGT peaking plants on the system when multi-mode operation of
CCGTs is not enabled.
Table 5.3: OCGT production (GWh) with increasing wind penetration
Installed Wind
Multi-mode not allowed
Multi-mode allowed
2000 MW
8.5
2
4000 MW
3.9
0.2
6000 MW
3.4
0.3
As wind penetration increases so too will the demand for replacement reserve, due to
the increased forecast error. The replacement reserve target can be met by fast-starting
offline units or from excess spinning reserve, if available. If sufficient excess spinning
reserve is not available to meet the replacement reserve target, the model must ensure
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
64
a number of fast-starting units are offline and available for operation to maintain a
secure system. Consequently, as a result of maintaining the replacement reserve target,
production from fast-start units (such as the multi-mode units in open-cycle mode) is
reduced. Additional simulations were conducted for the various wind penetrations with
no replacement reserve target, to investigate the extent that maintaining replacement
reserve suppressed the multi-mode units from running in open-cycle mode. For many
systems, such as the Irish system, this is more representative of current practice, where
no replacement reserve target formally exists. Table 5.4 shows the difference in the
average open-cycle production from multi-mode units that results when no replacement
reserve targets are enforced.
Table 5.4: Difference in Open-cycle production (GWh) from multi-mode units with no
replacement reserve target enforced
Installed Wind
4 Production
2000 MW
16.9%
4000 MW
7.2%
6000 MW
-0.5%
As seen, in the absence of a target for replacement reserve, open-cycle production
from the multi-mode units is utilized substantially more for the 2000 MW and 4000
MW wind power scenarios. However, with 6000 MW wind power, due to more frequent
part-loading of units, there is more frequently an excess of spinning reserve on the
system, as well as off-line fast-starting units (as per Table 5.3) which can contribute
to the replacement reserve target. Thus with 6000 MW wind power, the replacement
reserve target has little effect on the open-cycle operation of multi-mode units. Table
5.5 shows the average surplus spinning reserve available and the average replacement
reserve target per hour for each of the wind cases examined.
Table 5.5: Average hourly surplus spinning reserve (MW) available and replacement
reserve target (MW)
Installed Wind
Surplus spinning reserve
Replacement reserve target
2000 MW
65
500
4000 MW
120
580
6000 MW
240
700
Similarly, if additional peaking capacity, lower in merit relative to the CCGT units
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
65
Figure 5.4: Combined-cycle capacity factor (dashed line) and open-cycle production
(solid line) for each CCGT with multi-mode capability for the 2000 MW wind power
system
in open-cycle mode, is added to the system, open-cycle operation from the multimode CCGTs increases as the new peaking plants are now kept offline to meet the
replacement reserve target instead of the CCGTs in open-cycle mode. To demonstrate
this, 4 new OCGT units were added to the test system and the model was run for
the 2000 MW installed wind power scenario. The results showed a 32% increase in
open-cycle production from multi-mode CCGTs.
Figure 5.4 shows the capacity factor for each CCGT in combined-cycle mode and
its production over the year in open-cycle mode for the 2000 MW wind power scenario.
An inverse relationship is evident between the open-cycle production from a CCGT and
the capacity factor of the CCGT, which indicates that usage of the multi-mode function
is related to the amount of time the CCGT is offline. The more often a CCGT is not in
operation but available for dispatch, the more opportunities it has to run in open-cycle
mode, and this relationship would be expected regardless of the plant portfolio.
The percentage change in total production (combined-cycle plus open-cycle) that
results when multi-mode operation of CCGTs is enabled is shown in Table 5.6, for
each of the wind penetrations examined. Multi-mode operation increased production
for CCGT5, the lowest merit CCGT, which was seen to utilize the function most
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
66
Table 5.6: Percentage change in total production when multi-mode is enabled, shown
for each wind penetration
Installed
Wind
CCGT1
CCGT2
CCGT3
CCGT4
CCGT5
2000
MW
5.5
0
-3.3
-1.4
13.3
4000
MW
5.4
0.1
5
-7.3
38.5
6000
MW
-7.5
-0.1
-2.5
-37.1
11.1
frequently across all the wind penetrations examined. Total production from CCGT3
and CCGT4, which are mid-merit CCGTs, is reduced in all cases but one. There is a
risk (particularly for CCGTs that are frequently the marginal unit on the system, such
as CCGT3 and CCGT4) when offering open-cycle operation, of being dispatched from
combined-cycle to open-cycle operation at times of low net demand (demand minus
wind generation) to alleviate minimum load issues and then losing out to another
generator that can come online faster/cheaper, when the net demand increases again.
However, it is also likely that in a market environment, generators would strategise
when they would offer this multi-mode capability to avoid losing out on production.
CCGT1, the highest merit CCGT, benefits from increased production when multi-mode
operation is enabled on the system with 2000 MW and 4000 MW installed wind power.
This is due to increased exports and reduced production from the other CCGTs, as
opposed to increased production in open-cycle mode.
5.4.2
Benefits Arising from Multi-mode Operation
The efficiencies of the OCGT peaking units on the system are comparable with the
CCGT units in open-cycle mode. However, the CCGT units running in open-cycle
operation are assumed to have a lower gas price, to reflect the advantage of long-term
contracts. Their open-cycle capacity (as seen in Table 5.1) is also larger than the capacity of the OCGTs (103.5 MW each) and they benefit from avoided start-up costs when
transitioning from combined-cycle mode. Thus, when multi-mode operation of CCGTs
was enabled, production from OCGT peaking plant tended to be substituted by pro-
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
67
Figure 5.5: Average production from OCGT peaking units in each wind power scenario,
with multi-mode operation of CCGTs not allowed (light grey) and allowed (dark grey)
duction from the CCGTs in open-cycle mode. Figure 5.5, which shows the average
production from OCGTs for each wind penetration level when multi-mode operation of
CCGTs is allowed and not allowed, illustrates this point. Assuming open-cycle production from CCGTs is more economic than production from OCGTs, as is the case here, it
is possible that by enabling multi-mode operation of CCGTs sufficient flexibility could
be extracted from a system’s portfolio of plant to avoid building additional peaking
units, or equally that OCGT units would no longer be able to cover their costs and
so would be forced to retire from service. Both situations may then lead to increased
production from CCGTs in open-cycle mode.
Table 5.7 shows the total shortfall in replacement reserve over the year and the
number of hours in which this occurred, for each of the wind penetrations examined,
when multi-mode operation of CCGTs is, and is not, allowed. The additional faststarting generation available to the system when multi-mode operation of CCGT units
is allowed significantly reduces the shortfall in replacement reserve. This contributes
to a more secure system by preventing capacity shortfalls when wind forecasts prove
to be overly optimistic and also indicates that, depending on the market structure,
the generators may benefit from an additional revenue stream, via ancillary service
payments for the replacement reserve provided.
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
68
Table 5.7: Magnitude and frequency of replacement reserve shortfall, shown for various
levels of installed wind
Installed
Wind
MW
2000
4000
6000
Multi-mode CCGT
not allowed
MWh
No. hours
1688.7
13
2972.9
17
609.9
13
Multi-mode CCGT
allowed
MWh
No. hours
861.4
3
880.2
5
7.6
1
In addition to enhanced system security, the additional flexibility available to the
system when multi-mode operation of CCGT units is allowed will also yield operating
cost savings. Table 5.8 shows the total system production cost savings achieved by
enabling multi-mode operation of CCGTs. The total system cost is made up of fuel,
carbon and start-up costs for the Irish and British system combined, as they are cooptimized. In this case, these savings were achieved at no additional cost as each of
the CCGTs is currently capable of multi-mode operation.
Table 5.8: Total system cost saving (Me) resulting from multi-mode operation of
CCGTs
Installed Wind
Reduction in costs
2000 MW
1.55
4000 MW
0.51
6000 MW
2.65
The availability of less expensive peaking capacity when multi-mode operation of
CCGTs is enabled will tend to reduce price spikes. In addition, the model includes cost
penalties (these are not included in the system production costs) if demand, spinning
reserve or replacement reserve targets are not met. There were no hours when demand
was not met. However, the reduction in hours when the replacement reserve target is
not met, achieved by enabling multi-mode operation, as seen in Table 5.7, consequently
reduces the number of hours when this cost penalty (e10,000 for the given hour) is
incurred. This is seen in Table 5.9 which shows the average electricity price (excluding
hours with a cost penalty imposed) and the number of hours when the electricity price
exceeded e500/MWh (including hours with a cost penalty imposed) when multi-mode
operation of CCGTs is allowed, and not allowed, for each of the wind penetration levels.
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
69
Table 5.9: Average price and frequency of price spikes (>e500/MWh)
Installed
Wind
MW
2000
4000
6000
Multi-mode CCGT
not allowed
Average
No. Price
Price (e/MWh)
Spikes
49.76
27
48.10
27
45.21
21
Multi-mode CCGT
allowed
Average
No. Price
Price (e)
Spikes
49.70
8
47.96
9
45.11
8
Figure 5.6: Change in exports across the interconnector when multi-mode operation of
CCGTs is enabled
A direct consequence of the reduced prices is seen in Figure 5.6 which shows the
change in exports over the interconnector from the Irish to British systems that result
when multi-mode operation of CCGTs is allowed, for each of the wind scenarios examined. A substantial increase in exports is seen as a result of enabling multi-mode
operation of CCGTs, as the number of time periods when the electricity price on the
Irish system is less than the British system increases. Imports are largely unaffected.
The operation of the interconnector in the scenarios when multi-mode operation of
CCGTs was not allowed is shown in Table 5.10.
The increase in exports, resulting from multi-mode operation of CCGTs being en-
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
70
Table 5.10: Operation of interconnector when multi-mode is not allowed
Installed Wind
Import (MWh)
Export (MWh)
2000 MW
658,561
3,859,473
4000 MW
1,776,893
2,418,475
6000 MW
3,339,921
1,598,117
abled, supports a reduction in the level of wind curtailment, as more power is exported
to the British system during periods of high wind generation, thus avoiding generator
minimum load issues. The reduction in curtailment was significant, approximately 57%
and 82% on the 2000 MW wind system with the interconnector traded intra-day and
day-ahead respectively, but the actual percentage of the annual wind energy this represented was small (<0.01%). However, given that network congestion issues are not
modelled here it is likely that real levels of wind curtailment would be more significant
and consequently a reduction in curtailment levels arising from multi-mode operation
of CCGT units would be more advantageous. The increase in exports also supports a
reduction in CO2 emissions as generation on the British system, which is more carbon
intensive relative to the Irish case, is displaced. The observed CO2 reduction resulting
from multi-mode operation of CCGTs is small (≈100,000 tonnes or <0.05% of total
Irish and British system emissions). However, it was achieved at no additional cost
to the consumer or the generators as, in this case, the infrastructure (i.e. the bypass
stacks) is already in place.
In addition to enabling a new mode of operation, allowing multi-mode operation
of CCGTs may reduce cycling operation of these units. Table 5.11 shows the number
of start-ups (in combined-cycle mode), the utilization factor and the average duration
of time spent offline for CCGTs 3, 4 and 5 for the 2000 MW installed wind power
scenario. These units are seen to benefit from a reduced number of start-ups which
not only implies a start-up fuel saving, but also a reduction in plant wear-and-tear. As
discussed in Chapter 2, it is difficult to quantify the value of a reduction in cycling but
some studies have indicated that an avoided start-up could save generators substantial
amounts (up to $500,000 for a single start/stop cycle) (Lefton et al., 1998). An increase
in the utilization factor for the CCGT units (in CCGT mode) is also observed when
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
71
multi-mode operation is allowed. This implies a reduction in part-load operation, which
is particularly beneficial for CCGT plant, given HRSG components are susceptible to
differential thermal expansion resulting from flow instability, as well as water chemistry
issues, when operated at part-load (Wambeke, 2006). As the time spent online increases
when multi-mode operation is allowed, the average duration of the offline period will
be reduced. If the duration of time spent offline decreases the plant is more likely to be
in a warmer state when it starts up, thus alleviating the level of creep-fatigue damage
associated with start-ups (Lefton et al., 1997).
Table 5.11: Impact of Multi-mode on CCGT 3, 4 & 5 with 2000 MW installed wind
power
Start-ups
Utilization Factor
CCGT 3
CCGT 4
CCGT 5
No multi-mode
257
157
29
multi-mode
247
141
21
No multi-mode
0.94
0.86
0.67
Multi-mode
0.94
0.87
0.72
Average Duration of
No multi-mode
19
46
284
Offline Period (Hours)
Multi-mode
20
41
46
5.4.3
Sensitivity Studies
Usage of the multi-mode function is dependent on many factors, particularly the amount
of flexibility already present in the system. A sensitivity study was conducted to examine usage of the multi-mode function when the system was less flexible to meeting
demand. This involved running the model with 2000 MW wind power (as this level of
wind generation led to the greatest usage of CCGTs in open-cycle mode) and power
exchange across the interconnector fixed day-ahead as opposed to intra-day. Examining
the usage of the multi-mode function when the interconnector is scheduled day-ahead
versus intra-day illustrates how a less flexible system will more frequently utilize the
flexibility present in multi-mode CCGT operation. Figure 5.7 shows the average production from the multi-mode CCGTs in open-cycle mode and the average number of
instances the CCGTs utilized open-cycle operation, with the interconnector scheduled
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
72
Figure 5.7: Average production from a CCGT in open-cycle mode (line) and average number of instances generators utilized open-cycle operation (grey column), with
interconnector scheduled day-ahead and intra-day on the 2000 MW wind system
day-ahead and intra-day on the 2000 MW wind power system. The average production
from CCGTs in open-cycle mode with day-ahead scheduling of the interconnector is
seen to be more than three times greater than the system with intra-day scheduling of
the interconnector. By fixing the power exchange between the Irish and British systems
day-ahead, when there is greater uncertainty in the expected wind generation and demand, the system is forced to dispatch generators such as the multi-mode CCGT units,
as opposed to rescheduling imports/exports, to compensate for wind and load forecast
errors. Likewise, systems with seasonal hydro restrictions may see greater usage of
multi-mode CCGT operation during those periods when the operating flexibility of the
system is reduced.
In addition, the quality of wind and load forecasts employed by a system will also determine the usage of the multi-mode function. Additional simulations were completed
running the model in stochastic and perfect foresight modes. These represent different
means of including load and wind forecasts in the scheduling process; whereby stochastic optimization can be considered to represent a system employing ensemble forecasts,
deterministic optimization is representative of a system utilizing a single forecast and
the perfect forecast scenario is a hypothetical case where no forecast error exists. The
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
73
Figure 5.8: Average production from CCGT in open-cycle mode (GWh), shown for
different methods of optimization with 2000 MW wind power
robust solutions obtained by stochastic optimization showed less deployment of the
multi-mode function compared with the deterministic results. The stochastic solution,
optimized over several wind and load scenarios, typically has more units online to cover
all scenarios and therefore is more prepared to deal with unforseen shortfalls in wind
generation or increases in demand, without the need for starting peaking plant. The
capacity factors of the CCGT units are also higher for the stochastic case compared
to the deterministic case indicating that there was also less opportunity for these units
to run in open-cycle mode when the system is optimized stochastically. Running the
Wilmar model with perfect foresight of the system demand and wind profile also reveals
even less open-cycle operation from CCGTs, as in this case, with no forecast errors on
the system (except forced outages of generators), fast starting units are in less demand
relative to the deterministically optimized solution. Figure 5.8 compares the average
open-cycle operation from the multi-mode CCGTs, on the system with 2000 MW wind
power, when optimized with perfect foresight, stochastically and deterministically. The
average open-cycle production from a CCGT unit is seen to be 11% less on the stochastically optimized system and 35% less on the system with perfect forecast compared to
the deterministic case.
A sensitivity analysis was also conducted using a higher level of demand on the
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
74
system. In this case the original demand profile from AIGS (2008) with a 9.6 GW
peak, discussed in Chapter 3, was run for each wind scenario. The average production
from the multi-mode CCGTs in open-cycle mode over the year is shown in Figure 5.9
to be six to eight times greater on the 9.6 GW peak demand system, where peaking
capacity is in greater demand, compared to the 7.55 GW peak demand system, at
each of the wind power penetrations examined. In addition to the increased demand
resulting in increased open-cycle production from the multi-mode CCGTs (as well
as combined-cycle production), the other main difference between the scenarios is the
predominant direction of power transfer on the interconnector. With 2000 MW installed
wind capacity the Irish system is a net importer of power from Britain, at both levels
of demand examined. However, as more wind power is installed on the 7.55 GW peak
demand system the marginal electricity price is reduced sufficiently with respect to the
British system such that Ireland becomes a net exporter of power. Although increasing
wind power penetration on the 9.6 GW peak demand system also reduces the marginal
price it is still a net importer with 6000 MW installed wind power. Thus, on occasions
when forecast wind is overestimated and the system is in need of fast-starting plant,
the 7.55 GW peak demand system, being a net exporter, can more frequently choose
to curtail exports or start up a unit to compensate. In contrast, the 9.6 GW peak
demand system, being a net importer, more often only has the option to turn on faststarting plant. Hence, this implies that a system which tends to be a net exporter is
inherently more flexible, and has more options for dealing with variable wind power
than a system that is a net importer of power. In this scenario with higher demand,
each of the multi-mode CCGT units experienced increased total production (combinedcycle plus open-cycle) when multi-mode operation was allowed, suggesting that offering
multi-mode capability may prove more profitable on a system with a smaller capacity
margin.
Given the low deployment of the multi-mode functionality on the 7.55 GW peak
demand system and the high capacity factor in combined-cycle mode for CCGT 1 and
2, as seen in Figure 5.4, it would appear that there is insufficient incentive for all
CCGTs capable of multi-mode operation to offer this flexible capability. Thus, given
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
75
Figure 5.9: Average production from a CCGT in open-cycle mode on the 7.55 GW
peak demand system (light grey) and the 9.6 GW peak demand system (dark grey),
shown for various levels of installed wind power
that CCGTs 3, 4 and 5 have low capacity factors in combined-cycle mode, additional
simulations were conducted to investigate the resulting benefits if these units alone,
and if CCGT 5 alone, offered multi-mode capability. Table 5.12 shows the total system
cost (for Ireland and Britain) and the magnitude of the replacement reserve shortfall
over the year for these configurations (in addition to other configurations examined in
the paper). Examining the shortfall in the replacement reserve target for the different
configurations reveals that the majority (≈ 80%) of the reduction in replacement reserve
shortfall due to multi-mode capability is attributable to CCGT 5, while CCGTs 1 and 2
are seen to have no impact on the replacement reserve shortfall. Thus, CCGTs capable
of open-cycle operation, which have very low output in combined-cycle mode, have
value in providing replacement reserve.
All cases with 2000 MW wind power
7.55 GW Peak, No Multi-mode
7.55 GW Peak, 5 Multi-mode CCGTs
7.55 GW Peak, 3 Multi-mode CCGTs (3, 4 & 5)
7.55 GW Peak, 1 Multi-mode CCGT (5)
7.55 GW Peak, No Multi-mode, day-ahead interconnector trading
7.55 GW Peak, 5 Multi-mode CCGTs, day-ahead interconnector trading
7.55 GW Peak, No Multi-mode, stochastic
7.55 GW Peak, 5 Multi-mode CCGTs, stochastic
7.55 GW Peak, No Multi-mode, perfect foresight
7.55 GW Peak, 5 Multi-mode CCGTs, perfect foresight
9.6 GW Peak, Multi-mode not allowed
9.6 GW Peak, 5 Multi-mode CCGTs
Configuration
Total System
Cost
/ Saving
Me
13372.03 / 13370.48 / 1.55
13368.99 / 3.04
13371.73 / 0.3
13384.64 / 13382.98 / 1.66
13371.23 / 13371.27 /-0.04
13370.87 / 13369.38 / 1.49
13997.24 / 13996.16 / 1.08
Replacement
Reserve
Shortfall
MWh
1688.7
861.4
861.4
1032.4
2197.9
798
966.5
394
0
0
68345.9
63265.3
Avg. Top-up
Payment
(no. units)
Me
1.36 (2)
0.49 (3)
0
1.66 (2)
0.91 (2)
0.45 (1)
0
Table 5.12: Total system cost, replacement reserve shortfall and top-up payment, shown for various multi-mode configurations
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
76
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
77
As seen in Table 5.6, the multi-mode CCGTs may experience a reduction in total
production as a result of offering multi-mode capability to the market. This was also
observed to be the case for CCGTs 3 and 4, when only three units offered multi-mode
operation. This indicates that a system seeking to increase its flexibility via multimode operation of CCGTs, possibly to facilitate integration of variable renewables,
may need to reward these units either through ancillary service payments or another
market mechanism to restore their revenue to original levels (i.e. when multi-mode
operation was not allowed). The subsidy or “top-up payment” required to restore
the revenue of these units to their original level is estimated here as the loss in total
production multiplied by the average electricity price. The average “top-up payment”
required is shown in Table 5.12 with the number of units requiring this payment shown
in parenthesis. However, it should be noted that this represents the worst-case figure
given that the multi-mode CCGT unit offered this capability in all time periods, rather
than when it was profitable for them to do so, as would likely be the case in reality.
5.5
Summary
Amending the scheduling model used by TSOs to include the bids and technical characteristics of a CCGT unit in open-cycle mode, in addition to the conventional CCGT
unit, is a simple task. The CCGT unit in open-cycle mode can be defined as a new unit,
with a constraint added to ensure that the CCGT unit and the CCGT unit in open-cycle
mode cannot be scheduled to run at the same time. This chapter examined if allowing
CCGT units to operate in open-cycle mode, when this is technically feasible and cost
optimal, could deliver benefits to a system with a high wind penetration or to the generators themselves. It was shown that the additional fast-starting capacity available from
multi-mode operation of CCGTs reduced the replacement reserve shortfall, indicating
an opportunity for increasing system reliability. Low-merit CCGTs were seen to utilize
the multi-mode function more than high-merit CCGTs, as they are frequently offline
and available for dispatch, whilst the increased competition among generators, typical
at higher levels of wind generation, resulted in multi-mode operation of CCGTs being
Chapter 5. Multi-mode Operation of Combined-Cycle Gas Turbines
78
utilized less frequently. Peaking production from CCGTs in open-cycle mode displaced
peaking production from OCGTs, potentially reducing the need for such units to be
built. Sensitivity studies revealed that usage of the multi-mode function is dependent
on the level of flexibility inherent in the system. Optimizing the system stochastically
or allowing intra-day trading on interconnectors reduced the need for flexibility to be
extracted from generators and consequently resulted in less frequent deployment of the
multi-mode function.
The analysis in this chapter assumed that the CCGT units capable of multi-mode
operation offered this flexibility in all time periods, whereas in reality generators would
strategise when to offer open-cycle operation such that plant production levels are not
negatively impacted, as was seen for some units under certain scenarios in this chapter.
Nonetheless, it was shown that the payment required to restore generator revenue to
levels when they did not offer multi-mode operation, in those cases where generator
production was reduced, was typically less than the system cost saving, indicating a net
benefit to society. A cost saving is also associated with the reduction in replacement
reserve shortfall which has not been considered here.
CHAPTER
6
Power System Flexibility and the Impact on Plant Cycling
6.1
Introduction
P
OWER system flexibility is defined in IEA (2008) as the ability to respond rapidly
to large fluctuations in supply or demand. A flexible power system, therefore, is
inherently capable of supporting a larger penetration of variable renewables. As wind
generation continues to grow, the operating flexibility of conventional plant may prove
insufficient to meet an increasingly variable net demand. In addition, increased cycling
of these plants can lead to extensive damage to the plant’s components, particularly
for base-load plant, as described in Chapter 2. Thus, considerable interest surrounds
the idea of incorporating sources of flexibility into power systems to support a higher
penetration of wind power.
Energy storage facilities, interconnection to neighbouring power systems and demand side management schemes (DSM) are well cited sources of flexibility within a
79
Chapter 6. Options for Increasing Power System Flexibility
80
power system (IEA, 2008; Van Hulle and Gardner, 2008). Each of these flexibility
options can assist in balancing variations in the net load. The flexibility of interconnection is present in the ability to import electricity from, or export electricity to, a
neighbouring power system, thus reducing the burden of managing net load variability
domestically. Energy storage will charge when the electricity price is low and generate
when prices are high, which will tend to flatten the net load curve (somewhat). Low
prices are associated with high wind penetration, and if storage units charge during
these periods it will raise the system demand, requiring increased production from
conventional plant and possibly keeping them online when they may otherwise have
been forced off-line. Demand side management schemes, depending on their nature,
can allow demand to be shed completely or shifted in time to better suit the net load
profile. In the context of a system with a large wind penetration the ability to shed
or reschedule demand is useful if forecast wind fails to materialise or wind generation
begins to reduce rapidly and production from conventional plant cannot be ramped
up quickly enough to compensate or alternatively when high wind generation coincides
with low demand, potentially forcing generators to be shut-down.
It was found in Brown et al. (2008) that pumped storage on isolated systems can
allow a greater penetration of renewables and improve the dynamic security of the
system, however Tuohy and O’Malley (2009) also shows that, although pumped storage
can reduce wind curtailment, the increased use of base-load units can actually lead to
increased emissions. Hamidi and Robinson (2008) found that responsive demand on
a system with a high wind penetration makes greater use of the wind resource and
reduces emissions, whilst Keane et al. (2011) finds DSM substitutes production from
peaking units and can provide a valuable source of reserve. It was also noted in Malik
(2001) that the avoided cycling cost of thermal units is a major benefit of DSM. The
net benefits of wind generation can be increased significantly by increasing the level of
interconnection on the power system, as shown in Denny and O’Malley (2007), whilst
Göransson (2008) also shows that investment in transmission to a region sufficiently far
away to make wind speeds uncorrelated (supergrid), or to a region with excess flexible
capacity, can decrease the total system costs of a system with a high wind penetration.
Chapter 6. Options for Increasing Power System Flexibility
81
In addition, the next generation of fossil-fired generation is set to be more flexible
as plant manufacturers, in response to the changing needs of power companies, are now
launching high efficiency power plants which are suited to cycling operation (GE, 2011;
Siemens, 2008a,b). As discussed in Chapter 4, the high minimum loads of CCGT units
resulting from emissions limitations often lead to them being forced off-line during high
penetrations of wind generation. However, plant manufacturers have now developed
solutions (such as bypassing compressed air around the combustor into the turbine to
increase the fuel-to-air ratio inside the combustor) to achieve higher firing temperatures
at lower loads, thus reducing the part-load emissions. This could facilitate new CCGT
units to remain online during periods of high wind generation (Siemens, 2008a). The
new Siemens H class CCGT, for example, can operate stably at 100 MW, less than
20% of its rated output (Probert, 2011). As shown in Chapter 5, it is also plausible
that existing CCGT units may in the future be operated in open-cycle mode as well
as combined-cycle mode (assuming simple market changes are made), releasing an
additional source of flexibility to the system.
This chapter examines how these various forms of flexibility will alter the operation
of base-load plant and investigates which is most beneficial to scheduling a system with
a large supply of variable wind power to reduce cycling of these inflexible plants. The
effect on wind curtailment and CO2 emissions are also examined. In addition, other
forms of flexibility are discussed, namely battery electric vehicles, maintenance scheduling (with consideration of system flexibility), the ability to control wind generation and
faster markets.
6.2
Methodology
The approach employed here was to incorporate equal capacities of the various sources
of flexibility in turn into the test system. By comparing each scenario against the
base case, the impacts of each flexibility option on system operation, and in particular
the operation of base-load plant, could be determined. The test system used is the
Irish 2020 test system with a 7.55 GW peak and 6000 MW installed wind capacity, as
Chapter 6. Options for Increasing Power System Flexibility
82
described in Chapter 3. As per Chapter 4, the results have been normalized to give the
result for a typically sized CCGT or coal unit.
Five scenarios were developed altogether, each incorporating 500 MW of a flexible
resource into the base case test system. These scenarios included 500 MW interconnection, pumped storage, DSM, additional turndown capability for CCGTs (i.e. reduced minimum operating levels) and open-cycle capacity from multi-mode operation
of CCGTs, as summarised in Table 6.1.
Table 6.1: Scenarios Examined
Scenario
Scenario
Scenario
Scenario
Scenario
1
2
3
4
5
500
500
500
500
500
MW
MW
MW
MW
MW
Interconnection
Pumped Storage
DSM
Turndown
Multi-mode
In scenario 1 which included 500 MW interconnection, the transfer of electrical
energy between the interconnected systems can be rescheduled in every planning period.
In scenario 2, the 500 MW pumped storage was split into 4 identical 125 MW storage
units, with characteristics as shown in Table 6.2. The storage units in scenario 2
all pumped to, and generated from, the same reservoir. (Thus if one unit has not
pumped any water it can still generate, provided the other units have pumped water
into the reservoir.) Pumping at maximum output required 8.5 hours to fill the reservoir.
Running at minimum output, the storage units, as they can run independently, could
generate for 408 hours.
Table 6.2: Characteristics of new pumped storage units
Max generation (MW)
Min generation (MW)
Max storage content (total) (MWh)
Min storage content (total) (MWh)
Max charging (MW)
Min charging (MW)
Max contribution to TR1 (MW)
Round trip efficiency (%)
125
10
5000
920
120
120
50
78
Chapter 6. Options for Increasing Power System Flexibility
83
In scenario 3 which contained 500 MW DSM, the DSM was modelled as two 250
MW units, one a peak shifting unit and the other a peak clipping unit (the 50:50
ratio between shifting and clipping capacity was also used in KEMA (2005)). The
peak shifting unit corresponded to load which could be shifted in time during the
day without reducing the overall energy demand, for example refrigeration load. As
such any reduction in demand must be replaced within the day (i.e. the total energy
exchange is equal to zero). It was modelled as a storage unit with 100% efficiency, as
described in Chapter 3. When the storage unit generates it corresponds to a demand
reduction by DSM, and when the storage unit charges it corresponds to the demand
being replaced. The DSM peak shifting unit could contribute up to 42 MW of spinning
reserve when it was actively reducing demand (i.e. when the representative storage unit
was generating) and had a variable operating cost of e40/MWh. The peak clipping unit
corresponded to peak load which could be reduced at times of high electricity prices
and does not increase demand at another time, for example lighting demand. The peak
clipping unit had a variable operating cost of e80/MWh and could also deliver up to
42 MW of spinning reserve when it was actively reducing demand. The values for the
variable operating costs and spinning reserve capabilities of the peak shifting and peak
clipping units were taken from Keane et al. (2011).
In scenario 4, the minimum operating level for five CCGT units on the test system
was reduced by 100 MW each (from an average minimum operating level of 220 MW).
For scenario 5, two CCGT units on the system (CCGT 4 & 5 from Chapter 5) were
assumed to be capable of multi-mode operation, thus releasing 500 MW additional
open-cycle capacity to the system when the units were not running in combined-cycle
mode. (The open-cycle capacity of these CCGTs is altered here compared with Chapter
5 in order to provide 500 MW flexible capacity in total.)
Chapter 6. Options for Increasing Power System Flexibility
6.3
6.3.1
84
Results
Impact on the Operation of Base-load Units
The cycling activity of the CCGT and coal units on the base case test system was
described in Chapter 4. CCGTs were seen to undergo a large number of annual startups relative to the coal units. Given their high minimum operating levels they are
forced off-line during periods of high wind penetration. The coal units on the other
hand avoid start-stop cycling as they provide the cheapest fossil-fired generation to the
system and also their low minimum operating levels allow them to stay online during
periods of high wind generation. However, the high part-load efficiency of the coal units
means they are the main providers of spinning reserve on the system and so operate
at part-load levels frequently. Coal units are also subject to severe ramping during
periods of very high wind generation as they are some of the few thermal units online
to provide power balancing. Severe ramping is defined here as a change in output
greater than half the difference between a unit’s maximum and minimum output over
one hour (excluding hours when the unit is starting up or shutting down).
Figure 6.1: Change in start-ups and production for a typical CCGT unit in each scenario
relative to the base case
Chapter 6. Options for Increasing Power System Flexibility
85
Scenario 1 - Interconnection
Figure 6.1 shows the change in the average number of annual startups and production
for a typical CCGT unit, for each of the scenarios investigated. Of all the flexibility
options examined the addition of 500 MW interconnection on the test system resulted
in the greatest reduction in start-stop cycling (17 less starts per year) for a typical
base-load CCGT unit. The reduction in cycling for CCGTs was also seen in Figure
6.1 to be correlated with increased production (an additional 80 GWh, an increase of
approximately 3.4%). With 6000 MW installed wind power capacity on the system,
prices on the Irish system frequently undercut those in Britain to the extent that the
Irish system is a net exporter of electricity, as discussed in Chapter 4. The increase
in exports allows for increased production from base-load plant and also with the opportunity to export during periods of high wind power penetration CCGT units can
avoid being shut-down. Although not shown here, production from lower merit CCGTs
however, which are effectively in mid-merit operation, is displaced by the increased interconnection capacity, as import levels also tend to increase at times when these units
are the marginal units on the system.
Figure 6.2 shows the change in the average number of annual startups and production for a typical coal unit, for each of the scenarios investigated. The coal units in
the base case were at their minimum number of annual start-ups so no reduction in
coal plant start-ups was possible. However, the coal units did benefit from a large
reduction in ramping operation, as seen in Figure 6.3, which shows the average number
of hours severe ramping was required from CCGT and coal units over the year. The
reduction of 118 hours (38%) of severe coal ramping relative to the base scenario was
the largest reduction in ramping of all scenarios examined. Likewise the reduction in
CCGT ramping of 47 hours (42%) relative to the base case was the largest observed
over all the scenarios investigated given that the increased export capacity will allow
more opportunities to balance net load variability through exchanges with the British
system. Thus, the additional flexibility that interconnection provides is particularly
beneficial to a system that is a net exporter, however, as seen in Chapter 4, it can
Chapter 6. Options for Increasing Power System Flexibility
86
exacerbate cycling on a system that is a net importer.
Figure 6.2: Change in start-ups and production for a typical coal unit in each scenario
relative to the base case
Scenario 2 - Pumped Storage
The addition of 500 MW of pumped storage increased the annual start-ups for a typical
CCGT unit by 9 relative to the base case, as seen in Figure 6.1. This increase in cycling
for a typical CCGT is correlated with slightly increased CCGT production (+1%) and
online hours (not shown), implying that although the CCGT units are being cycled more
they are gaining new opportunities for generation due to the introduction of the new
storage units. The increase in start-ups for a typical CCGT, with the additional pumped
storage on the system, was seen to arise as the addition of storage led to increased levels
of exports, requiring CCGT units to be started up to meet the additional demand.
Table 6.3: Operation of new storage units
Utilization factor for generation (%)
Utilization factor for spinning reserve (%)
Unit 1
40.9
38.5
Unit 2
44.5
40.4
Unit 3
43.4
42.9
Unit 4
43.1
45.3
However, the production for a typical coal unit on the system, shown in Figure 6.2,
was seen to decrease by 3.6%, while annual start-ups were seen to increase (+5), due to
the additional pumped storage capacity. Examining the operation of these new storage
Chapter 6. Options for Increasing Power System Flexibility
87
units revealed that they were used as much to provide spinning reserve to the system
as generation to the system. Table 6.3, which provides the utilization factor for each of
these new storage units on the system (total generation divided by maximum generation
possible during online hours) as well as the spinning reserve utilization factor (defined
here as total spinning reserve provided divided by maximum spinning reserve possible
during online hours), illustrates this trend. Consequently, the demand for spinning
reserve from coal units, which are the main thermal providers of primary reserve on the
system, is reduced and as such, as was seen in Chapter 4 also, these units can now be
cycled off-line on occasion as the requirement for them to be online providing spinning
reserve is reduced. Therefore, the amount of spinning reserve provided from coal units
drops 12% with the introduction of 500 MW pumped storage. Increased instances of
severe coal ramping were also observed in Figure 6.3.
Figure 6.3: Change in the number of hours severe ramping was required by a typical
CCGT or coal unit in each scenario relative to base case
Scenario 3 - Demand Side Management
The schedule for the DSM peak shifting and peak clipping units are set day-ahead
and cannot be revised intra-day. This limits the flexibility they can provide to the
system and can lead to sub-optimal decisions due to forecast uncertainty. As shown in
Chapter 6. Options for Increasing Power System Flexibility
88
Chapter 3, day-ahead wind generation is more often over-forecast than under-forecast,
thus reducing the net load predicted for the following day. This will thereby tend to
reduce the amount by which DSM will be utilised, particularly the expensive peak
clipping DSM unit. As such, the peak clipping unit has a capacity factor of 1.3% over
the year, while the peak shifting unit has a capacity factor of 5.7%. The clipping unit
is never dispatched at its maximum output, but provides its maximum contribution to
spinning reserve (42 MW) in every hour that it is utilised. Similarly the peak shifting
unit provides its maximum contribution to spinning reserve (42 MW also) in 89% of
the time that it is online. Thus the main functionality of the DSM units is in providing
reserve rather than reducing the demand. This was seen to have a detrimental effect
on the cycling of the base-load generation, despite its limited utilization.
The addition of 500 MW of DSM increased start-ups for a typical CCGT by 18,
relative to the base case, as seen in Figure 6.1, while starts for a typical coal unit
increased by 7, as seen in Figure 6.2. These were the largest increases observed across
all scenarios. Figure 6.3 also showed a large increase in the instances of severe ramping
for a typical coal unit (+294). As was seen previously with pumped storage, when DSM
units contribute to the spinning reserve target there is less requirement for thermal units
to be online providing spinning reserve. Thus, the system with DSM will tend to commit
less generation day-ahead. When forecast wind generation then fails to materialize the
following day, conventional generation needs to be started at short notice, giving rise
to the large increase in start-ups. (Production from peaking units also increased by
almost 400%). Ramping is also increased, particularly for coal units as seen in Figure
6.3.
To examine if DSM could bring about a reduction in base-load cycling, assuming
that it did not contribute to spinning reserve, sensitivities were run in which (i) the peak
clipping and peak shifting units did not provide spinning reserve, (ii) the peak clipping
and peak shifting units did not provide spinning reserve and their dispatches could be
rescheduled intra-day, and (iii) same as (ii), but with the variable operating cost for
the clipping unit reduced from e80/MWh to e60/MWh and the variable operating
cost for the shifting unit reduced from e40/MWh to e20/MWh. Table 6.4 shows the
Chapter 6. Options for Increasing Power System Flexibility
89
Figure 6.4: Change in start-ups for a typical CCGT and coal unit for each DSM scenario
capacity factor of the DSM peak clipping and shifting units for each of these scenarios.
Removing the ability to provide spinning reserve reduced the utilization of these units,
while allowing their dispatch to be changed intra-day increased utilization of the peak
clipping unit, but reduced utilization of the peak shifting unit. Reducing the variable
cost of DSM increased its utilization relative to the original scenario. These relatively
small reductions in the utilization of the DSM units had large impacts on cycling of
base-load plant as seen in Figure 6.4 and Figure 6.5 which show the change in starts
and production from the base case for each of the DSM scenarios examined.
Table 6.4: Capacity factor of DSM units for various scenarios
DSM
DSM, no reserve
DSM, no reserve, with rescheduling
DSM no reserve, with rescheduling, reduced cost
Peak clipping
1.3%
0.16%
0.35%
3.68%
Peak shifting
5.7%
4.9%
4.8%
7.5%
For the new sensitivities the level of plant cycling is comparable with the base case.
Only a minor reduction in CCGT start-ups was achieved (-2) and this was seen to
correspond to reduced production from those units. Given the need for DSM shifting
units to have zero impact on net energy over the day, its utilization is limited and as
such, as shown in these results, it does not hold benefits for cycling of base-load plant.
Chapter 6. Options for Increasing Power System Flexibility
90
Figure 6.5: Change in production for a typical CCGT and coal unit for each DSM
scenario
Scenario 4 - Turndown
In this scenario the CCGTs whose minimum operating level was reduced were seen to
utilise the increased turndown over an average of 330 hours throughout the year. By
reducing the minimum operating level of five CCGTs on the system, those CCGTs were
seen to benefit from reduced annual start-ups (annual start-ups for a typical CCGT
were down by 15) and subsequently increased levels of production (+95 GWh), as
seen in Figure 6.1. However, these units did experience increased instances of severe
ramping, as they are now kept online during periods of high wind generation when
previously they were shut-down. As such, Figure 6.3 shows an increase of 103 hours
when severe ramping was required from a typical CCGT unit. As might be expected
with CCGTs gaining increased production, production for a coal unit was consequently
reduced (-30 GWh), as seen in Figure 6.2. Increased production from the five CCGTs
also reduced production from peaking capacity (-27%), thus resulting in reduced CO2
emissions as seen in Table 6.6.
Scenario 5 - Multi-mode operation of CCGTs
The total production over the year from the multi-mode CCGTs in open-cycle mode
was 27.5 GWh, almost 4 times as much as the highest merit OCGT peaking unit in the
base case. Overall, the impact of including 500 MW of additional open-cycle capacity,
Chapter 6. Options for Increasing Power System Flexibility
91
via multi-mode operation of 2 CCGT units, had a small impact on the system dispatch.
The multi-mode CCGTs, when dispatched, were typically online around evening peak
hours and tended to impact production from low-merit CCGT units and peaking units.
As seen in Figure 6.1, the number of annual start-ups for a typical CCGT unit and the
number of instances that a typical CCGT or coal unit was required to perform severe
ramping, as seen in Figure 6.3, decreased indicating avoided cycling damage.
6.3.2
Impact on Wind Curtailment and CO2 Emissions
The available wind power on the test system in the test year was 18.4 TWh. Table 6.5
shows the amount of available wind that was curtailed in each of the scenarios. It is
clear that pumped energy storage, having the most flexible energy storage potential of
the options examined, was most effective at minimising wind curtailment events on the
system.
Table 6.5: Curtailment of wind in each scenario
Scenario
Base Case
Interconnection
Pumped storage
DSM
Turndown
Multi-mode
Wind Curtailed
(GWh)
% Change from
Base Case
148.4
61.5
56.0
117.3
108.9
153.6
-58.5
-62.3
-20.9
-26.6
3.49
The total Irish and British CO2 emissions for each scenario can be seen in Table
6.6. Each scenario is seen to reduce CO2 relative to the base case, however, the overall
changes are small. The largest CO2 reduction occurred in scenario 1 with increased
interconnector capacity. Emissions increased on the Irish system due to the increased
production to meet increased export levels, however the production that was displaced
on the British system was more CO2 intensive, thus yielding a net reduction.
Chapter 6. Options for Increasing Power System Flexibility
92
Table 6.6: CO2 emissions in each scenario
Scenario
Base Case
Interconnection
Pumped storage
DSM
Turndown
Multi-mode
6.4
CO2 emissions
(Mtonnes)
Change from
base case (Mtonnes)
199.4
198.9
199.1
199.1
199.2
199.4
-0.5
-0.3
-0.3
-0.2
0
Summary of Results
This chapter so far has investigated how commonly cited sources of power system
flexibility will interact with base-load generation on a power system with a high wind
energy penetration and alleviate or aggravate plant cycling. The results have been
somewhat counter-intuitive as several of the flexibility options examined, including
storage, were shown to contribute to plant cycling. The limited utilization of DSM,
had little impact on cycling (although a large increase in cycling if it is assumed to
provide reserve). Interconnection resulted in avoided cycling for both CCGT and coal
plant while, increased turndown for CCGTs was also seen to benefit CCGT operation.
The decision to invest in any of these options will be based on capital costs and
expected revenues. Benefits to the power system such as a reduction in production
costs, emissions or wind energy curtailment are often considered also. This chapter has
shown that plant cycling is another important factor to be weighed up, regardless of
whether the effects are positive or negative, particularly considering the high cycling
costs that have been found, as discussed in Chapter 2.
Chapter 6. Options for Increasing Power System Flexibility
6.5
6.5.1
93
Other Flexibility Options
Battery Electric Vehicles
Plug-in hybrid electric vehicles (PHEVs) and fully electric vehicles (EVs) provide an
opportunity to reduce emissions and decrease the dependence of the transport sector on
petroleum products. Consequently, many countries have announced national targets for
electric vehicles, for example, the Department of Energy (D.O.E) in the US is seeking
1,000,000 vehicles on the road by 2015, while in Ireland the target is for 10% of the
vehicle fleet (≈250,000 vehicles) to be electrified by 2020.
Plug-in hybrid electric vehicles (PHEVs) and fully electric vehicles (EVs) can also
deliver flexibility to power systems via the energy storage capacity present in the batteries. By employing a ‘smart charging’ strategy, whereby the system operator manages
the charging of electric vehicles, the net load profile can be flattened somewhat by
charging vehicles during the valleys, as depicted in Figure 6.6. This is particularly
beneficial on a windy night when base-load units may be forced off-line to accommodate high wind power penetration. Thus, EVs should facilitate more base-load and less
part-load operation from generators alleviating cycling issues and reducing emissions,
as was found to be the case in Göransson et al. (2010).
Figure 6.6: Illustration of load valley filling by EV charging
Chapter 6. Options for Increasing Power System Flexibility
94
However, as shown in Hadley and Tsvetkova (2009) and Göransson et al. (2010), if
a significant number of these vehicles are introduced without any control over the time
of charging, i.e. a typical owner charges the vehicle on arriving home from work and the
battery is charged until full, the evening peak demand will be exaggerated, requiring
more production from peaking units and thus resulting in higher CO2 emissions.
Assuming a smart charging scheme is in place, the system operator also has the
ability to stop vehicle charging temporarily if, for example, wind generation on the
system unexpectedly dropped off. Likewise, if wind generation unexpectedly picked
up the system operator (or a demand aggregator) can begin charging vehicles with
depleted or partially charged batteries. Thus, EVs can effectively deliver both positive
and negative spinning reserve (not actually ‘spinning’ but with an equivalent activation
time) to a power system (Kiviluoma and Meibom, 2011). However, it has been shown
that the marginal benefits of EVs will saturate at a point as there is a limit to the
amount of reserve that is required and the amount by which the net load profile can
be flattened (Kiviluoma and Meibom, 2009).
Vehicle-to-Grid (V2G) schemes have also been investigated, whereby it is possible
for electrical energy present in the battery to be delivered to the grid. In this case
EVs can deliver positive spinning reserve by not only reducing charging, but by actually providing electrical energy to the power system. However, repeatedly reversing
the flow of electricity between the battery and the grid will result in some level of
degradation to the battery which must be taken into consideration. When this, and
the cost of the bidirectional power electronics required, were taken into consideration
in Dallinger et al. (2011), it was found that it was not economical to provide positive
spinning reserve from EVs by discharging the battery.
6.5.2
Maintenance Scheduling
Another area where improvements in plant cycling could be gained (without the need
for costly additions to the power system) is maintenance scheduling. One of the duties
of a system operator is to agree an outage schedule with the power producing companies
Chapter 6. Options for Increasing Power System Flexibility
95
which allows each generating unit to fulfill its maintenance requirements without compromising the reliability of the power system. This process typically involves generators
submitting their outage requests for the year ahead to the system operator, who then
determines the impact of the aggregate outage requests on system reliability, based on
some reliability index (Feng and Wang, 2010; Shahidehpour and Marwali, 2000). The
loss of load expectation (LOLE), expected duration of unmet demand (EDUD), expected unsupplied energy or expected lack of available reserve are typical indices used
to determine the impact on system reliability (Mukerji et al., 1991). If the requested
maintenance schedules do not cause the system reliability to fall below some defined
standard (for example, the Irish system operator uses an LOLE of 8 hours per year),
they will be approved. Otherwise, if maintenance requests are causing periods of reliability concern, the generator(s) involved must revise their outage request(s) in order
to preserve system reliability. Typically generators seek to schedule their outages such
that their overall revenue is maximized, or in other words they request outages for
periods with the lowest electricity prices and hence the lowest electricity demand.
Traditionally, system operators have focussed on ensuring that there is sufficient
capacity to meet demand at all times during the year. However, a system with a large
wind penetration will also need to maintain a certain level of operational flexibility, in
addition to plant capacity, in order to maintain a reliable system. For example, if a
large quantity of fast-starting or fast-ramping plant is unavailable due to maintenance,
a system may still have sufficient capacity available to serve the load, however, should
a sudden drop in wind power output occur, there may not be sufficient fast response
generation available to compensate, or inflexible generators may be forced to operate
outside their normal operational limits. This type of operation, particularly when
required frequently of base-load generators, is associated with equipment deterioration,
increased maintenance costs and a reduction in reliability, as discussed in Chapter 2. By
ensuring that there is sufficient operating flexibility available within the generation fleet
to meet net load variations at all times during the year, excessive cycling of conventional
plant may be reduced/avoided.
One approach to evaluating the level of flexibility present in power systems was
Chapter 6. Options for Increasing Power System Flexibility
96
discussed in Lannoye et al. (2010), in a which a new metric, the insufficient ramping
resource expectation (IRRE), based on the loss of load expectation (LOLE) metric for
generation adequacy was presented. Utilizing such a metric in conjunction with multiple
net load projections (perhaps based on historical demand and wind power data from
several years) would go some way to ensuring that a power system maintained sufficient
flexibility throughout the year.
6.5.3
Control of Wind Power Output
By controlling the pitch angle of wind turbine blades it is possible to curtail wind power
output or limit its upward ramp rate. Curtailment of wind power is often viewed as a
negative outcome of a system having too little flexibility. However, there are occasions
when wind curtailment is the most economic solution to meeting demand. For example,
consider a system which has forecast a surge in wind power output, followed a short
time later by a drop-off in wind power output. If accommodating this ‘short-lived’ high
penetration of wind means switching off thermal plant, that will need to be restarted
shortly afterwards when the wind penetration begins to decline, the resulting startup fuel costs, cycling costs and carbon costs may instead make it more favourable
to curtail the wind power output for the short period. Ideally this is achieved by
the system operator sending a dispatch instruction to the wind generators. Presently
Bonneville Power Administration and Alberta Electric Service Operator are utilising
ramp controls on wind generation under certain reliability criteria. However, some
systems do not have the ability to control wind farm output (for example in Ireland a
large proportion of the wind generation is connected to the distribution system, which
cannot be controlled by the TSO), which can result in uneconomic system operation
and plant cycling.
6.5.4
Market Options
The power output from a wind farm is variable as the energy source itself, i.e. the
wind, is variable. However, the correlation in wind speeds between any two given
Chapter 6. Options for Increasing Power System Flexibility
97
sites decreases as the distance between those sites increases. Thus, when the power
output from various wind farms dispersed over a large area is aggregated, the overall
variability is less than the variability of the individual wind farms. This indicates
that a system which is interconnected to a neighbouring system can benefit from the
principle of statistical independence and thus reduce the burden on its thermal plant
to manage net load variability. The US is divided into 130 balancing areas, each
responsible for matching generation to demand in that area. Many studies have shown
that consolidating some of these balancing areas can benefit the integration of variable
renewables (NREL, 2011).
With a high wind penetration ‘faster’ markets are also advantageous. Currently
many markets are settled on an hourly basis, which can restrict access to flexible resources on the system. For example, in an hourly market a fast starting generator
cannot be started up within the hour to meet an increase in net load. Instead it would
have to wait until the beginning of the next hour to be dispatched, while online units
would have to ramp their output to meet the increased net load instead. Milligan et al.
(2010) finds the benefits of faster markets include greater access to flexibility and reduced a ramping requirement from conventional units.
CHAPTER
7
Unit Commitment with Dynamic Cycling Costs
7.1
T
Introduction
HE increased levels of cycling that base-load plant will be forced to undergo due
to increasing penetrations of wind generation have been shown in Chapter 4
and have also been seen in Göransson and Johnsson (2009). As discussed in Chapter
2, this can lead to high levels of damage accumulating within the plant’s components
ultimately resulting in increased maintenance requirements and forced outage rates.
Cycling related costs will arise via increased maintenance costs for generators, loss of
revenue resulting from longer and more frequent outages, increased fuel costs due to
reduced plant efficiency, as well as capital costs due to component replacement. Studies
indicate that the magnitude of these cycling related costs are high but, as discussed
in Chapter 2, accurately quantifying them is a challenging task given the range of
components affected, the unit specific nature of the analysis and the lengthy time lag
that is typically seen before cycling damage becomes apparent through component
98
Chapter 7. Unit Commitment with Dynamic Cycling Costs
99
failure (Lefton, 2004).
Not considering these costs however will result in the uneconomic dispatch of plants,
yet still markets currently do not include specific cycling cost components in their
bidding mechanisms, or at best cycling costs are bundled into a generator’s start-up
or ramping costs. Depending on the operating regime of a plant, these cycling related
costs can accumulate rapidly and are therefore dissimilar to plant characteristics such
as heat rate, which typically vary over a much longer time-scale. Therefore, to examine
the impact of these costs accurately, they should be modelled in a dynamic manner
such that they accumulate within the optimization process based on how the unit is
being operated and can thereby influence dispatch decisions.
This chapter presents a novel formulation that allows these cycling costs to be modelled dynamically, which can be integrated into a MIP (mixed integer programming)
unit commitment and economic dispatch model. This facilitates more accurate modelling of these costs and examination of how they accumulate in line with the operating
regime of a plant. The formulation sets up a cycling cost which increments with each
additional plant start-up or ramp, with the resulting cost function being linear, piecewise linear or step-shaped. This new approach to modeling cycling costs is particularly
suitable for long-term planning studies where it can be used to reflect the ageing effect
on a plant over time. It may also have applications for real-world dispatch models
where it can discourage the same unit from being repeatedly dispatched to cycle, as
this will incur an incremental cost to reflect the wear-and-tear to that unit and can
consequently alter its position in the merit order. A case study is included to determine how implementing dynamic cycling costs over a period of one year will affect the
resulting dispatch relative to a scenario where cycling costs are not considered.
7.2
Formulation of Dynamic Cycling Costs
A detailed formulation for implementing dynamic cycling costs which increase in line
with unit operation is presented here. Cycling costs are subdivided into costs for
Chapter 7. Unit Commitment with Dynamic Cycling Costs
100
(A) start-ups and (B) ramps. The formulation utilizes three main steps: (i) a binary
variable is set to indicate that damaging operation has occurred at time step t, (ii) a
counter tracks how much of that type of operation has occurred up to that point, and
(iii) an incrementing cycling cost is incurred at that time step. Linear, piece-wise linear
and step-shaped cost functions for both starts and ramps are detailed here.
7.2.1
Cycling Costs Related to Start-ups
Linear
Constraints 7.1 - 7.3 allow a dynamic, linearly incrementing cost for wear-and-tear
related to start-ups to be modelled. Based on the online binary variable, vg (t), constraint 7.1 sets the start-up, sg (t), and shut-down, zg (t), binary variables equal to 1
appropriately, when a unit ‘g’ is started or shut down at time t. Constraint 7.2 increments a counter, NgS (t), to track how many start-ups have been performed by that
unit. Constraint 7.3 determines the start-up related cycling cost, CgS (t), with the final
term ensuring that a cost is only incurred when the decision is made to start the unit
at time ‘t’ (i.e. sg (t) = 1). Figure 7.1 provides an example of this linearly increasing
cost function, where the incremental cost, costSg , is set equal to 100.
sg (t) − zg (t) = vg (t) − vg (t − 1), ∀ t ∈ T, ∀ g ∈ G
(7.1)
NgS (t) ≥ NgS (t − 1) + sg (t), ∀ t ∈ T, ∀ g ∈ G
(7.2)
¡
¢
CgS (t) ≥ NgS (t).costSg − M. 1 − sg (t) , ∀ t ∈ T, ∀ g ∈ G
(7.3)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
101
Figure 7.1: Linearly increasing start-up related cycling cost
Piecewise Linear
By defining i thresholds, T hSg (i), each corresponding to a cumulative number of plant
start-ups, at which point the start-up related cycling cost, CgS (t), will increase by
incremental cost costSg (i) for each additional start, a piecewise linear incremental cost
function can be modelled. Constraint 7.4 is a modified form of constraint 7.2 which
counts the cumulative number of start-ups. For i > 1, the start-up counter, NgS (t, i),
will not have a positive value until NgS (t, 1) has reached T hSg (i). T hSg (1) must equal 1.
Constraint 7.5 determines the total cycling cost. Figure 7.2 provides an example of a
piecewise linearly increasing cost function, where costSg (1) is set equal to 100, costSg (2)
is set equal to 150 and T hSg (2) equals 4.
NgS (t, i)
µ
¶
S
≥ Ng (t − 1, 1) + sg (t) + 1 − T hSg (i),
(7.4)
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
CgS (t)
¶
Ig µ
X
¡
¢
S
S
S
≥
Ng (t, i). costg (i) − costg (i − 1)
i
¡
¢
− 1 − sg (t) .M, ∀ t ∈ T, ∀ g ∈ G
(7.5)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
102
Figure 7.2: piecewise linearly increasing start-up related cycling cost
Step Function
Alternatively, if less information is known regarding the shape of the cost function
an appropriate simplification may be to define a step function, where CgS (t) does not
increment until T hSg (i) is reached. Again, it is required that T hSg (1) is equal to 1.
NgS (t, i) is determined by constraint 7.6 and in this case can be greater than or less
than 0 (it was previously defined as a positive variable only). Constraint 7.7 sets the
binary variable stepg (t, i) equal to 1 when NgS (t, i) has exceeded T hSg (i), and constraint
7.8 determines the cycling cost. Figure 7.3 provides an example of this incrementing,
step-shaped cost function, where costSg (t, 1) is set equal to 100, costSg (t, 2) is set equal
to 150 and T hSg (2) equals 4.
NgS (t, i)
µ
¶
S
= Ng (t − 1, 1) + sg (t) + 1 − T hSg (i),
(7.6)
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
NgS (t, i) − stepg (t, i).M ≤ 0,
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
(7.7)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
CS (t) ≥ costSg (i).stepg (t, i) −
¡
¢
1 − sg (t) .M,
103
(7.8)
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
Figure 7.3: Step increasing start-up related cycling cost
Hot and Cold Starts
Either the linear, piecewise linear or step formulations can be extended to differentiate
between hot and cold start-ups for units. Constraint 7.9 will set the binary variable
cold plus its
scold
g (t) equal to 1 only if a unit is started at time t, having been offline for t
minimum downtime, DTg . In constraints 7.2, 7.4 and 7.6 ‘+ sg (t)’ is replaced with ‘+
sg (t) + α.scold
g (t)’. A scaling factor, α, is chosen based on the ratio of cycling damage
caused by a hot start relative to a cold start, and thus normalizes NgS (t, i) to count in
terms of hot starts.
Tgcold +DTg
scold
g (t) ≥ vg (t) −
X
n=1
vg (t − n), ∀ t ∈ T, ∀ g ∈ G
(7.9)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
7.2.2
7.2.2.1
104
Cycling Costs Related to Ramping
Define one ramp level
The simplest form of incurring cycling costs related to ramping duty is to define a
change in output, Rg , between consecutive time periods, greater than which, damaging
transients will occur within the unit. Constraints 7.10 and 7.11 ensure that the binary
variable r(t) is set to 1 when a change in output exceeding Rg occurs. To avoid double
counting cycling costs when large ramps are experienced in the start-up or shut-down
process, the final term ensures that the constraints are non-binding when the unit is in
the start-up or shut-down process. If the ramp-related cycling costs are likely to exceed
the start-up or shut-down cost, constraint 7.12 is needed to prevent the model setting
s(t) and z(t) both equal to 1 in constraint 7.1, in order to make constraints 7.10 and
7.11 non-binding.
¡
¢
pg (t) − pg (t − 1) − M.rg (t) ≤ Rg + sg (t).M , ∀ t ∈ T, ∀ g ∈ G
(7.10)
¡
¢
pg (t − 1) − pg (t) − M.rg (t) ≤ Rg + zg (t).M , ∀ t ∈ T, ∀ g ∈ G
(7.11)
sg (t) + zg (t) ≤ 1, ∀ t ∈ T, ∀ g ∈ G
(7.12)
Utilizing the binary variable, rg (t), a counter is defined, as before, to incur an
incrementing, ramp-related cycling cost, CgR (t). Using the formulation from Section
7.2.1, the ramp-related cycling cost function may be linear, piecewise linear or stepshaped. Constraints 7.2 and 7.3 are replaced with the analogous ramp terms shown in
Table 7.1 to implement a linearly incrementing cost. Constraints 7.4 and 7.5, or 7.6
to 7.8, are replaced with the analogous ramp terms as shown in Table 7.1 to define a
piecewise linear, or a step shaped, incrementing ramp related cycling cost respectively.
Chapter 7. Unit Commitment with Dynamic Cycling Costs
105
Table 7.1: Analogous Terms
Linear
Piecewise
Linear &
Step
7.2.2.2
Starts
Ramps
Bi-directional Ramps
sg (t)
rg (t)
xg (t)
costSg
NSg (t)
CSg (t)
costR
g
R
Ng (t)
CR
g (t)
costX
g
NX
g (t)
sg (t)
rg (t)
xg (t)
costSg (i)
NSg (t,i)
ThSg (i)
CSg (t)
stepSg (t)
costR
g (i)
NR
g (t,i)
ThR
g (i)
R
Cg (t)
stepR
g (t)
costX
g (i)
CX
g (t)
NgX (t,i)
ThX
g (i)
CX
g (t)
stepX
g (t)
Define multiple ramp levels
The previous formulation, where one level Rg is set to define a ramp, can be expanded
to incur a dynamic ramp-related cycling cost, for j ramps of different magnitudes,
Rg (j). Constraint 7.13 ensures that for a ramp less than Rg (1), rg (t, j) will equal zero
for all j. A ramp greater than Rg (1), but less than Rg (2), will set rg (t, 1) equal to one,
and so forth. The final term ensures that the constraint is non-binding when the unit
¡
is starting up. A corresponding constraint is needed for down ramps, where pg (t)¢
¡
¢
pg (t − 1) in constraint 7.13 is replaced with pg (t − 1)-pg (t) and M.s(t) is replaced
with M.z(t). Constraint 7.14 ensures that the binary variable, rg (t, j), which indicates
that a ramp ≥ Rg (j) has occurred, can only have a value of 1 for one ramp level j, at
any given time. As before, constraint 7.12 is required to prevent sg (t) and zg (t) both
being set to 1 to make constraint 7.13 and its corresponding constraint non-binding.
Chapter 7. Unit Commitment with Dynamic Cycling Costs
106
j
X
¡
¢
¡
¢
pg (t) − pg (t − 1) < Rg (1). 1 −
rg (t, j) + Rg (2).rg (t, 1)
j=1
+... + Rg (j).rg (t, j − 1) + P¯g .rg (t, j) + M.sg (t),
(7.13)
where Rg (1) < Rg (2) < Rg (j)... < P¯g , ∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
j
X
rg (t, j) ≤ 1, ∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
(7.14)
j=1
As with hot and cold starts, scaling factors are used to normalize NgR (t) to count in
terms of one ramp level, as shown in constraint 7.15, where r(t, j) is expressed in terms
of r(t, 1). Constraint 7.16 determines the total ramp-related cycling cost, shown here
with a constant cost increment, costR
g , with the final term ensuring that a cost is only
incurred in a time period when a ramp (> Rg (1)) occurs.
NgR (t) = NgR (t − 1) + rg (t, 1) + β.rg (t, 2) + .... + γ.rg (t, j),
(7.15)
∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
CgR (t) ≥ NgR (t).costR
g −
j
X
¡
¢
1−
rg (t, j) .M
j=1
(7.16)
∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
To combine this formulation of j ramp levels with i cost thresholds (i.e piecewise
linear) constraints 7.15 and 7.16 are replaced by constraints 7.17 and 7.18, such that
R
R
once NgR (t, i) reaches T hR
g (i), Cg (t, i) will begin incrementing by costg (i).
Chapter 7. Unit Commitment with Dynamic Cycling Costs
¡
NgR (t, i) = NgR (t − 1, 1) + rg (t, 1) + β.rg (t, 2)
¢
+.... + γ.rg (t, j) + 1 − T hR
g (i)
107
(7.17)
∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g , ∀ i ≤ Ig
CgR (t)
¶
Ig µ
X
¡
¢
R
R
R
≥
Ng (t, i). costg (i) − costg (i − 1)
−
i
j
X
(7.18)
rg (t, j).M, ∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
j=1
To include a step-shaped ramp related cycling cost function, constraints 7.6-7.8 are
replaced with the analogous terms for ramping from Table 1.
7.2.2.3
Bi-directional ramps
Bi-directional ramping, typically experienced by a load-following unit, is thought to
be significantly more severe than ramps in one direction. A more detailed analysis
of cycling costs can include costs for bi-directional ramping as follows. Constraints
7.19 and 7.20 set the binary variables upg (t) and downg (t) to indicate the direction of
ramping. Only ramps of magnitude greater than Rg are considered as there will be
some level of ramping capability a generator can undertake relatively free of wear-andtear. Constraints 7.21 and 7.22 determine when a unit experiences large load changes
in opposite directions between two consecutive time periods.
pg (t) − pg (t − 1) ≤ Rg + M.upg (t) + M.sg (t), ∀ t ∈ T, ∀ g ∈ G
(7.19)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
pg (t − 1) − pg (t) ≤ Rg + M.downg (t) + M.zg (t), ∀ t ∈ T, ∀ g ∈ G
108
(7.20)
upg (t) + downg (t − 1) − M.xg (t) ≤ 1, ∀ t ∈ T, ∀ g ∈ G
(7.21)
upg (t − 1) + downg (t) − M.xg (t) ≤ 1, ∀ t ∈ T, ∀ g ∈ G
(7.22)
The binary variable xg (t) can now be used to increment a counter which in turn
can incur a dynamic bi-directional ramp-related cycling cost. To implement a linearly
incrementing cost function constraints 7.2 and 7.3 are replaced with the analogous ramp
terms shown in Table 1. Again, constraints 7.4 and 7.5, or constraints 7.6 to 7.8, are
replaced with the analogous ramp terms, as shown in Table 1, to implement a piecewise
linearly incrementing cost or a step-shaped incrementing cost respectively.
If dynamic cycling costs for ramping and bi-directional ramping are implemented
together it is necessary to avoid double counting ramping costs. This is achieved
by subtracting [r(t, j).costR (j) + r(t − 1, j).costR (j)] from the total cycling cost for
bi-directional ramping, CgX (t), when the reverse directional ramp is detected (when
xg (t)=1).
7.3
Model and Test System
To examine how cycling costs, modelled dynamically, will impact plant dispatch the
new formulation was implemented in a conventional MIP unit commitment model based
on Carrión and Arroyo (2006) and Arroyo and Conejo (2000). The unit commitment
problem was formulated as
Chapter 7. Unit Commitment with Dynamic Cycling Costs
M inimize
XX
cpg (t) + csg (t) + CgS (t) + CgR (t)
109
(7.23)
t∈T g∈G
subject to
X
pg (t) = D(t), ∀ t ∈ T
(7.24)
pg (t) ≤ P̄g .vg (t), ∀ t ∈ T
(7.25)
pg (t) ≥ P g .vg (t), ∀ t ∈ T
(7.26)
g∈G
As per Carrión and Arroyo (2006) and illustrated by Figure 7.4, a piecewise linear
approximation of a quadratic production cost function for each unit was adopted as
represented by:
N Lg
cpg (t) = Ag vg (t) +
X
Flg δl g(t), ∀ t ∈ T, ∀ g ∈ G
(7.27)
l=1
N Lg
pg (t) =
X
δl g(t) + P g vg (t), ∀ t ∈ T, ∀ g ∈ G
(7.28)
l=1
δ1 (g, t) ≤ T1g − P g , ∀ t ∈ T, ∀ g ∈ G
(7.29)
δl (g, t) ≤ Tlg − Tl−1g , ∀ t ∈ T, ∀ g ∈ G ∀ l = 2...N Lg − 1
(7.30)
δN Lg (g, t) ≤ P̄g − TN Lg −1 − Tl−1g , ∀ t ∈ T, ∀ g ∈ G
(7.31)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
δl (g, t) ≥ 0, ∀ t ∈ T, ∀ g ∈ G ∀ l = 1...N Lg
110
(7.32)
where Ag = ag + bg P g + cg P 2g .
Figure 7.4: Piecewise linear production cost (Carrión and Arroyo, 2006)
Start-up costs which were dependent on the period of time the unit had been offline
were modelled as follows:
¡
¢
csg (t) ≥ vg (t) − vg (t − 1) .hcg ∀ t ∈ T, ∀ g ∈ G
csg (t)
Tgcold +DTg
X
¡
≥ vg (t) −
¢
vg (t − n) .ccg , ∀ t ∈ T, ∀ g ∈ G
(7.33)
(7.34)
n=1
Minimum up time constraints were formulated by constraints 7.35, 7.36 and 7.37.
Equation 7.35 is only included if the number of hours a unit must remain online to
satisfy its minimum uptime, Bg , is greater than or equal to 1.
t≤Bg
X¡
¢
1 − vg (t) = 0, ∀ g ∈ G
t
(7.35)
Chapter 7. Unit Commitment with Dynamic Cycling Costs
111
t+U Tg −1
X
vg (n) ≥ U Tg .sg (t), ∀ g ∈ G, ∀ t = Bg + 1...T̄ − U Tg + 1
(7.36)
n=t
T̄
X
¡
¢
vg (n) − sg (t) ≥ 0, ∀ g ∈ G, ∀ t = T̄ − U T + 2...T̄
(7.37)
n=t
¡
¢
where Bg = max 0, vg (T)U Tg -hup
g +vg (T) .
Minimum down time constraints were formulated using constraints 7.38, 7.39 and
7.40. Equation 7.35 is only included if Lg ≥ 1.
t≤Lg
X¡
¢
vg (t) = 0, ∀ g ∈ G.
(7.38)
t
t+DTg −1
X
vg (n) ≥ DTg .zg (t), ∀ g ∈ G, ∀ t = Lg + 1...T̄ − DTg + 1
(7.39)
n=t
T̄
X
¡
¢
1 − vg (n) − zg (t) ≥ 0, ∀ g ∈ G, ∀ t = T̄ − DT + 2...T̄
(7.40)
n=t
¢
¡
+(1 − vg (T)) .
where Lg = max 0, (1 − vg (T)).DTg -hdown
g
The formulation was applied to the 10 unit test system used in Carrión and Arroyo
(2006); Kazarlis et al. (1996); Damousis et al. (2004), which was duplicated to give
a 20 unit system, thus facilitating a larger case study. The technical and economic
characteristics of these units are given in Table 7.2 and Table 7.3. (The initial state
Chapter 7. Unit Commitment with Dynamic Cycling Costs
112
is the number of hours a unit is assumed to have been online for at the start of the
optimization.) The fuel cost curves for the test units are given in Appendix D. The
peak demand (1500 MW) was doubled (3000 MW) and a historical year-long hourly
demand profile for the Irish system was scaled to produce a demand profile with a 3000
MW peak. The model was run for the test year, optimizing each day at an hourly
resolution.
Table 7.2: Generator Data
Units
P̄g
(MW)
Pg
(MW)
U Tg
(h)
DTg
(h)
Initial State
(h)
1-4
5-8
9-10
11-12
13-14
15-20
455
130
162
80
85
55
150
20
25
20
25
10
8
5
6
3
3
1
8
5
6
3
3
1
8
-5
-6
-3
-3
-1
Table 7.3: Generator production cost data
Units
ag
($/h)
bg
($/MWh)
cg
($/M W 2 h)
hcg
($/h)
ccg
($/h)
tcold
g
(h)
1-2
3-4
5-6
7-8
9-10
11-12
13-14
15-16
17-18
19-20
1000
970
700
680
450
370
480
660
665
670
16.19
17.26
16.60
16.50
19.70
22.26
27.74
25.92
27.27
27.79
0.00048
0.00031
0.00200
0.00211
0.00398
0.00712
0.00079
0.00413
0.00222
0.00173
4500
5000
550
560
900
170
260
30
30
30
9000
10000
1100
1120
1800
340
520
60
60
60
5
5
4
4
4
2
2
0
0
0
Generator cycling costs are difficult to determine and largely uncertain as discussed
in Section I. The figures used here, shown in Table 7.4, to implement dynamic cycling costs for the test system, are a conservative assumption based on those shown
in Lefton et al. (2006) and are intended to illustrate how dynamic cycling costs could
Chapter 7. Unit Commitment with Dynamic Cycling Costs
113
impact system operation, rather than provide an accurate estimate of such costs.
Piecewise linear costs for starts and ramps were implemented with the incremental
S
cost (costSg (i) or costR
g (i)) increasing by 10% and 20% when the start counter (Ng (t, 1)),
S
or ramp counter (NgR (t, 1)), exceeded 100 (T hSg (2) or T hR
g (2)) and 200 (T hg (3) or
T hR
g (3)) respectively. The scaling factor, α, was chosen to be 2, i.e. each cold start
incremented NgS (t, 1) by 2 (while a hot start incremented NgS (t, 1) by 1). Two ramp
levels, Rg (1) and Rg (2) corresponding to 20% and 40% of the difference between maximum and minimum output for a unit, were modelled. Scaling factors were chosen such
that ramps greater than Rg (1) or Rg (2) incremented NgR (t, 1) by 1 or 2 respectively.
Table 7.4: Incremental cycling costs $, (i=1)
Units
Base-load (Units 1-4)
Mid-merit (Units 5-10)
Peaking (Units 11-20)
7.4
costSg (i)
costR
g (i)
300
60
30
15
3
1.5
Results
This section examines how plant dispatches are affected when (i) a cycling cost related
to start-ups is implemented, (ii) a cycling cost related to ramping is implemented and
(iii) cycling costs related to start-ups and ramping are implemented simultaneously.
7.4.1
Start-up Related Cycling Costs Results
Implementing a dynamic cycling cost for plant start-ups, as shown in Table 7.4, was
seen to result in an overall reduction in plant start-ups. This is seen in Table 7.5,
which reveals reducing starts for base-load and mid-merit units. For base-load units,
the reduction in starts was correlated with increased production as, having the largest
incremental cycling costs, these units avoided shut-downs and gained more online hours.
This is seen via the average capacity factor shown in Table 7.6. Mid-merit units how-
Chapter 7. Unit Commitment with Dynamic Cycling Costs
114
ever, who also had reduced starts, saw reduced production indicating that they were
utilised less often. As these units were started up and shut down, and subsequently
incurred cycling costs, it became more economical after some point to dispatch peaking
units. Thus, starts and production increased for peaking units when a dynamic cycling
cost for start-ups was modelled. Figure 7.5 illustrates the cumulative start-ups for the
mid-merit and peaking units over the year when (i) cycling costs were modelled and (ii)
when cycling costs were not modelled. Starts are seen to accumulate rapidly between 0
and 2000 hours and from hours greater than 7000, as these are the winter months and
thus have higher demand, requiring more plant start-ups. Up to 1000 hours, the level of
cycling costs incurred by the mid-merit and peaking units is seen to have no impact on
the number of start-ups. However, beyond 1000 hours the cycling costs which are accumulated by mid-merit begins to have an effect on their position in the merit order and
consequently peaking plant are seen to be dispatched more frequently. Modelling dynamic cycling costs related to plant start-ups was also found to have the knock on effect
of increasing generator ramping. Over the year a 22% increase in ramping (NgR (t, 1))
was observed relative to the case when no cycling costs were modelled as generators
were more frequently ramped down to minimum output, rather than shut-down, in an
effort to avoid the increasing cycling costs.
Table 7.5: Impact of dynamic cycling costs for start-ups on total annual starts
No cycling
costs modeled
Cycling cost for
starts modeled
Base-load (Units 1-4)
Mid-merit (Units 5-10)
Peaking (Units 11-20)
34
1372
577
12
1005
838
Total
1983
1855
Units
Units within the same class, i.e. base-load, mid-merit or peaking, were also seen to
converge to a similar number of annual start-ups, as indicated by the reduced standard
deviation of annual start-ups seen in Table 7.7. This indicates that once a unit has
been cycled and its cycling cost is incremented, the next time a unit needs to be cycled
the costs will have now changed such that a different unit (most likely the next in the
Chapter 7. Unit Commitment with Dynamic Cycling Costs
115
Table 7.6: Impact of dynamic cycling costs for start-ups on average plant capacity
factors (%)
No cycling
Cycling cost for
costs modeled
starts modeled
Base-load (Units 1-4)
92.59
92.73
Mid-merit (Units 5-10)
27.82
25.42
Peaking (Units 11-20)
0.85
2.23
Units
Figure 7.5: Cumulative plant start-ups over the year, shown when dynamic cycling
costs for starts were (i) modelled and (ii) not modelled
merit order) may be scheduled. This leads to the burden of cycling operation being
more evenly distributed across the units. Over a long horizon, i.e. several years, this
effect can lead to a shift in the merit order, a trend which is somewhat emerging in
Figure 7.5.
To facilitate a sensitivity analysis, multiples of the initial incremental cycling costs,
costSg (1), that were shown in Table 7.4, were also examined. As the incremental cost
was increased the reduction in start stop cycling that is achieved by modelling dynamic
cycling costs quickly saturated as seen in Figure 7.6, thus indicating that the majority
of plant cycling is unavoidable. Table 7.8 shows a breakdown of the total number of
Chapter 7. Unit Commitment with Dynamic Cycling Costs
116
Table 7.7: Impact of dynamic cycling costs on plant start-ups by unit type
No cycling
Cycling cost for
cost modelled
starts modelled
Units
Avg
Std. Dev
Avg
Std. Dev
Base-load (Units 1-4)
8.5
9.9
3
3.6
Mid-merit (Units 5-10)
228.7
75.7
167.5
26.1
Peaking (Units 11-20)
57.7
73.1
83.8
27.5
starts by unit group, which again reveals that increasing starts for peaking units are
correlated with increasing incremental cycling cost, as it becomes more favourable to
dispatch these units due to the relatively larger cycling costs associated with the midmerit units. (The ripples in the curve shown in Figure 7.6 result from the increasing
starts for peaking units, as seen in Table 7.8.)
Figure 7.6: Impact of dynamic cycling cost on total start-ups, shown for various multiples of costSg (i)
A scenario where cycling costs were only modelled for a subset of the total fleet
was also examined. The 6 largest units on the system (units 1, 2, 3, 4, 9, 10) were
chosen based on the assumption that these units would be most impacted by cycling
operation and thus most likely to bid a wear-and-tear cost into the market to reflect
this. The results showed that although the number of annual start-ups was reduced for
these units, the start-ups for other units increased by an amount much greater than
Chapter 7. Unit Commitment with Dynamic Cycling Costs
117
Table 7.8: Impact of dynamic cycling costs for starts on total plant start-ups, shown
for various multiples of costSg (i)
Base-load
Mid-merit
Peaking
Units 1-4
Units 5-10
Units 11-20
No cycling cost
34
1372
577
costSg (i)*0.5
13
1104
781
costSg (i)*1
costSg (i)*2
costSg (i)*3
costSg (i)*10
12
1005
838
13
941
896
13
907
948
13
869
992
the reduction achieved for the units which bid a cycling cost, as seen in Table 7.9. This
would indicate the need for a uniform policy relating to the bidding of cycling costs to
be implemented in markets, such that all units reflect their cycling costs, or do not, to
avoid the situation where only some generators are bidding cycling costs which leads
to inefficient operation and excessive costs.
Table 7.9: Change in starts when a subset of units bid cycling costs for start-ups
∆ Starts
Units 1, 2, 3, 4, 9, 10
All other units
7.4.2
-86
+256
Ramping Related Cycling Costs Results
Implementing a dynamic cycling cost for plant ramping (shown in Table 7.4) resulted
in a 90% reduction in ramping overall as seen in Table 7.10. As described previously,
assuming a ramp greater than 20% or 40% of the difference between a unit’s maximum and minimum output increments the ramp counter, NgR (t), by a value of 1 or
2 respectively. The total value of NgR (t) at the end of the test year, summed for all
units, is shown in Table 7.10. Base-load units which carried out the greatest amount of
ramping when cycling costs were not modelled, saw the greatest reduction in ramping
Chapter 7. Unit Commitment with Dynamic Cycling Costs
118
operation when cycling costs for ramps were implemented. The drastic reduction in
ramping that was achieved by implementing dynamic ramping costs, however, led to
increased start-stop cycling as might be expected, although only by 3.3% over the year.
The most notable change to the overall dispatch that resulted from the introduction
of dynamic ramping costs was a slight reduction in production from base-load plant
allowing for increased production from mid-merit and peaking units as seen in Table
7.11, thereby spreading the ramping requirement over more units. Thus, including the
ramping cost was also seen to result in a slightly greater number of units online (5.94
per hour on average when dynamic ramping costs were modelled, versus 5.92 when no
cycling costs were modelled).
Table 7.10: Impact of dynamic cycling costs for ramping on total annual ramping
(NgR (t, 1))
No cycling
Cycling cost for
costs modeled
ramps modeled
Base-load (Units 1-4)
3717
120
Mid-merit (Units 5-10)
2214
1224
Peaking (Units 11-20)
795
623
Total ramping
6726
1967
Units
Table 7.11: Impact of dynamic cycling costs for ramping on average plant capacity
factors (%)
No cycling
Cycling cost for
costs modeled
ramps modeled
Base-load (Units 1-4)
92.59
92.21
Mid merit (Units 5-10)
27.82
28.61
Peaking (Units 11-20)
0.85
1.02
Units
7.4.3
Start-up and Ramping Cycling Costs Results
Implementing dynamic cycling costs (as shown in Table 7.4) for starts and ramping
simultaneously, reduced both types of cycling operation relative to the case when no
Chapter 7. Unit Commitment with Dynamic Cycling Costs
119
cycling costs were modelled, as shown in Table 7.12. Base-load units, having the largest
cycling costs, see the greatest reductions in cycling operation. Nonetheless, neither
total starts nor total ramps were reduced in this scenario as much as starts alone
or ramps alone were reduced when cycling costs for starts or ramps were modelled
individually. However, when cycling costs for start-ups only were modelled, ramping
operation increased and likewise when cycling costs for ramping only were modelled,
starts increased, thus when the cycling costs that would have been incurred, assuming
the costs given in Table 7.4 increment as described in Section 7.3, the case in which
cycling costs for start-ups and ramping were modelled simultaneously had the lowest
overall cycling costs, as shown in Figure 7.7. This would indicate that modelling cycling
costs for starts and ramping simultaneously most cost effectively reduces cycling and
as such one should not be considered without the other.
Table 7.12: Impact on total annual starts and ramps when dynamic cycling costs for
both start-ups and ramping were modelled
Units
No cycling costs
Cycling cost for starts
modeled
and ramps modeled
Starts
Ramps
Starts
Ramps
34
3717
12
144
Mid merit (Units 5-10)
1372
2214
1003
2069
Peaking (Units 11-20)
577
795
855
1456
Total
1983
6726
1870
3669
Base-load (Units 1-4)
Finally, when total system costs are examined for the scenario including cycling costs
and compared to the total system cost for the scenario in which cycling costs were not
modeled, but were calculated and added afterwards, it can be seen that modeling
cycling costs leads to lower system costs overall. This is shown in Figure 7.8. In this
example, the cost saving seen is considerable i.e. 14%.
Chapter 7. Unit Commitment with Dynamic Cycling Costs
120
Figure 7.7: Cycling costs (that would have been incurred) shown for various scenarios
Figure 7.8: Total system costs shown for various scenarios
7.5
Summary
Interest concerning cycling costs is growing and this paper sets out a formulation that
can utilize knowledge of incremental wear-and-tear costs related to plant start-ups or
ramping, to implement a dynamic incrementing cycling cost. The formulation covers
linear, piecewise linear and step-shaped cycling cost functions, the appropriate choice
for a user being determined by the level of knowledge of the generator’s cycling costs.
The formulation for piecewise linear incremental cycling costs related to plant start-
Chapter 7. Unit Commitment with Dynamic Cycling Costs
121
ups and ramps was implemented for a test system. Although the incremental costs
chosen are approximations, the results reveal certain trends that are likely for power
systems where generators undergo regular cycling and reflect the resulting wear-andtear costs in their bids. For example, dynamically modeling cycling costs for generator
starts was seen to reduce the number of starts, but caused ramping operation to be
increased (and vice-versa), whilst modeling cycling costs for only a subset of the generation fleet was seen to induce much higher levels of cycling in the remaining generation.
It was also seen that as cycling costs accumulated over time changes in the merit order
occurred, and that modeling cycling costs led to an overall saving for the system as
cycling operation was subsequently reduced.
CHAPTER
8
Conclusions
T
HIS thesis presented research related to the cycling of base-load generation with
increasing penetrations of wind energy on a power system. In Chapter 1 the
evolution of power systems to incorporate higher levels of wind generation against a
background of deregulation and increased competition is discussed. The likelihood
of increased generator cycling resulting has been found in many studies, such as GE
(2010); NREL (2010); NYISO (2010), and is beginning to become apparent in real world
systems (MMU, 2010). The physical consequences for increased cycling are explored in
Chapter 2 and thus provides the motivation for this research.
Chapter 4 outlined how the operation of CCGT and coal units will be impacted by
increasing levels of wind generation on a power system. Base-load CCGT units were
seen to undergo a large increase in start-stop cycling as wind penetration increased,
while coal units, being the most base-load generation, tended to remain online but
were subject to increased ramping and part-load operation. Thus, both CCGT and
coal units would be expected to experience increasing costs and forced outage rates
122
Chapter 8. Conclusions
123
over time due to wear and degradation of components from cycling operation.
Sensitivity analyses were conducted to examine the level of cycling occurring when
storage and interconnection were removed (individually) from the system. The results
showed reduced cycling for base-load plant in both cases. Without storage on the system, there is an increased requirement on base-load units to be online providing reserve
to the system, resulting in reduced start-stop cycling, while without interconnection the
entire system demand must be met domestically yielding increased production from and
reduced cycling of base-load units.
Having observed the decreasing production and online hours for CCGT units, Chapter 5 examined a new mode of operation for these units. Many CCGT units are fitted
with a bypass stack which allows the steam cycle to be bypassed and the gas turbine to
be run in open-cycle mode; a highly flexible, although less efficient, mode of operation.
The benefits of allowing CCGTs to operate in this manner, when technically possible
and economically optimal, included increased availability of replacement reserve. Production from peaking plant was also seen to be displaced when multi-mode operation of
CCGTs was introduced, indicating a reduced need for these units to be built and consequently a saving to society. The results also showed that low-merit CCGTs utilized
the multi-mode function more than high-merit CCGTs, as they are frequently offline
and available for dispatch, whilst the increased competition among generators, typical
at higher levels of wind generation, resulted in multi-mode operation of CCGTs being
utilized less frequently.
Chapter 6 examined how incorporating various sources of flexibility onto a power
system would impact cycling of base-load units and interestingly some were found to
have negative impacts on plant cycling. Pumped storage and DSM (assuming it provided reserve) increased coal cycling as the requirement on these units to remain online
for reserve provision was reduced. Interconnection and lower minimum operating levels
for CCGT units were found to reduce CCGT cycling and yield increased production
for these units.
Chapter 7 presented a novel formulation to allow cycling costs to be represented in
Chapter 8. Conclusions
124
a dynamic manner. Implementation of this formulation in a unit commitment model
allowed a case study to be conducted. The results showed that modelling dynamic
cycling costs will result in a reduction in cycling operation, however, if cycling costs are
modelled for a subset of generation only (the 6 largest units on the test system in this
case), the resulting level of cycling is significantly higher than the case when no cycling
costs were modelled. This indicates the importance of a uniform approach to bidding
cycling costs in electricity markets. It was also found that as cycling costs accumulated
over time, changes in the merit order became apparent. Specifically, as mid-merit units
were started up and shut down, and subsequently accumulated cycling costs, it was
found that after some point it became more economical to dispatch peaking units,
which had lower incremental cycling costs. This highlights the importance of investing
in flexible generation and retrofitting existing plant to be more capable of frequent
cycling.
8.1
Future Work
The analysis completed in Chapter 4, which examined the impact of increasing wind
penetrations on the operation of base-load plant, was conducted with an hourly time
resolution model, ie. the Wilmar Planning Tool. Each of the generators modelled on
the test system was capable of ramping from its minimum to maximum output (or vice
versa) in under one hour, so ramp rate constraints were non-binding. However, at a
time resolution under one hour modelling generator ramp rates would almost certainly
have an impact on the resulting dispatch, particularly as wind energy penetration
increases and the magnitude of net load ramps also increase. Consequently, a new
version of the Wilmar model, which operates with a 15 minute time step, has been
in development at the Electricity Research Centre in conjunction with this work. A
change to the structure of the model requires a change to the structure of the scenario
trees which are inputted into the model. Thus, an updated Scenario Tree Tool is also
being developed which will allow greater flexibility in making alterations to the model
structure, such as the time step, frequency of rolling planning, length of optimization
Chapter 8. Conclusions
125
Figure 8.1: CO2 emissions increase linearly with production
horizon or the number of branches in the scenario tree. Future work should utilise this
new version of the model to analyse if generator cycling is currently underestimated
using an hourly time step.
The analysis completed in Chapters 4 to 6 included estimates of the CO2 emissions
from generators based on the fuel consumption of the generators. Each fuel type was
assigned a carbon content (tonnes/GJ) and this was used to determine the CO2 emissions of the fuel consumed (GJ) by each generator in each hour. Thus, CO2 emissions
increased linearly (or piece-wise linearly if multiple heat rate slopes were modelled) as
production from a generator (and therefore fuel consumption) increased, as shown in
Figure 8.1 for a CCGT unit from the test system described in Chapter 3.
However, in reality generator fuel consumption, and thus emissions, are nonlinear
and Figure 8.1 represents a common modelling simplification for linear models. Motivated by the need to understand the link between emissions and generator cycling,
recent work conducted at NREL (as part of the Western Wind and Solar Phase 2 Study)
has utilised CEMs data to analyse the emissions from generators at various levels of
production. The results of this work determines the increase in emissions or ‘emissions
penalty’ that is incurred, relative to one hour of full-load operation, when a unit is
(i) operated at part-load (defined as 50% of max generation), (ii) ramped (defined as
Chapter 8. Conclusions
126
5% capacity change in one hour) and (iii) started-up (Brinkman, 2011). The findings
are detailed in Table 8.1. Future work could perform a similar analysis on emissions
penalties using data for generators on the Irish system. Reproducing Table 8.1 for the
Irish system would allow for more accurate analysis of the impact of generation cycling
on system emissions. Also as CO2 costs can represent almost a quarter of total system
costs, more detailed analysis of CO2 costs is warranted.
Table 8.1: CO2 emissions penalties for cycling operation (Brinkman, 2011)
Unit type
Coal
CCGT
OCGT
Part-load
penalty
5.1%
15.6%
12.4%
Ramping
penalty
0.4%
0.3%
0.3%
Start-up
penalty
110%
32%
32%
A technical approach has been taken in this thesis to examine the issue of baseload cycling with increasing wind penetration. However, many interesting policy and
market design issues have been indirectly raised, for example the fact that variable
wind generation may perversely support inflexible generation or that generators may
seek to bid cycling costs into the market and by doing so can avoid cycling operation.
If the goal for power systems is to achieve a high penetration of renewable generation,
in order to improve security of supply and reduce emissions without compromising
system reliability, the portfolio of conventional generation will need to become more
flexible, which may require incentives. The evolution of existing portfolios into more
flexible portfolios, in light of these concerns, is an interesting research area that warrants
investigation.
References
AIGS, 2008. All Island Renewable Grid Study, Workstream 2B, [Online] Available:
http://www.dcenr.gov.ie.
Al-Sunaidy, A., Green, R., 2006. Electricity deregulation in OECD (Organization for
Economic Cooperation and Development) countries. Energy 31 (6-7), 769–787.
Albright, D., Albright, D., Albright, J., 1999. Generator field winding shorted turn
detection technology. Generatortech, Inc.
Anderson, R., van Ballegooyen, H., 2003. Steam turbine bypass systems. Combined Cycle Journal,
Fourth Quarter,
[Online] Available:
http://www.psimedia.info/CCJ.htm.
APPrO, 2006. The Association of Power Producers of Ontario, Adopting a Ramp
Charge to Improve Performance of the Ontario Market, [Online] Available:
http://www.ieso.ca/imoweb/pubs/consult/mep/MP WG-20060707-ramp-cost.pdf.
Arroyo, J., Conejo, A., 2000. Optimal response of a thermal unit to an electricity spot
market. IEEE Transactions on Power Systems 15 (3), 1098–1104.
Arroyo, J., Conejo, A., 2004. Modeling of start-up and shut-down power trajectories of
thermal units. IEEE Transactions on Power Systems 19 (3), 1562–1568.
AWEA, 2011a. American Wind Energy Association,
U.S. wind industry
continues
growth,
Press
Release,
[Online]
Available:
http://www.awea.org/newsroom/pressreleases/release 07APR11 .cfm.
AWEA, 2011b. American Wind Energy Association, U.S. Wind Industry First Quarter
2011 Market Report, [Online] Available: http://www.awea.org.
Axford, M., 2009. Recession reduces demand for electricity. Power, [Online] Available:
http://www.powermag.com.
127
References
128
Balling, L., Hofmann, D., 2007. Fast cycling towards bigger profits. Modern power
systems, [Online] Available: http://www.modernpowersystems.com.
Bird, L., Bolinger, M., Gagliano, T., Wiser, R., Brown, M., Parsons, B., 2005. Policies
and market factors driving wind power development in the United States. Energy
Policy 33 (11), 1397–1407.
Bixby, R., Fenelon, M., Gu, Z., Rothberg, E., Wunderling, R., 2000. MIP: Theory
and Practice Closing the Gap. In: Proc. of 19th IFIP TC7 Conference on system
modelling and optimization. Vol. 174. pp. 19–49.
Black & Veatch, 2011. 2011 Strategic Directions Survey Results, [Online] Available:
http://www.bv.com/Downloads/Resources/Brochures/.
Blevins, B., 2007. Combined-cycle unit modeling in the nodal design, [Online] Available:
http://www.ercot.com.
Bloomberg, 2011. Bloomberg New Energy Finance, Wind turbine prices
fall to their lowest in recent years, Press Release, [Online] Available:
http://www.bnef.com/PressReleases/view/139.
Braun, M., 2004. Environmental external costs from power generation by renewable
energies. Thesis, Stuttgart University, Stuttgart, Germany.
Brinkman, G., 2011. NREL, Wind impact on emissions?, [Online] Available:
http://www.uwig.org/kcworkshop/Brinkman-KCWork.pdf.
Brown, P., Lopes, J., Matos, M., May 2008. Optimization of pumped storage capacity
in an isolated power system with large renewable penetration. IEEE Transactions on
Power Systems 23 (2), 523–531.
Brown, T. B., 1994. Assessing the effect of thermal transients on the life of boiler plant.
In: Proc. of International Conference on Life Management of Power Plants. pp.
137–143.
CAISO, 2010a. California ISO, Integration of renewable resources - operational requirements and generation fleet capability at 20 percent RPS, [Online] Available:
http://www.caiso.com/2804/2804d036401f0ex.html.
CAISO, 2010b. California ISO, Multi-stage generating (MSG) unit modeling, [Online]
Available: http://www.caiso.com/2078/2078908392d0.html.
Carrino, A., Jones, R., 2011. Coal plants challenged as gas plants surge. Power, [Online]
Available: http://www.powermag.com/.
Carrión, M., Arroyo, J., 2006. A computationally efficient mixed-integer linear formulation for the thermal unit commitment problem. IEEE Transactions on Power
Systems 21 (3), 1371–1378.
Castano, I., 2011. China installing wind-power capacity as fast as it can [Online] Available: http://www.renewableenergyworld.com.
References
129
CER, 2010. Commission for Energy Regulation,
cast Model and PLEXOS Validation Report
http://www.allislandproject.org.
Redpoint Validated Fore2010, [Online] Available:
Charles River Associates, 2010. Spp witf wind integration study. prepared for Southwest
Power Pool, [Online] Available: http://www.uwig.org.
Colpier, U., Cornland, D., 2002. The economics of the combined cycle gas turbine- an
experience curve analysis. Energy Policy 30 (4), 309–316.
Dallinger, D., Krampe, D., Wietschel, M., 2011. Vehicle-to-grid regulation reserves
based on a dynamic simulation of mobility behavior. IEEE Transactions on Smart
Grid (available online).
Damousis, I., Bakirtzis, A., Dokopoulos, P., 2004. A solution to the unit-commitment
problem using integer-coded genetic algorithm. IEEE Transactions on Power Systems
19 (2), 1165–1172.
Danneman, E., Beuning, S., 2011. Wind integration System and generation issues,
[Online] Available: http://www.energy-tech.com.
Dany, G., 2001. Power reserve in interconnected systems with high wind power production. In: Proc. of IEEE PowerTech Conference. Vol. 4.
Denny, E., O’Malley, M., 2007. Quantifying the total net benefits of grid integrated
wind. IEEE Transactions on Power Systems 22 (2), 605–615.
DOE,
2009.
United
States
Department
of
Energy,
States
with
Renewable
Portfolio
Standards,
[Online]
Available:
http://apps1.eere.energy.gov/states/maps/renewable portfolio states.cfm.
Doherty, R., O’Malley, M., 2005. A new approach to quantify reserve demand in systems
with significant installed wind capacity. IEEE Transactions on Power Systems 20 (2),
587–595.
Dupacova, J., Growe-Kuska, N., Romisch, W., 2003. Scenario reduction in stochastic
programming: An approach using probability metrics. Mathematical Programming
95 (3), 493–511.
EirGrid, 2009. Generation Adequacy Report 2010 - 2016, [Online] Available:
http://www.eirgrid.com.
EirGrid, SONI, 2010a. All Island TSO Facilitation of Renewbles Studies, [Online] Available: http://www.eirgrid.com/renewables/facilitationofrenewables/.
EirGrid, SONI, 2010b. Solver Choice in the SEM: A Comparative Study of
Lagrangian Relaxation vs. Mixed Integer Programming, [Online] Available:
http://www.sem-o.com.
Energy-Tech, 2004. Cycling of
http://www.energy-tech.com.
Combined-Cycle
Plants,
[Online]
Available:
References
130
EPRI, 2001a. Correlating cycle duty with cost at fossil fuel power plants. EPRI, Palo
Alto, CA. 1004010.
EPRI, 2001b. Damage to power plants due to cycling. EPRI, Palo Alto, CA. 1001507.
EPRI, 2002. Determining the cost of cycling and varied load operations: Methodology.
EPRI, Palo Alto, CA. 1004412.
Eskom, 2007. Eskom Holdings Ltd., Ankerlig power station conversion and transmission
integration project, [Online] Available: http://www.eskom.co.za/.
EU, 2008. European Union, Climate change: Commission welcomes final adoption of Europe’s climate and energy package, Press Release, [Online] Available:
http://europa.eu/rapid/pressReleasesAction.do?reference=IP/08/1998.
EWEA, 2011a. European Wind Energy Association, Large Scale Integration of Wind
Energy in the European Power Supply: Analysis, Issues and Recommendations, [Online] Available: Availablehttp://www.ewea.org/index.php?id=178.
EWEA, 2011b. European Wind Energy Association, Wind Energy and the Grid, [Online] Available: http://www.ewea.org/index.php?id=196.
EWEA, 2011c. European Wind Energy Association, Wind in power, 2010 European
statistics, [Online] Available: http://ewea.org/index.php?id=1665.
Feng, C., Wang, X., 2010. A competitive mechanism of unit maintenance scheduling in
a deregulated environment. IEEE Transactions on Power Systems 25 (1), 351–359.
Flynn, M., Walsh, M., O’Malley, M., 2000. Efficient use of generator resources in emerging electricity markets. IEEE Transactions on Power Systems 15 (1), 241–249.
French, D. N., 1993. Metallurgical failures in fossil fired boilers. John Wiley & Sons
Inc.
GE, 2005. GE Energy and Consulting, The effects of integrating wind power
on transmission system planning, reliability, and operations, prepared for New
York State Energy Research and Development Authority, [Online] Available:
http://www.nyserda.org/publications.
GE, 2010. GE Energy Applications & Systems Engineering and EnerNex Corporation and AWS Truepower, New England Wind Integration Study, prepared for ISO New England, [Online] Available:
http://www.uwig.org/CRA SPP WITF Wind Integration Study Final Report.pdf.
GE, 2011. General Electric, FlexEfficiency Power Plant Technology, [Online] Available:
http://www.ge-flexibility.com/index.jsp.
Göransson, L., 2008. Wind power in thermal power systems - large-scale integration.
Licentiate thesis, Dept. of Energy and Environment, Chalmers University of Technology, Goteburg, Sweeden.
References
131
Göransson, L., Johnsson, F., 2009. Dispatch modeling of a regional power generation
system-integrating wind power. Renewable Energy 34 (4), 1040–1049.
Göransson, L., Karlsson, S., Johnsson, F., 2010. Integration of plug-in hybrid electric
vehicles in a regional wind-thermal power system. Energy Policy 38 (10), 5482–5492.
Hadley, S., Tsvetkova, A., 2009. Potential impacts of plug-in hybrid electric vehicles on
regional power generation. The Electricity Journal 22 (10), 56–68.
Hamidi, V., Robinson, F., 2008. Responsive demand in networks with high penetration
of wind power. In: Proc. of IEEE/PES Transmission and Distribution Conference
and Exposition.
Hatch, 2008. Nova Scotia Wind Integration Study prepared for Nova Scotia Department
of Energy, [Online] Available: http://www.gov.ns.ca.
Holttinen, H., 2005. Impact of hourly wind power variations on the system operation
in the Nordic countries. Wind Energy 8 (2), 197–218.
Holttinen, H., Milligan, M., Kirby, B., Acker, T., Neimane, V., Molinski, T., 2008.
Using standard deviation as a measure of increased operational reserve requirement
for wind power. Wind Engineering 32 (4), 355–377.
Holttinen, H., Pedersen, J., 2003. The effect of large-scale wind power on a thermal
system operation. In: Proc. of 4th International Workshop on Large-scale Integration
of Wind Power for Offshore Wind Farms.
IEA, 2008. International Energy Agency, Empowering variable renewables, options for
flexible electricity systems, [Online] Available: http://www.iea.org.
IEA,
2010.
International
Energy
Agency,
IEA
Wind
Energy,
Annual
Report
2009,
[Online]
Available:
http://www.ieawind.org/AnnualReports PDF/2009/2009AR 92210.pdf.
Kazarlis, S., Bakirtzis, A., Petridis, V., 1996. A genetic algorithm solution to the unit
commitment problem. IEEE Transactions on Power Systems 11 (1), 83–92.
Keane, A., Tuohy, A., Meibom, P., Denny, E., Flynn, D., Mullane, A., O’Malley, M.,
2011. Demand side resource operation on the Irish power system with high wind
power penetration. Energy Policy 39 (5), 2925 – 2934.
Kehlhofer, R., Rukes, B., Hannemann, F., Stirnimann, F., 2009. Combined-cycle gas
& steam turbine power plants. Pennwell Books.
KEMA, 2005. A Scoping Study: Demand Side Measures for Small Business and Residential Customers on Ireland’s Electrical System prepared for Sustainable Energy
Ireland, [Online] Available: http://www.seai.ie.
King, J., 1996. Recent experience in condition assessments of boiler header components
and supports. In: Proc. of ASME Pressure Vessels and Piping Conference.
References
132
Kirby, B., Milligan, M., 2008. Facilitating wind development: The importance of electric
industry structure. The Electricity Journal 21 (3), 40–54.
Kitto Jr, J., Bryk, S., Piepho, J., 1996. Upgrades and enhancements for competitive
coal-fired boiler systems. In: Proc. of the 1996 International Joint Power Generation
Conference.
Kiviluoma, J., Meibom, P., 2009. Coping with wind power variability: How plug-in
electric vehicles could help. In: Proc. of the 8th International Workshop on LargeScale Integration of Wind Power into Power Systems as well as on Transmission
Networks for Offshore Wind Farms. pp. 336–340.
Kiviluoma, J., Meibom, P., 2011. Methodology for modelling plug-in electric vehicles in
the power system and cost estimates for a system with either smart or dumb electric
vehicles. Energy 36 (3), 1758–1767.
Lannoye, E., Milligan, M., Adams, J., Tuohy, A., Chandler, H., Flynn, D., O’Malley,
M., 2010. Integration of variable generation: Capacity value and evaluation of flexibility. In: Proc. of IEEE Power and Energy Society General Meeting.
Lefton, S., 2004. Profitable operation requires knowing how much it
costs to cycle your unit. Combined Cycle Journal, [Online] Available:
http://www.combinedcyclejournal.com/.
Lefton,
S.,
2011.
Power
plant
asset
management
Presented
at
UWIG
Spring
Workshop,
Kansas
City,
[Online]
Available:
http://www.uwig.org/kcworkshop/Lefton-KCWork.pdf.
Lefton, S., Besuner, P., 2006. The cost of cycling coal fired power plants. Coal Power
Magazine, pp. 16–20.
Lefton, S., Besuner, P., Agan, D., 2006. The real cost of on/off cycling. Modern power
systems, [Online] Available: http://www.modernpowersystems.com.
Lefton, S., Besuner, P., Grimsrud, G., 1995. Managing utility power plant assets to
economically optimize power plant cycling costs, life, and reliability. In: Proc. of 4th
IEEE Conference on Control Applications. pp. 195–208.
Lefton, S., Besuner, P., Grimsrud, G., 1997. Cycling fossil fired units proves costly
business. Electric Light and Power 75 (7).
Lefton, S., Besuner, P., Grimsrud, P., Bissel, A., Norman, G., 1998. Optimizing power
plant cycling operations while reducing generating plant damage and costs at the Irish
Electricity Supply Board, Aptech Engineering Service Technical Report (TP123),
[Online] Available: http://forgoodpower.com/technical papers.html.
Lu, B., Shahidehpour, M., 2004. Short-term scheduling of combined cycle units. IEEE
Transactions on Power Systems 19 (3), 1616–1625.
Malik, A., 2001. Modelling and economic analysis of DSM programs in generation
planning. International Journal of Electrical Power & Energy Systems 23 (5), 413–
419.
References
133
Meibom, P., 2006. WILMAR - Wind Power Integration in Liberalised Electricity Markets, [Online] Available: http://www.wilmar.risoe.dk/Results.htm.
Meibom, P., Barth, R., Brand, H., O’Malley, M., 2011. Stochastic optimization model
to study the operational impacts of high wind penetrations in ireland. IEEE Transactions on Power Systems (available online).
Meibom, P., Weber, C., Barth, R., Brand, H., 2009. Operational costs induced by fluctuating wind power production in germany and scandinavia. IET Renewable Power
Generation 3 (1), 75–83.
Milligan, M., Kirby, B., Beuning, S., 2010. Combining Balancing Areas’ Variability:
Impacts on Wind Integration in the Western Interconnection. National Renewable
Energy Laboratory.
MMU, 2010. Market Monitoring Unit, Power Plant Cyling in SEM, [Online] Available:
http://www.allislandproject.org/en/mmu decision documents.aspx.
Möhrlen, C., Jørgensen, J., Pinson, P., Madsen, H., Runge Kristoffersen, J., 2007. High
Resolution Ensemble for Horns Rev: A project overview. In: Proc. European offshore
wind energy conference.
Monteiro, C., Bessa, R., Miranda, V., Botterud, A., Wang, J., Conzelmann, G., 2009.
Wind power forecasting: State-of-the-art 2009 Argonne National Laboratory (ANL),
[Online] Available: http://www.dis.anl.gov/pubs/65613.pdf.
Moore, W., 2006. Include generators and exciters in your outage inspections. Power,
[Online] Available: http://www.powermag.com.
Mukerji, R., Merrill, H., Erickson, B., Parker, J., Friedman, R., 1991. Power plant
maintenance scheduling: optimizing economics and reliability. IEEE Transactions
on Power Systems 6 (2), 476–483.
Narula, R., Massy, M., Singh, J., 2002. Design Consideration for Combined Cycle
Plants for the Deregulated Market - An EPC Contractors Perspective. Proceedings
of International Gas Turbine Institute ASME TURBO EXPO.
Norgaard, P., Holttinen, H., 2004. A multi-turbine power curve approach. In: Nordic
Wind Power Conference.
NREL,
2010.
National
Renewable
Energy
Laboratory,
Western
wind
and
solar
integration
study,
[Online]
Available:
http://www.nrel.gov/wind/systemsintegration/wwsis.html.
NREL, 2011. National Renewable Energy Laboratory, The Role of Large
Balancing Areas in Integrating Solar Generation, [Online] Available:
http://www.nrel.gov/docs/fy11osti/50059.pdf.
NYISO, 2010. Growing Wind - Final Report of the NYISO Wind Integration Study,
[Online] Available: http://www.nyiso.com.
References
134
OSIsoft, 2007. Cutting the Cost of Flexible Operation in a Competitive Power Market,
[Online] Available: http://www.osisoft.com/.
Oswald, J., Raine, M., Ashraf-Ball, H., 2008. Will British weather provide reliable
electricity? Energy Policy 36 (8), 3212–3225.
Padhy, N., 2004. Unit commitment - a bibliographical survey. IEEE Transactions on
Power Systems 19 (2), 1196–1204.
Probert, T., 2011. Fast starts and flexibility:
Let the gas turbine battle
commence.
Power Engineering International,
[Online] Available:
http: // www. powerengineeringint. com/ 19 (6).
Risø National Laboratory, 1989. European wind resources at 50 metres a.g.l., [Online]
Available: http://www.windatlas.dk/Europe/landmap.html.
Salam, S., 2007. Unit commitment solution methods. In: Proc. of World Academy of
Science, Engineering and Technology. Vol. 26. pp. 320–325.
SEAI, 2010. Sustainable Energy Authority Ireland, History of Wind Energy, [Online]
Available: http://www.seai.ie/Renewables/Wind Energy/History of Wind Energy/.
Sen, S., Kothari, D., 1998. Optimal thermal generating unit commitment: a review.
International Journal of Electrical Power & Energy Systems 20 (7), 443–451.
Shahidehpour, M., Marwali, M., 2000. Maintenance scheduling in restructured power
systems. Springer.
Shibli, A., Starr, F., 2007. Some aspects of plant and research experience in the use of
new high strength martensitic steel p91. International Journal of Pressure Vessels
and Piping 84 (1-2), 114–122.
Siebert, N., 2008. Development of methods for regional wind power forecasting. Ph.D.
thesis, Mines ParisTech, Paris, France.
Siemens, 2008a. Low Load Operational Flexibility for Siemens G-class Gas Turbines.
In: Proc. of Power-Gen International.
Siemens, 2008b. Operational flexibility enhancements of combined-cycle power plants.
In: Proc. of Power-Gen Asia.
Söder, L., 2004. Simulation of wind speed forecast errors for operation planning of multiarea power systems. In: IEEE International Conference on Probabilistic Methods
Applied to Power Systems. pp. 723–728.
Starr, F., 2003. Background to the design of HRSG systems and implications for CCGT
plant cycling. Operation Maintenance and Materials Issues 2 (1).
Strbac, G., Shakoor, A., Black, M., Pudjianto, D., Bopp, T., 2007. Impact of wind
generation on the operation and development of the UK electricity systems. Electric
Power Systems Research 77 (9), 1214–1227.
References
135
Streiffert, D., Philbrick, R., Ott, A., 2005. A mixed integer programming solution for
market clearing and reliability analysis. In: Proc. of IEEE Power Engineering Society
General Meeting. pp. 2724–2731.
Troy, N., Denny, E., O’Malley, M., 2010. Base-load cycling on a system with significant
wind penetration. IEEE Transactions on Power Systems 25 (2), 1088–1097.
Tuohy, A., Meibom, P., Denny, E., O’Malley, M., 2009. Unit commitment for systems
with significant wind penetration. IEEE Transactions on Power Systems 24 (2),
592–601.
Tuohy, A., O’Malley, M., 2009. Impact of pumped storage on power systems with
increasing wind penetration. In: Proceedings of 2009 IEEE PES General Meeting.
Calgary, Alberta, Canada.
Ummels, B., Gibescu, M., Pelgrum, E., Kling, W., Brand, A., 2007. Impacts of wind
power on thermal generation unit commitment and dispatch. IEEE Transactions on
Energy Conversion 22 (1), 44–51.
Van Hulle, F., Gardner, P., 2008. Wind Energy - The Facts, Part 2 Grid Integration,
[Online]. Available: http://www.wind-energy-the-facts.org/.
Wambeke, S., 2006. Risks to HRSGs in low-load operation. Combined Cycle Journal
Second Quarter, [Online] Available: http://www.combinedcyclejournal.com/.
Watson, W., 1996. The success of the combined cycle gas turbine. In: Proc. of the
IEEE Conference on Opportunities and Advances in International Electric Power
Generation. pp. 87–92.
Xcel
Energy,
2010.
Integrating
Wind
Cost
of
Cycling
Analysis
for
Harrington
Station
Unit
3
[Online]
Available:
http://www.blankslatecommunications.com/Images/Aptech-HarringtonStation.pdf.
Appendix A. Probability distribution of net load ramps
Figure 8.2: Probability distribution of hourly net load ramps on the 7.55 GW peak
system
136
Appendix A. Probability distribution of net load ramps
137
Figure 8.3: Probability distribution of hourly net load ramps on the 9.6 GW peak
system
Appendix B. Cycling data for CCGT and coal units
Figure 8.4: Start-ups and capacity factor for a typical low-merit CCGT unit on the
7.55 and 9.6 GW peak demand systems, with increasing wind penetration
138
Appendix B. Cycling data for CCGT and coal units
139
Table 8.2: Start-up data for CCGT and coal units on the 7.55 GW peak demand system
Statistic
CCGT
Coal
Wind energy penetration
15%
29%
43%
15%
29%
43%
Max. value
161
186
197
65
54
43
Min. value
18
41
70
8
9
6
Average value
72.4
90.6
115.4
28.4
26.4
20.6
Std. Deviation
63.8
63.6
52.3
26.4
22.6
15.4
Table 8.3: Capacity factor data for CCGT and coal units on the 7.55 GW peak demand
system
Statistic
CCGT
Coal
Wind energy penetration
15%
29%
43%
15%
29%
43%
Max. value
0.81
0.75
0.64
0.77
0.71
0.69
Min. value
0.73
0.59
0.44
0.72
0.67
0.61
Average value
0.79
0.71
0.59
0.75
0.70
0.66
Std. Deviation
0.034
0.066
0.086
0.024
0.175
0.029
Table 8.4: Start-up data for CCGT and coal units on the 9.6 GW peak demand system
Statistic
CCGT
Coal
Wind energy penetration
11%
23%
34%
11%
23%
34%
Max. value
116
170
198
56
81
67
Min. value
7
23
44
8
9
5
Average value
32.4
63.6
93
27.2
36.4
32.4
Std. Deviation
47.1
63.4
65.3
25.4
36.3
30.9
Appendix B. Cycling data for CCGT and coal units
140
Table 8.5: Capacity factor data for CCGT and coal units on the 9.6 GW peak demand
system
Statistic
CCGT
Coal
Wind energy penetration
11%
23%
34%
11%
23%
34%
Max. value
0.90
0.85
0.77
0.83
0.79
0.76
Min. value
0.85
0.76
0.65
0.77
0.75
0.72
Average value
0.87
0.82
0.74
0.80
0.76
0.74
Std. Deviation
0.02
0.03
0.05
0.02
0.02
0.02
Appendix C. Base-load cycling with/without storage/interconnection
Figure 8.5: Number of hours online for an average CCGT and coal unit with/without
storage and an increasing wind penetration on the 9.6 GW peak demand system
141
Appendix C. Base-load cycling with/without interconnection
142
Figure 8.6: Number of start-ups for an average CCGT and coal unit with/without
storage and an increasing wind penetration on the 9.6 GW peak demand system
Figure 8.7: Number of hours online for an average CCGT and coal unit with/without
interconnection and an increasing wind penetration on the 9.6 GW peak demand system
Appendix D. Fuel Cost Curves
Figure 8.8: Fuel cost curves for test units
143
Appendix D. Fuel Cost Curves
Figure 8.9: Fuel cost curves for test units
144
Appendix E. Publications
1. Troy, N., Denny, E. and O’Malley, M. “Base-load cycling on a system with significant wind penetration”, IEEE Transactions on Power Systems, vol. 25, issue
2, pp. 1088 - 1097, 2010.
2. Troy, N., Flynn, D. and O’Malley, M. “Multi-mode Operation of Combined-Cycle
Gas Turbines with Increasing Wind Penetration”, IEEE Transactions on Power
Systems, In Press
3. Troy, N., Flynn, D., Milligan M. and O’Malley, M. “Unit Commitment with
Dynamic Cycling Costs”, IEEE Transactions on Power Systems, in review.
145
1088
IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 25, NO. 2, MAY 2010
Base-Load Cycling on a System
With Significant Wind Penetration
Niamh Troy, Graduate Student Member, IEEE, Eleanor Denny, Member, IEEE, and Mark O’Malley, Fellow, IEEE
Abstract—Certain developments in the electricity sector may result in suboptimal operation of base-load generating units in countries worldwide. Despite the fact they were not designed to operate in a flexible manner, increasing penetration of variable power
sources coupled with the deregulation of the electricity sector could
lead to these base-load units being shut down or operated at partload levels more often. This cycling operation would have onerous
effects on the components of these units and potentially lead to
increased outages and significant costs. This paper shows the serious impact increasing levels of wind power will have on the operation of base-load units. Those base-load units which are not
large contributors of primary reserve to the system and have relatively shorter start-up times were found to be the most impacted as
wind penetration increases. A sensitivity analysis shows the presence of storage or interconnection on a power system actually exacerbates base-load cycling until very high levels of wind power
are reached. Finally, it is shown that if the total cycling costs of
the individual base-load units are taken into consideration in the
scheduling model, subsequent cycling operation can be reduced.
Index Terms—Costs, interconnected power systems, power
system modeling, pumped storage power generation, thermal
power generation, wind power generation.
I. INTRODUCTION
S higher penetrations of wind power are achieved, system
operation becomes increasingly complex, as variations
in the net load (load minus wind) curve increase [1]. Wind is
a variable energy source and fluctuations in output must be
offset to maintain the supply/demand balance, thus resulting in
a greater demand for operational flexibility from the thermal
units on the system [2]. These units must also carry additional
reserves to maintain system reliability should an unexpected
drop in wind occur, as the power output from wind farms is
also relatively difficult to predict [3]. However, even when
state-of-the-art methods of forecasting are employed, the next
day hourly predicted wind output can vary by 10%–15% of
A
Manuscript received May 25, 2009; revised September 24, 2009. First published January 08, 2010; current version published April 21, 2010. This work
was conducted in the Electricity Research Centre, University College Dublin,
Ireland, which is supported by Airtricity, Bord Gais, Bord na Mona, Cylon Controls, the Commission for Energy Regulation, Eirgrid, Electricity Supply Board
(ESB) International, ESB Networks, ESB Power Generation, Siemens, SWS
Group, and Viridian. This work was supported by a Charles Parsons Energy Research Award from the Department of Communications, Energy and Natural Resources administered by Science Foundation Ireland. Paper no. TPWRS-003772009.
N. Troy and M. O’Malley are with the School of Electrical, Electronic, and
Mechanical Engineering, University College Dublin, Dublin, Ireland (e-mail:
[email protected]; [email protected]).
E. Denny is with the Department of Economics, Trinity College Dublin,
Dublin, Ireland (e-mail: [email protected]).
Digital Object Identifier 10.1109/TPWRS.2009.2037326
the total wind capacity as reported in [4], which can result
in thermal units being over- and under-committed [2]. Furthermore, in certain systems wind is allowed to self-dispatch,
so forecast output is not included in the day-ahead schedule.
This can lead to increased transmission constraints which
will further intensify plant cycling and has been shown to
displace energy from combined cycle gas turbines (CCGTs) in
particular [5]. The culmination of adding more variability and
unpredictability to a power system is that thermal units will
undergo increased start-ups, ramping and periods of operation
at low load levels collectively termed “cycling”[6]–[9].
In addition to wind, the competitive markets in which these
units operate are also a significant driver of plant cycling;
increased levels of competition brought about by widespread
deregulation results in all types of generators being forced
into more market-orientated, flexible operation to increase
profits [10]. The severity of plant cycling, will be dependent
on the generation mix and the physical characteristics of the
power system. It is widely reported that the availability of
interconnection and storage can assist the integration of wind
on a power system [11], [12]. Interconnection can allow imbalances from predicted wind power output to be compensated
via imports/exports whereas some form of energy storage can
enable excess wind to be moderated in time to correlate with
demand. This should relieve cycling duty on thermal units as
the onus on them to balance fluctuations is relieved.
Although all conventional units will be impacted to some degree by wind integration, it is cycling of base-load units that is
particularly concerning for system operators and plant owners
alike. As these units are designed with minimal operational
flexibility, cycling these units will result in accelerated deterioration of the units’ components through various degeneration
mechanisms such as fatigue, erosion, corrosion, etc, leading to
more frequent forced outages and loss of income. The start/stop
operation and varying load levels result in thermal transients
being set up in thick-walled components placing them under
stress and causing them to crack. The interruptions to operation
caused by cycling disrupts the plant chemistry and results
in higher amounts of oxygen and other ionic species being
present, leading to corrosion and fouling issues. A multitude
of other cycling related issues have been documented in the
literature [13]–[19]. Excessive cycling of base-load units could
potentially leave them permanently out of operation prior to
their expected lifetimes.
Hence cycling of base-load units will impose additional costs
on the unit, the most apparent being increased operations and
maintenance (O&M) and capital costs resulting from deterioration of the components. However, fuel costs will also increase
with cycling operation as the unit will be starting up more frequently, and also because the overall efficiency of the unit will
0885-8950/$26.00 © 2010 IEEE
TROY et al.: BASE-LOAD CYCLING ON A SYSTEM WITH SIGNIFICANT WIND PENETRATION
deteriorate. Environmental penalties will arise as a result of increased fuel usage, while income losses arise as the unit will undergo longer and more frequent outages [17], [19], [20]. Quantifying these costs is particularly difficult given the vast array
of components affected. Also, cycling related damage may not
be immediately apparent. Studies have suggested it can take up
to seven years for an increase in the failure rate to become apparent after switching from base-load to cycling [21]. The uncertainty surrounding cycling costs can lead to these costs being
under-valued by generators, which in turn can lead to increased
cycling.
This paper examines the effect that increasing penetration of
wind power will have on the operation of base-load units. The
role that interconnection and storage play in alleviating or aggravating the cycling of base-load units is investigated across
different wind penetration scenarios. Finally, the effect of increasing start-up costs (to represent increasing depreciation) on
the operation of base-load units is examined. Section II details the methodology used in the study. Section III reports the
results and discusses the impact of modeling assumptions on
these results. Section IV provides some discussion surrounding
how wind and plant cycling is treated in electricity markets.
Section V concludes the paper.
II. METHODOLOGY
A. Modeling Tool
Simulations were carried out using a scheduling model called
the Wilmar Planning Tool, which is described extensively in
[22] and [23]. The Wilmar Planning Tool was originally developed to model the Nordic electricity system and was later
adapted to the Irish system as part of the All Island Grid Study
[23]. It is currently employed in the European Wind Integration
Study [24]. The Wilmar Planning Tool was the tool of choice
for this study as it combined the benefits of mixed integer optimization with stochastic modeling. The main functionality of
the Wilmar Planning Tool is embedded in the Scenario Tree Tool
and the Scheduling Model.
The Scenario Tree Tool generates scenario trees containing
three inputs to the scheduling model: wind, load and demand
for replacement reserve. Realistic possible wind forecast errors
are generated using an auto regressive moving average (ARMA)
approach which considers the historical statistical behavior of
wind at individual sites. Historical wind speed series taken from
the various sites are then added to the wind speed forecast error
scenarios to generate wind speed forecast scenarios. These are
then transformed to wind power forecast scenarios. Load forecast scenarios are generated in a similar manner. A multi dimensional ARMA model, as in [25], is used to simulate the wind
correlation between sites. A scenario reduction technique similar to that in [26] is employed to reduce the large number of
possible scenarios generated.
In the modeling tool reserve is categorized as primary or replacement. Primary reserve, which is needed in short time scales
(less than five minutes), is supplied only by synchronized units.
The system should have enough primary reserve to cover an
outage of the largest online unit occurring at the same time as
a fast decrease in wind power production. Positive primary reserve is provided by increased production from online units or
pumped storage, whilst negative primary reserve is provided by
1089
decreased production from online units or by pumped storage
when in pumping mode. The demand for replacement reserve,
which is reserve with an activation time greater than 5 min, is
determined by the total forecast error which is defined according
to the hourly distribution of wind power and load forecast errors
and the possibilities of forced outages. A forced outage time series for each unit is also generated by the scenario tree tool using
a semi-Markov process based on given data of forced outage
rates, mean time to repair and scheduled outages is produced.
Any unit that is offline and can come online in under one hour
can provide replacement reserve.
The Scheduling Model minimizes the expected cost of the
system over the optimization period covering all scenarios generated by the scenario tree tool and subject to the generating
units’ operational constraints, such as minimum down times (the
minimum time a unit must remain offline following shut-down),
synchronization times (time taken to come online), minimum
operating times (minimum time a unit must spend online once
synchronized) and ramp rates. In order to maintain adequate
system inertia and dynamic reactive support at times of high
wind, a minimum number of large base-load units must be online at all times. Details of the objective function which contains
fuel, carbon and start-up costs are given in Appendix A and further details are included in [22]. The Generic Algebraic Modeling System (GAMS) was used to solve the unit commitment
problem using the mixed integer feature of the Cplex solver. For
all the simulations in this study the model was run with a duality
gap of 0.01%.
Rolling planning is used to re-optimize the system as new
wind and load information becomes available. Starting at noon
the system is scheduled over 36 hours until the end of the next
day. The model steps forward with a three hour time step with
new forecasts used in each step. In each planning period a three
stage stochastic optimization model is solved having a deterministic first stage, a stochastic second stage with three scenarios covering three hours and a stochastic third stage with six
scenarios covering a variable number of hours according to the
planning period in question. The state of the units at the start of
any time step must be the same as the state of the units at the
end of the previous time step.
B. Test System
The 2020 Irish system was chosen as a test case for this study
because its unique features make it suitable for investigating
base-load cycling. It is a small island system, with limited interconnection to Great Britain, a large portion of base-load plant
and significant wind penetration. Thus, potential issues with cycling of base-load units may arise on this system at a lower wind
penetration.
Various portfolios were developed in the Wilmar Planning
Tool for the All Island Grid Study [27] to investigate the effects
of different penetrations of renewables on the Irish system for
the year 2020. Portfolios 1, 2, and 5 from [27] were used in
this study and are outlined in Table I as the “moderate wind”,
“high wind”, and “very high wind” cases. A “no wind” case has
also been added. As seen in Table I, the test system is a thermal
system, with a small portion of inflexible hydro capacity and the
base-load is composed of coal and combined cycle gas turbine
(CCGT) generation. The three wind cases examined have 2000
MW, 4000 MW, and 6000 MW wind installed on the system,
1090
IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 25, NO. 2, MAY 2010
TABLE I
INSTALLED CAPACITY (MW) BY FUEL TYPE
TABLE III
CHARACTERISTICS OF A TYPICAL CCGT
AND COAL UNIT ON THE TEST SYSTEM
TABLE II
FUEL PRICES (C/GJ) BY FUEL TYPE
power to be examined. The model was run stochastically, for one
year, for the “no wind” case and each of the three wind cases to
examine the effect that increasing wind power penetration will
have on the operation of base-load units, as these are the units
with the most limited operational flexibility and as such, will
suffer the greatest deterioration from increased cycling.
To conduct a sensitivity analysis investigating the role that
storage and interconnection play in altering the impact of increasing wind penetration on base-load operation, the model
was run stochastically, for one year, for the “no wind” case and
each of the three wind cases, first, without any pumped storage
on the system and second, without any interconnection on the
system. In order to fairly compare systems without storage/interconnection to the systems with storage/interconnection, the
systems must maintain the same reliability. Thus it was necessary to replace the pumped storage units and interconnector
with conventional plant. The 292 MW of pumped storage was
replaced with three 97.5-MW open cycle gas turbine (OCGT)
units and the 1000 MW of interconnection was replaced with
nine 100-MW OCGT units (as 100 MW is always used as primary reserve, the maximum import capacity is 900 MW). The
characteristics of these units were set such that they could deliver the same capacity over the same time period as the interconnection/storage units they replaced. Thus, in terms of flexibility the systems with storage/interconnection were no more
or less flexible than the systems without storage/interconnection. The OCGT units which replaced the storage units were
capable of delivering the same amount to primary reserve (132
MW in total). The OCGT units that replaced the interconnection
did not contribute to primary reserve but instead 100 MW was
subtracted from the demand for primary reserve in each hour.
This is the assumption used when the interconnector is in place.
The cost of running these units is generally greater than the
cost of imports or production from a storage unit thus production from storage/interconnection is not shifted directly to these
units. This is advantageous in this type of study, as the operation
of other units on the system without storage/interconnection can
be observed whilst the system adequacy is not undermined by
reduced capacity, thus facilitating sensitivity analysis. For example, had a CCGT unit been used to replace the interconnector,
it would likely provide the energy that had been previously delivered by the interconnector but this would not allow examination of how the existing units on the system would be affected
which supply 11%, 23%, and 34% of the total energy demand
and represent 19%, 32%, and 42% of the total installed capacity
on the system, respectively.
The 2020 winter peak forecast is 9.6 GW and the summer
night valley is 3.5 GW. Losses on the transmission system are
included in the load. The test system includes four 73 MW
pumped storage units with a round-trip efficiency of 75% and
a maximum pumping capacity of 70 MW each and two 83 MW
CHP units with “must-run” status as they provide heat for industrial purposes. The 2020 fuel prices used are shown in Table II
and a carbon price of 30/ton was assumed. The gas prices
shown in Table II are the averages over the year and the other
fuel prices remain constant throughout. As this study is primarily concerned with the operation of base-load units, the characteristics of those units are shown in Table III.
A simplified model of the British power system is included
in which units are aggregated by fuel type. Wind and load is assumed to be perfectly forecast on the British system. The model
includes 1000 MW of HVDC interconnection between Ireland
and Great Britain and it is scheduled on an intra-day basis, i.e.,
it is rescheduled in every rolling planning period. Flows on the
interconnector to Britain are optimized such that the total costs
of both systems are minimized. A maximum of 873 MW can be
imported as 100 MW is used as primary reserve at all times and
there are 3% losses on the remainder.
C. Scenarios Examined
Different wind cases, as described in the previous section,
were used in this study to allow various penetrations of wind
TROY et al.: BASE-LOAD CYCLING ON A SYSTEM WITH SIGNIFICANT WIND PENETRATION
1091
TABLE IV
FLUCTUATIONS IN WIND POWER OUTPUT WITH INCREASING WIND
TABLE V
NUMBER OF THERMAL UNITS ONLINE WITH INCREASING WIND PENETRATION
(AVERAGED AT EACH HOUR SHOWN OVER A TWO-WEEK PERIOD IN APRIL)
Fig. 1. Annual number of start-ups and capacity factor for an average CCGT
and coal unit with increasing wind penetration.
in the absence of interconnection. The results from the systems
without storage and interconnection were compared to the base
case (i.e., with storage and interconnection).
The final part of the study examined the effect that increasing
the start-up costs of the base-load units will have on their operation. It was assumed the cost of starting these units would increase, as they experienced more wear and tear, from increased
cycling. Given the uncertainty surrounding what this increase
in costs might be [17], [19], the operation of the base-load units
was examined over a range of start-up costs. The start-up cost
of each of the base-load units on the system was increased by
a multiple of its original value and the model was run for one
year. The process was repeated with the start-up costs incremented by a greater multiple of the original amount each time.
This was carried out for the “moderate” (19% installed wind
capacity) and “very high” (42% installed wind capacity) wind
cases.
To examine the results, the base-load units were categorized
as coal or CCGT. As the total capacity of the coal and CCGT
units varied across the portfolios, the results for the individual
units in each group were normalized by their capacity to obtain
the result per MW for each unit. The average result per MW
was then obtained and this was multiplied by the capacity of a
typical coal or CCGT unit (chosen to be 260 MW and 400 MW,
respectively) to give the result for a typical coal or CCGT unit
as shown as follows:
(1)
where is the result for the th unit,
unit and is the number of units
is the capacity of the th
III. RESULTS
A. Effect of Increasing Wind Penetration on the Operation of
Base-Load Units
As the wind penetration on a power system is increased, large
fluctuations in the wind power output will become more frequent, as seen in Table IV. In addition, generation from thermal
units is increasingly displaced, thus the number of units online
will decrease. This is shown in Table V.
Therefore the onus on thermal units to compensate fluctuations in the wind power output becomes more demanding with
increasing wind penetration. Fig. 1 shows the annual number
of start-ups and capacity factor for an average sized CCGT and
coal unit of 400 MW and 260 MW, respectively, as wind penetration increases. The capacity factor is the ratio of actual generation to maximum possible generation in a given time period.
As the wind penetration grows and the variability and unpredictability involved in system operation is increased, the operation of a base-load CCGT unit is severely impacted. Moving
from 0% to 42% installed wind capacity the annual start-ups for
a typical CCGT unit rise from 22 to 98, an increase of 340%.
This increase in CCGT start-ups corresponds to a plummeting
capacity factor as seen in Fig. 1. Thus increasing levels of wind
effectively displaces CCGT units into mid-merit operation.
Similar to a CCGT unit, start-ups for a coal unit increase with
wind penetration up to 32% installed wind capacity, albeit not
as drastically as a CCGT unit. However, at penetrations greater
than 32% installed wind capacity, this correlation diverges and
the start-ups for a coal unit begin to decrease, as seen in Fig. 1.
As wind penetration grows, demand for primary reserve will
grow. Due to high part-load efficiencies, as indicated by the minimum load heat rates seen in Table III, coal units are the main
thermal providers of primary reserve on this system. In addition
to this they have low minimum outputs so at times of high wind
more coal units can remain online to meet the minimum units
online constraint thus minimizing wind curtailment. Coal units
are also highly inflexible; once taken offline it is a minimum
of ten hours (minimum down time plus synchronization time as
seen in Table III) before the unit can be online and generating
again. The combination of these characteristics, increases the
need for these units to be kept online to provide primary reserve
to the system as high levels of wind are reached. Thus, despite
the fact that the cost of starting a CCGT unit on this system is
greater than the cost of starting a coal unit as seen in Table III,
the CCGT unit has the greatest increase in start-stop cycling
with increasing wind as it does not supply a large amount of reserve to the system, has a large minimum output and can come
online in a shorter time compared to a coal unit.
As CCGT units are taken offline more frequently with increasing wind penetration, the requirement on coal units to provide reserve to the system is driven even higher. Thus, although
the capacity factor of a coal unit decreases as wind increases,
1092
IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 25, NO. 2, MAY 2010
Fig. 2. Utilization factor and annual number of hours where severe ramping is
performed for an average CCGT and coal unit with increasing wind penetration.
Fig. 3. Number of hours online for an average CCGT and coal unit with/
without storage and an increasing wind penetration.
the rate of decrease is much less than for a CCGT as seen in
Fig. 1. Therefore, as wind penetration exceeds approximately
32% installed capacity a crossover point occurs and the inflexible coal units now become the most base-loaded units on the
system whilst the relatively more flexible CCGT are forced into
two-shifting, as seen by the capacity factors in Fig. 1. Thus, if
capacity factor is indicative of the revenue earned by these units,
the units with the most limited operational flexibility are the
most rewarded at high levels of wind. This would suggest that
some form of incentive may be needed to secure investment in
flexible plants (for example OCGTs), which are commonly reported as beneficial to system operation with large amounts of
wind [28], [29].
Fig. 2 shows the utilization factor for an average base-load
coal and CCGT unit and the number of hours they perform
severe ramping as wind penetration increases. The utilization
factor is the ratio of actual generation to maximum possible
generation during hours of operation in a given period. Severe
ramping is defined in this paper as a change in output greater
than half the difference between a unit’s maximum and minimum output over one hour. Hours when the unit was staring
up or shutting down were not included. Although coal units will
avoid heavy start-stop cycling as wind levels grow by being the
main thermal providers of primary reserve and highly inflexible, they do experience increased part-load operation. This is
indicated by a drop in utilization factor from 0.94 to 0.88 as
wind levels increase from 0% to 42% installed wind capacity,
as seen in Fig. 2. The utilization factor for a CCGT unit also
decreases with increasing levels of wind as seen in Fig. 2, however, it remains high in comparison with a coal unit, indicating
the small contribution of reserve it provides to the system and
correspondingly the infrequent periods of part-load operation.
As seen in Fig. 2, both types of unit experience a dramatic increase in hours where severe ramping is required, as wind penetration exceeds 32% installed capacity. As wind penetration
moves from 32% to 42% installed wind capacity a coal unit
experiences the greatest increase in severe ramping operation
going from 4 to 78 h, compared to an increase from 4 to 32
h for a CCGT unit, as these units are now offline more often.
The sharp increase in ramping corresponds to the substantial increase in wind fluctuations seen in Table IV between 32% and
42% installed wind capacity, which must be compensated by a
smaller number of online units. Such an increase in part-load
operation and ramping can lead to fatigue damage, boiler corrosion, cracking of headers and component depreciation through
a variety of damage mechanisms. This is of major concern to
plant managers.
The results reported are for “average” CCGT and coal units.
In order to show how these results correspond to the actual results for the real units modeled, the maximum value, minimum
value, average value and standard deviation of the number of
start-ups and capacity factor for the modeled CCGT and coal
units are given in Appendix B.
B. Sensitivity Analysis
Section III-A showed the serious impact increasing levels of
wind will have on the operation of base-load units. The extent of
this impact will be determined by the generation portfolio and
the characteristics of the system. This section provides a sensitivity analysis of the effect of the portfolio on the results, by
examining the operation of the base-load units with increasing
levels of wind power when storage and interconnection are removed from the system.
1) No Storage Case: Fig. 3 shows the number of hours online for an average CCGT and coal unit on systems with and
without pumped storage and an increasing wind penetration. On
the system without pumped storage the base-load units spend
more hours online compared to the system with storage, until
a very high wind penetration (greater than 32% installed capacity for a CCGT and greater than 42% installed capacity for
a coal unit) is reached. The presence of pumped storage on a
system will displace the primary reserve contribution required
from conventional units and thus reduce the need for them to be
online. Correspondingly, an average base-load unit spends more
hours online on the system without pumped storage as there is
more requirement on the unit to be online providing primary
reserve to the system. As coal units, in this case, are the main
thermal provider of primary reserve to the system they are the
most affected by the addition of a storage unit, as seen for a
typical coal unit in Fig. 3. The difference in hours online for a
typical CCGT unit on the system with storage compared to the
system without storage is small as they are not large contributors to primary reserve.
However, at very high wind penetrations a crossover point occurs when large fluctuations in wind power output occur more
frequently, as seen in Table IV, and now the system with pumped
TROY et al.: BASE-LOAD CYCLING ON A SYSTEM WITH SIGNIFICANT WIND PENETRATION
Fig. 4. Number of start-ups for an average CCGT and coal unit with/without
storage and an increasing wind penetration.
storage is more equipped to balance these fluctuations. As the
demand for reserve is sufficiently large at very high wind penetrations, such that reserve from many thermal units is needed
in addition to the reserve from the storage units, storage will no
longer be a factor in base-load units going offline. Thus, at very
high levels of wind, base-load units now spend more hours online on the system with storage compared to the system without
storage.
Fig. 4 shows the number of start-ups for an average base-load
CCGT and coal unit on a system with and without pumped
storage as wind penetration increases. Almost no difference in
the number of start-ups for a typical CCGT unit is seen on the
systems with and without storage until installed wind reaches
greater than 32%. However, the number of start-ups for a typical coal unit is seen to be much greater on the system with
storage compared to the system without storage, again indicating that storage will most adversely affect the units that provide the largest portion of primary reserve to the system. Again
a crossover point is reached at some very high wind penetration
after which start-ups rise rapidly on the system without storage
due to large and frequent fluctuations in wind power output. This
occurs at 32% installed wind for a CCGT and greater than 42%
installed wind capacity for a coal unit. Thus, until very high
wind penetrations are reached the existence of a pumped storage
unit is shown to actually exacerbate cycling of base-load units.
2) No Interconnection Case: Fig. 5 compares the number
of hours spent online by a typical CCGT and coal unit on systems with and without interconnection, as wind is increased.
The base-load units are seen to spend significantly more hours
online on the system without interconnection compared to the
system with interconnection until a very high wind penetration
is reached.
Due to a large portion of base-load nuclear plant and cheaper
gas prices compared with Ireland, the market price for electricity
tends to be cheaper in Great Britain. As a consequence Ireland
tends to be a net importer of electricity from Great Britain and
as such will import electricity before turning on domestic units.
Thus interconnection to Great Britain displaces conventional
generation on the Irish system, forcing units down the merit
order and exacerbating plant cycling. Without the option to import electricity, as in the “no interconnection case”, all demand
must be met by domestic units requiring more units to be online
generating more often. Thus a typical CCGT and coal unit are
1093
Fig. 5. Number of hours online for an average CCGT and coal unit with/
without interconnection and an increasing wind penetration.
Fig. 6. Number of start-ups for an average CCGT and coal unit with/without
interconnection and an increasing wind penetration.
seen in Figs. 5 and 6 to spend more hours online and have less
start-ups on the system without interconnection.
However, as seen in Fig. 5 at some wind penetration between
32% and 42% installed wind capacity for a CCGT unit and
greater than 42% installed capacity for a coal unit, a crossover
point will occur when the units spend more hours online on the
system with interconnection. As very high wind penetrations
are reached, the electricity price in Ireland undercuts British
prices more often making exports economically viable. Thus at
very high penetrations of wind, the system with interconnection can deal with large fluctuations in the wind power output
via imports/exports more favorably and avoid plant shut-downs.
Thus interconnection is shown not to benefit the operation of
base-load units on a system that is a net importer until wind
penetration increases to such point that exports are economically viable.
C. Effect of Increasing Start-Up Costs
Having shown in Sections III-A and B the severe impact increasing wind penetration will have on the operation of the baseload units, this section now examines how the increasing costs
imposed on these units by cycling operation, will subsequently
affect their operation. A component of a unit’s start-up cost
should be the cost of wear and tear inflicted on the unit during
the start-up process [16]. However, given the uncertainty in determining such a cost, this aspect is often neglected, leading to
the units being scheduled to start more frequently, yielding more
1094
Fig. 7. Number of base-load start-ups for increasing start-up costs.
cycling related damage. This section examines how the operation of the base-load units changes as the start-up costs are incrementally increased to represent the increasing depreciation
of the unit.
1) Start-Ups: The number of start-ups for an average CCGT
and coal unit is shown in Fig. 7, as start-up costs are increased,
with 19% and 42% installed wind capacity, respectively. Increasing the start-up costs of a CCGT unit results in a substantial
reduction in start-stop cycling, particularly at the higher wind
penetration. This indicates a feedback effect, whereby increased
cycling will lead to increased costs, but when these costs are
included in the cost function, cycling will subsequently be reduced. With 42% installed wind capacity, increasing the start-up
costs by a factor of 6 sees the start-ups for a CCGT drop from
98 to 27, a decrease of 72%. Doubling the start-up costs of a
coal unit in the low wind case reduced start-ups by 19, a 68%
reduction. No further reduction in coal start-ups was possible
as these units were then at their minimum number of annual
start-ups (governed by scheduled and forced outages).
A greater reduction in cycling is achieved by increasing
start-up costs on the system with 42% installed wind capacity
compared to the system with 19% installed wind capacity, as
this system can export more due to lower electricity prices.
Increasing the start-up costs of the base-load units in Ireland
by a factor of 6, results in a 29% increase in exports on the
system with 42% installed wind capacity as it becomes more
economical to allow the base-load units in Ireland to stay online
and avoid shut-downs by increasing exports to Britain.
2) Ramping and Part-Load Operation: Fig. 8 shows the
number of hours that severe ramping is required by an average
CCGT and coal unit, as start-up costs are increased with 19%
and 42% installed wind capacity. Fig. 9 shows the utilization
factor for an average CCGT and coal unit, with 19% and 42%
installed wind capacity as their start-up costs are increased. The
trade-off for the reduction in start-stop cycling of base-load
units, achieved by increasing the start-up costs, is an increase
in ramping activity as seen in Fig. 8 and part-load operation
as seen in Fig. 9, which will also leads to plant deterioration
although it is reported to be less costly compared with start-ups
[30].
By increasing the start-up costs of the base-load units,
start-ups are reduced and these units are kept online more, but
at the expense of more flexible units which are taken offline.
As a result the number of hours when the base-load units are
IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 25, NO. 2, MAY 2010
Fig. 8. Number of hours of severe ramping duty for increasing start-up costs.
Fig. 9. Utilization factor for increasing start-up costs.
the only thermal units online increases with increasing start-up
costs. During such hours there will be a considerable ramping
requirement on these units to balance fluctuations in the wind
power output. As there will be even less thermal units online
in the 42% installed wind capacity case compared to the 19%
installed capacity case the greatest increase in ramping is
observed for the 42% installed wind capacity case as start-up
costs are increased, as seen in Fig. 8. Some inconsistencies
in the trend can occur because “severe ramping” is defined
discretely, as seen for a CCGT with 42% installed wind.
As the base-load units are being kept online more often, as
their start-up costs are increased, they will experience increased
part-load operation as indicated by the reduction in utilization
factor in Fig. 9. As start-up costs are increased sufficiently it
becomes more economical to run these units at part-load, than
to take them offline and forgo expensive start-up costs at a
later time. The greater increase in part-load operation occurs
on the system with 42% installed wind capacity compared to
the system with 19% installed wind capacity, corresponding
to the large reduction in start-ups seen at 42% installed wind
capacity. The difference in start-ups and ramping for a CCGT
and coal unit between 19% installed wind and 42% installed
wind is also seen in Figs. 1 and 2 for the original start-up costs
and for brevity is not discussed again here.
D. Effect of Modeling Assumptions
The model used was limited to hourly time resolution. The
lack of intra-hourly data may have lead to the severity of the
TROY et al.: BASE-LOAD CYCLING ON A SYSTEM WITH SIGNIFICANT WIND PENETRATION
cycling being seriously underestimated, for example the severe
ramping events. The frequency of severe ramping events found
in the study may be underestimated as severe ramps may have
occurred over shorter time frames than one hour. Also, such
a sizeable ramp occurring over a period shorter than one hour
would have a much more damaging effect on the unit.
For all simulations, rolling planning with a three hour time
step was used. Had the system been re-optimized more regularly, the wind and load forecasts would have been updated more
often. However, [22] shows this would have minimal impact on
the operation of the base-load units examined here so a three
hour time step was deemed adequate.
IV. DISCUSSION
How electricity markets evolve to manage plant cycling is beyond the scope of this paper, however, this section offers some
discussion as to how cycling costs could be represented and
areas for future market development with a large wind penetration. In many electricity markets generators submit complex
bids for energy in addition to the technical characteristics of
the plant. If the current trend for wind development continues,
plant cycling, as shown in this paper, will inevitably becoming
an increasing concern and generators may subsequently alter
their bids or plant characteristics in order to minimize cycling
damage. Section III-C examines how by taking the cost of cycling into consideration in a unit’s start-up cost, subsequent cycling can be reduced. Generators in SEM, the Irish electricity
market, are directed to include cycling costs in their start-up
costs so this approach was taken in this paper.
Cycling costs could also be included in no-load or energy
costs, or even defined as a new market product such as ramping
costs [31]. However, increasing the energy cost will also increase the marginal cost of the unit, which risks changing the
position of the unit in the merit order and inducing further cycling. Alternatively cycling costs could be incorporated in a
unit’s shut-down costs. The Wilmar Planning Tool used in this
study does not model shut-down costs at present. Future work
could investigate the effect of incorporating shut-down costs in
the scheduling algorithm on a generators dispatch.
As cycling costs are difficult to quantify, generators may use
the opportunity to exercise market power. For example a generator may increase the start-up costs excessively in order to avoid
shut-down, although this strategy may result in them being left
offline following a trip or scheduled shut-down because of their
excessive start-up cost. Thus some may instead favor setting a
maximum number of start-ups a unit can carry out over a period
of time, however, this approach would unfairly reward inflexible
units and provide no incentive to improve operational flexibility.
In some electricity markets generators submit simple bids.
This can result in increased start-ups for generators as no explicit consideration of the cost of starting the unit is taken. Incorporating wind in such a market would induce further cycling,
indicating that a move to complex bidding could be beneficial.
Longer scheduling horizons that take future wind forecasts into
consideration may also reduce plant start-ups, however the forecast error increases with the time horizon. Thus enabling a later
gate closure in a market with a significant wind penetration,
1095
which would allow the most up-to-date wind forecasts to be employed, could be more effective at reducing unnecessary plant
start-ups [32].
V. CONCLUSIONS
Increasing wind penetration on a power system will lead to
changes in the operation of the thermal units on that system, but
most worryingly to the base-load units. The base-load units are
impacted differently by increasing levels of wind, depending on
their characteristics. CCGT units see rapid increases in startstop cycling and plummeting capacity factor and are essentially
displaced into mid-merit operation. On the test system examined coal units are the main thermal providers of primary reserve
to the system and as a result see increased part-load operation
and ramping. This increase in cycling operation will lead to increased outages and plant depreciation.
Certain power system assets are widely reported to assist
the integration of wind power. This paper examined if storage
and interconnection reduced cycling of base-load units by
comparing a system with storage and interconnection to a
system without storage and without interconnection, across a
range of wind penetrations. It was found that until very high
penetrations of wind are reached storage will actually displace
the need for base-load units to be online providing reserve to
the system. This results in increased cycling of base-load units
compared to the system without storage. Similarly, for a system
that is a net importer, interconnection will actually displace
generation from domestic units, also resulting in increased
cycling of base-load units compared to a system without interconnection. At very large penetrations of wind a crossover
point exists, where larger and more frequent fluctuations in the
wind power output, can be dealt with more effectively on a
system with interconnection and storage and thus the system
with storage and interconnection becomes the most favorable
to the operation of base-load units.
Having shown how the operation of the base-load units is
dramatically affected by increasing levels of wind power and
assuming this would lead to added costs in various guises, the
effect that increasing start-up costs for base-load units had on
their subsequent operation was examined. This showed that as
the cost of starting a base-loaded CCGT unit increased, startstop cycling of the unit was subsequently reduced. However, a
reduction in start-ups is seen to be correlated with an increase
in part-load operation and ramping.
APPENDIX A
WILMAR OBJECTIVE FUNCTION
The objective function shown in (A1) consists of operating
fuel cost, start up fuel cost (if a unit starts in that hour), emissions costs and penalties incurred for not meeting load or reserve targets. If a unit is online at the end of the day, its start-up
costs are subtracted from the objective function to ensure that
there are still units online at the end of the optimization period.
The decision variable is given in the first three lines, showing
whether a unit is online or offline. Further detail on the formulation of the unit commitment problem is given in [22].
1096
IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 25, NO. 2, MAY 2010
Indices:
TABLE VI
VARIATION IN CCGT START-UPS WITH INCREASING WIND
F
i,I
r,R
s,S
START
t,T
USEFUEL
Fuel.
Unit group.
Region.
Scenario.
Units with start-up fuel consumption.
Time.
Unit using fuel.
TABLE VII
VARIATION IN COAL START-UPS WITH INCREASING WIND
Parameters:
EMISSION
END
k
L
LOAD
PRICE
REP
SPIN
TAX
Rate of emission.
Endtime of optimization period.
Probability of scenario.
Infeasibility penalty.
Penalty for loss of load.
Fuel price.
Penalty for not meeting replacement reserve.
Penalty for not meeting primary reserve.
Emission tax.
TABLE VIII
VARIATION IN CCGT CAPACITY FACTOR WITH INCREASING WIND
Variables:
CONS
OBJ
U
V
ONLINE
QDAY
QINTRA
QREP
QSPIN
+, -
Fuel consumed.
Objective function.
Relaxation variable.
Decision variable—on or off.
Integer on/off for unit.
Day ahead demand not met.
Intra day demand not met.
Replacement reserve not met.
Primary reserve not met.
Up, down regulation.
TABLE IX
VARIATION IN COAL CAPACITY FACTOR WITH INCREASING WIND
APPENDIX B
SUMMARY OF NON-NORMALIZED BASE CASE RESULTS
Tables VI–IX indicate the variation in start-ups and capacity factor of the CCGT and coal units in the base case (i.e.,
Tables VI–IX relate to Fig. 1), for each of the wind penetrations.
The maximum value, minimum value, average and standard
deviation are shown. It can be seen that the CCGT units have
a greater spread in start-ups compared to the coal units and
the standard deviation of start-ups is least at the highest wind
case for both types of units. For capacity factor the spread in
results across the units increased as the wind increased, with
the CCGT units again having a greater variation compared to
the coal units, however, there are more CCGT units than coal
units in each of the wind cases.
REFERENCES
(A1)
[1] H. Holttinen, “Impact of hourly wind power variations on the system
operation in the Nordic countries,” Wind Energy, vol. 8, no. 2, pp.
197–218, Apr./Jun. 2005.
[2] B. C. Ummels, M. Gibescu, E. Pelgrum, W. Kling, and A. Brand, “Impacts of wind power on thermal generation unit commitment and dispatch,” IEEE Trans. Energy Convers., vol. 22, no. 1, pp. 44–51, Mar.
2007.
TROY et al.: BASE-LOAD CYCLING ON A SYSTEM WITH SIGNIFICANT WIND PENETRATION
[3] G. Dany, “Power reserve in interconnected systems with high wind
power production,” in Proc. IEEE Power Tech Conf., vol. 4, 6 pp, 2001.
[4] Ahlstrom, L. Jones, R. Zavadil, and W. Grant, “The future of wind
forecasting and utility operations,” IEEE Power and Energy Mag., vol.
3, no. 6, pp. 57–64, Nov.–Dec. 2005.
[5] The Effect of Integrating Wind Power on Transmission System Planning, Reliability and Operations, Report prepared for New York State
Energy Research and Development Agency, 2005. [Online]. Available:
http://www.nyserda.org/publications/wind_integration_report.pdf.
[6] P. Meibom, C. Weber, R. Barth, and H. Brand, “Operational costs induced by fluctuating wind power production in Germany and Scandinavia,” Proc. IET Renew. Power Gen., vol. 3, no. 1, pp. 75–83, Jan.
2009.
[7] M. Braun, “Environmental external costs from power generation by renewable energies,” Master’s thesis, Stuttgart Univ., Stuttgart, Germany,
2004.
[8] H. Holttinen, V. T. T. Finland, J. Pedersen, and E. Denmark, “The effect of large scale wind power on thermal system operation,” in Proc.
4th Int. Workshop Large-Scale Integration of Wind Power and Transmission Networks for Offshore Wind Farms, Billund, Denmark, Oct.
2003.
[9] L. Goransson and F. Johnsson, “Dispatch modeling of a regional power
generating system—Integrating wind power,” Renew. Energy, vol. 34,
no. 4, pp. 1040–1049, Apr. 2009.
[10] L. Balling and D. Hoffman, Fast Cycling Towards Bigger Profits,
Modern Power Systems, 2007. [Online]. Available: http://www.modernpowersystems.com.
[11] Variability of Wind Power and Other Renewables—Management Options and Strategies, International Energy Agency. [Online]. Available:
http://www.iea.org/textbase/papers/2005/variability.pdf.
[12] Large Scale Integration of Wind Energy in the European Power Supply:
Analysis, Issues and Recommendations, European Wind Energy Association. [Online]. Available: http://www.ewea.org/index.php?id=178.
[13] R. Viswanathan and J. Stringer, “Failure mechanisms of high temperature components in power plants,” Trans. ASME, vol. 122, pp. 246–255,
Jul. 2000.
[14] V. Viswanathan and D. Gray, Damage to Power Plants Due to Cycling,
EPRI, Palo Alto, CA, 2001, Tech. Rep. 1001507.
[15] J. Gostling, “Two shifting of power plant: Damage to power plants due
to cycling—A brief overview,” OMMI, vol. 1, no. 1, Apr. 2002. [Online]. Available: http://www.ommi.co.uk/.
[16] K. D. Le, R. R. Jackups, J. Feinstein, H. Thompson, H. M. Wolf, E. C.
Stein, A. D. Gorski, and J. S. Griffith, “Operational aspects of generation cycling,” IEEE Trans. Power Syst., vol. 5, no. 4, pp. 1194–1203,
Nov. 1990.
[17] F. J. Berte and D. S. Moelling, “Assessing the true cost of cycling is a
challenging assignment,” Combined Cycle J., pp. 23–25, 2003.
[18] C. Johnston, “An approach to power station boiler and turbine life
management,” in Proc. World Conf. NDT, Montreal, QC, Canada, Sep.
2004.
[19] S. A. Lefton, P. M. Besuner, and G. P. Grimsrud, “Managing utility
power plant assets to economically optimize power plant cycling costs,
life, and reliability,” in Proc. 4th IEEE Conf. Control Applications, Albany, NY, Sep. 1995.
[20] E. Denny and M. O’Malley, “The impact of carbon prices on generation
cycling costs,” Energy Pol., vol. 37, no. 4, pp. 1204–1212, Apr. 2009.
[21] S. A. Lefton, P. M. Besuner, G. P. Grimsrud, A. Bissel, and G. L.
Norman, Optimizing Power Plant Cycling Operations While Reducing
Generating Plant Damage and Costs at the Irish Electricity Supply
Board. Sunnyvale, CA: Aptech Eng. Service, 1998.
[22] A. Tuohy, P. Meibom, E. Denny, and M. O’Malley, “Unit commitment
for systems with significant wind penetration,” IEEE Trans. Power
Syst., vol. 24, no. 2, pp. 592–601, May 2009.
[23] Wind Variability Management Studies, All Island Renewable Grid
Study—Workstream 2B, 2008. [Online]. Available: http://www.
dcmnr.gov.ie.
[24] European Wind Integration Study. [Online]. Available: http://www.
wind-integration.eu/.
1097
[25] L. Soder, “Simulation of wind speed forecast errors for operations planning of multi-area power systems,” in Proc. 2004 IEEE Int. Conf. Probabilistic Methods Applied to Power Systems, Ames, IA, Sep. 2004, pp.
723–728.
[26] J. Dupacova, N. Growe-Kuska, and W. Romisch, “Scenario reduction
in stochastic programming: An approach using probability metrics,”
Math. Program., vol. 95, no. 3, pp. 493–511, 2003.
[27] High Level Assessment of Suitable Generation Portfolios for the AllIsland System in 2020, All Island Renewable Grid Study—Workstream
2A, 2008. [Online]. Available: http://www.dcmnr.gov.ie.
[28] B. Kirby and M. Milligan, “Facilitating wind development: The importance of electric industry structure,” Elect. J., vol. 21, no. 3, pp. 40–54,
Apr. 2008.
[29] G. Strbac, A. Shakoor, M. Black, D. Pudjianto, and T. Bopp, “Impact of
wind generation on the operation and development of the UK electricity
systems,” Elect. Power Syst. Res., vol. 77, no. 9, pp. 1214–1227, Jul.
2007.
[30] Editorial, “Profitable operation requires knowing how much it costs to
cycle your unit,” Combined Cycle J., pp. 49–52, 2004.
[31] M. Flynn, M. Walsh, and M. O’Malley, “Efficient use of generator resources in emerging electricity markets,” IEEE Trans. Power Syst., vol.
15, no. 1, pp. 241–249, Feb. 2000.
[32] C. Hiroux and M. Saguan, “Large-scale wind power in European electricity markets: Time for revisiting support schemes and market designs,” Energy Policy, to be published.
Niamh Troy (GS’09) received the B.Sc. degree in applied physics from the University of Limerick, Limerick, Ireland. She is currently pursuing the Ph.D. degree at the Electricity Research Centre in the University College Dublin, Dublin, Ireland.
Eleanor Denny (M’07) received the B.A. degree
in economics and mathematics, the M.B.S. degree
in quantitative finance, and the Ph.D. degree in
wind generation integration from University College
Dublin, Dublin, Ireland, in 2000, 2001, and 2007,
respectively.
She is currently a Lecturer in the Department
of Economics at Trinity College Dublin and has
research interests in renewable generation and
integration, distributed energy resources, and system
operation.
Mark O’Malley (F’07) received the B.E. and Ph.D.
degrees from University College Dublin, Dublin, Ireland, in 1983 and 1987, respectively.
He is a Professor of electrical engineering at
University College Dublin and is director of the
Electricity Research Centre with research interests
in power systems, control theory, and biomedical
engineering.
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
IEEE TRANSACTIONS ON POWER SYSTEMS
1
Multi-Mode Operation of Combined-Cycle Gas
Turbines With Increasing Wind Penetration
Niamh Troy, Member, IEEE, Damian Flynn, Senior Member, IEEE, and Mark OMalley, Fellow, IEEE
Abstract—As power systems evolve to incorporate greater penetrations of variable renewables, the demand for flexibility within
the system is increased. Combined-cycle gas turbines are traditionally considered as relatively inflexible units, but those which
incorporate a steam bypass stack are capable of open-cycle operation. Facilitating these units to also operate in open-cycle mode can
benefit the power system via improved system reliability, while reducing the production needed from dedicated peaking units. The
utilization of the multi-mode functionality is shown to be dependent on the flexibility inherent in the system and the manner in
which the system is operated.
Index Terms—Power system modeling, thermal power generation, wind power generation.
I. INTRODUCTION
C
OMBINED-CYCLE gas turbines (CCGTs) are a type of
power generating unit that achieve high efficiencies (up
to 60%) by capturing the waste heat from a gas turbine in a heat
recovery steam generator (HRSG) and using it to produce superheated steam to drive a steam turbine [1]. The high efficiencies
achieved, combined with their ease of installation, short-build
times, and relatively low gas prices, have made the CCGT a popular technology choice [2], [3]. In the Republic of Ireland, for
example, 43% of the installed thermal capacity is CCGT technology, while in the markets of Texas (ERCOT) and New England (NEPOOL), CCGTs represent 37% of the total installed
capacity.
The operational flexibility of a CCGT unit is limited by the
steam cycle, which contains many thick-walled components,
necessary to withstand extreme temperatures and pressures [4],
[5]. To avoid differential thermal expansion across these components and the subsequent risk of cracking, these components
must be brought up to temperature slowly, resulting in slower
start-up times and ramp rates for the unit overall [6]. However,
by incorporating a bypass stack upstream of the HRSG at the
design stage, a CCGT unit has the option to bypass the steam
cycle and run in open-cycle mode, whereby exhaust heat from
Manuscript received March 02, 2011; revised June 24, 2011; accepted July
22, 2011. This work was conducted in the Electricity Research Centre, University College Dublin, Ireland, which is supported by the Commission for Energy
Regulation, Bord Gais Energy, Bord na Mona Energy, Cylon Controls, EirGrid,
the Electric Power Research Institute (EPRI), ESB Energy International, ESB
Energy Solutions, ESB Networks, Gaelectric, Siemens, SSE Renewables, and
Viridian Power & Energy. This work was supported by Science Foundation Ireland under Grant Number 06/CP/E005. Paper no. TPWRS-00128-2011.
The authors are with the School of Electrical, Electronic, and Mechanical Engineering, University College Dublin, Dublin, Ireland (e-mail:
[email protected]; [email protected]; [email protected]).
Digital Object Identifier 10.1109/TPWRS.2011.2163649
the gas turbine is ejected directly into the atmosphere via the
bypass stack [6]. This reduces the power output and efficiency
of the plant but offers greater operational flexibility. Running
in open-cycle mode, the gas turbine has a short start-up time of
15 to 30 min and is capable of changing load quickly. However,
bypass stacks are not always incorporated because they can potentially lead to leakage losses, thus reducing plant efficiency,
while also introducing additional capital costs [1].
As international energy policy drives ever greater penetrations of renewable energy, wind power is set to represent a
larger portion of the generation mix [7]. This is driving a greater
demand for flexibility within power systems in order to deal
with high penetrations of variable and difficult to predict energy sources [8], [9]. Storage, interconnection, and responsive
demand are commonly cited as flexible options for dealing with
variability issues [10]–[12]; however, these options have considerable costs associated with them. Facilitating open-cycle
operation of CCGT units that have the technical capability to
run in open-cycle mode (i.e., those with a bypass stack) can
also deliver much needed flexibility to a system with a high
wind penetration. This resource is often technically available,
but inaccessible due to market arrangements.
In order to derive the greatest benefits from a CCGT unit that
can run in open-cycle mode, it is necessary for the scheduling
algorithm to explicitly consider both modes of operation for the
unit, i.e., open-cycle and combined-cycle [13]. These will have
greatly different technical and cost characteristics and so need
to be declared individually. Currently most markets do not facilitate CCGT units to submit multiple bids representing different modes of operation; thus, presently open-cycle operation
of a CCGT unit is typically limited to periods when the steam
section is undergoing maintenance. However, some U.S. systems have begun addressing this issue to varying degrees, with
ERCOT and CAISO seeking to implement configuration-based
modeling of CCGTs [14], [15].
The option to run in open-cycle mode could also provide
benefits for the generators. Renewable integration studies have
shown that CCGT units will experience significant decreases
in running hours and thus will receive less revenue from the
market as they are displaced by greater levels of wind generation which has an almost zero marginal cost [16]–[20]. Due
to their high minimum loads, CCGTs are shut down frequently
with high wind penetrations as they cannot reduce output sufficiently to accommodate the wind power output [16]. By facilitating CCGT units to operate in open-cycle mode, these units
may have a new opportunity to capture revenue from increased
operation during periods when they might otherwise be offline.
For example, if a CCGT unit has been forced offline by high
0885-8950/$26.00 © 2011 IEEE
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
2
IEEE TRANSACTIONS ON POWER SYSTEMS
wind generation on the system, it may have the opportunity to
run as a peaking unit.
This paper builds on preliminary work in [21] and includes
improved modeling of CCGTs from that in [21] to examine
if a power system with a high wind penetration can benefit
from the additional flexibility introduced when these units are
facilitated to operate in open-cycle mode, when technically
feasible and economically suitable. The all-island Irish 2020
system [22] is considered here as it is expected to contain both
a large share of wind power and CCGT units. In addition, as
it is a small, island system that is weakly interconnected, the
challenges of maintaining the supply/demand balance with a
high wind penetration are exacerbated, and so the solutions
found can hold insights for other systems pursuing large-scale
wind power. Section II describes the modeling tool used in this
study and also the changes that were made to model multi-mode
operation of CCGTs. Section III outlines the test system used.
Section IV describes the results of the study and Section V
concludes the paper.
II. MODELING TOOL
The Wilmar Planning Tool is a stochastic, mixed integer
unit commitment and economic dispatch model, originally
developed to model the Nordic electricity system and later
adapted to the Irish system as part of the All Island Grid Study
[22]–[25]. The main functionality of the Wilmar Planning Tool
is embedded in the Scenario Tree Tool and Scheduling Model.
The Scenario Tree Tool utilizes historical wind power or wind
speed data, load data, and wind and load forecasts for different
time horizons to identify an auto regressive moving average
(ARMA) series which can then simulate wind and load forecast errors for various time horizons [26]. These simulated wind
and load forecasts errors are paired in a random way before a
scenario reduction technique, following the approach of [27], is
applied. The wind and load forecast errors are combined with
scaled up wind and load time series to produce wind power production and load forecast scenarios. For each scenario, the demin) is calcumand for replacement reserve (activation time
lated based on a comparison of the hourly power balance considering perfect forecasts and no forced outages with the power
balance considering scenarios of wind and load forecast errors
as well as forced outages. A percentile of the deviation between
the compared power balances must be covered by replacement
reserves; in this case, the 90th percentile is chosen based on current practice [23]. A forced outage time series for each unit is
also generated by the Scenario Tree Tool using a semi-Markov
process based on historical plant data of forced outage rates,
mean time to repair, and scheduled outages.
The model can also be run in deterministic and perfect foresight modes whereby only one wind generation and load scenario is planned for. In deterministic mode, this scenario is the
expected value of wind and load. The expected value of wind
is found by summing, for all (post-reduction) scenarios, the
product of the wind power forecasts and their probability of
occurring. The expected value of load and replacement reserve
is found similarly [24]. Consequently, the scenario planned for
will differ from the realized scenario. This mode is typical of
the scheduling process currently practiced by most system operators, i.e., only one scenario is planned for and it will contain
some level of forecast error. Perfect foresight mode contains no
forecast error for wind generation or load but forced outages still
occur, as with all other modes.
The Scheduling Model minimizes the expected costs for all
scenarios, subject to system constraints for reserve and the minimum number of units online (6 units in the Republic of Ireland and 2 units in Northern Ireland). These costs include fuel,
carbon, and start-up fuel costs (always assumed to be hot starts).
In addition to replacement reserve, one category of spinning
reserve, namely tertiary operating reserve (TR1), is modeled,
which has a response time of 90 s to 5 min and is only supplied
by online units. Enough spinning reserve must be available to
cover an outage of the largest online unit occurring concurrently
with a fast decrease in wind power production over the TR1 time
frame, as described in [28].
Generator constraints such as minimum down times, synchronization times, minimum operating times, and ramp rates must
also be obeyed. Rolling planning is employed to re-optimize the
system as new wind generation and load information become
available. Starting at noon each day, the system is scheduled
over 36 h until the end of the next day. The model steps forward
with a 3-h time step and reschedules the units based on information from new forecasts. The model produces a year-long dispatch at an hourly time resolution for each individual generating
unit. Further detail on the model and formulation of the unit
commitment problem can be found in [23]. The Generic Algebraic Modeling System (GAMS) is used to solve the unit commitment problem using the mixed integer feature of the Cplex
solver (version 12). For all simulations in this study, the model
was run with a duality gap of 0.5%. A year-long simulation takes
h when run in deterministic mode or
h in stochastic
mode, on an Intel core quad 3-GHz processor with 4 GB of
RAM.
A. Modeling Multi-Mode Operation of CCGTs
In order to examine the potential for multi-mode operation
”, of all CCGT units capable of proof CCGT units a set, “
longed open-cycle operation, i.e., those with bypass stacks, was
” corresponds to these CCGT units
defined. The set “
when run in open-cycle mode. CCGT units comprised of two
” units, as inor more gas turbines will have multiple “
dicated by index “a”. The relation “multi-mode” is defined to
” with the corresponding member(s)
pair each member of “
”. To ensure the mutually exclusive operation of
of “
these “
” units and the corresponding “
” units, the
constraint shown in (1) was added to the model, where
is the state binary variable which describes the online status of
the unit. This allows the model to dispatch, when economically
” (combined-cycle mode) or any/all of
optimal, either the “
” units (open cycle mode), for all
the corresponding “
scenarios “s” and time steps “t”, but not both simultaneously as
they are in reality the same unit:
(1)
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
TROY et al.: MULTI-MODE OPERATION OF COMBINED-CYCLE GAS TURBINES WITH INCREASING WIND PENETRATION
3
Equation (2), taken from [29], sets the state binary variables
or
equal to 1 for all units “i”, when a unit is started
up or shut down, respectively:
(2)
When modeling multi-mode operation of CCGT units, two
new circumstances arise when calculating the start-up fuel
, which must be explicitly represented.
consumption,
Firstly, when a “
” unit transitions from conventional combined-cycle operation into open-cycle operation no start-up fuel
” unit as represented by inequality
is consumed by the “
(3), where
is the start-up energy used by each unit
” unit starts from zero
(measured in MWh). When the “
and
), the first term
production (
on the right-hand side of inequality (3) determines the fuel used
by the unit while the second term equals zero. Alternatively,
when the unit switches from combined-cycle to open-cycle
and
), the second term
operation (
causes the right-hand side of (3) to equal zero. Setting
as a positive variable and using an inequality condition ensures
” unit is shutting down and the corresponding
that when a “
” unit is not starting up,
will be 0:
“
Fig. 1. CCGT start-up from open-cycle mode.
and (6), where
is a unit’s minimum stable operating
is a unit’s maximum capacity (MW),
level (MW) and
” unit or the “
” unit is onensure that if either the “
line, then the “
” unit cannot contribute to the portion
of replacement reserve that is provided from offline units. This
” unit is online
is necessary to avoid the situation where a “
” unit to conand the model allows the corresponding “
tribute to offline replacement reserve:
(3)
(5)
The second circumstance relates to the unit transitioning
from open-cycle to combined-cycle operation. In this case, the
start-up fuel consumed is less than the start-up fuel used in
bringing the CCGT online from zero production, as some of
this start-up fuel has already been used to bring the unit online
in open-cycle mode and the gas section of the plant is in a hot
state. As an approximation, the start-up fuel used to bring the
unit into combined-cycle operation from open-cycle operation
” and a
is the difference between the start-up fuel for the “
”, as seen in
fraction, , of the start-up fuel for the “
(4). Based on the operating experience of generators, was
” unit is started from
chosen to be 0.5 here. When the “
zero production (
and
), the first
term on the right-hand side of (4) provides the start-up fuel
consumed, while the second term equals zero. When the unit
switches from open-cycle to combined-cycle operation, the
second term is included, thus approximating the start-up fuel
consumed in this situation:
(6)
(4)
In the Wilmar model, any unit can contribute to the target
for replacement (non-spinning) reserve, provided that an offline
unit can come online in time to provide reserve for the hour
in question and the reserve available from an online unit is not
needed to meet spinning reserve targets. In Wilmar, the contribution from online and offline units to the replacement reserve
(MW), are calculated individually. In this case the
target,
” units cannot provide offline replacement reserve as they
“
” units
have long start-up times, but the corresponding “
can, given their fast start-up times. The constraints shown in (5)
Improved modeling of plant start-ups was also implemented
following the formulation given in [29]. This allows for those
units with start-up times greater than 1 h to be block-loaded over
the course of their start-up time. In earlier versions of the Wilmar
model, units remained at zero production for the duration of
start-up process. The addition of this feature significantly increased the computation time, so only the start-up process of the
CCGT units was modeled in detail. Other units with a start-up
time greater than 1 h, namely the coal-fired units, typically have
fewer starts over the year and lower minimum operating levels
relative to the CCGTs and so modeling their start-up process in
detail would have little impact on the results.
When the bypass stack is utilized to switch from combined-cycle to open-cycle operation, the transition is automatic
and occurs without shutting down the gas turbine or reducing
its power output. However, the transition from open-cycle to
combined-cycle operation is dependent on the temperature state
of the boiler. Therefore, if the CCGT unit has been operating
for a period of time in open-cycle mode and is then scheduled to
switch to combined-cycle mode, its output must adjust in order
to achieve the correct HRSG inlet temperature, as depicted in
Fig. 1. This was implemented by setting the allowable power
from [29]) for each interval of the CCGT’s
output (
start-up process, which begins at hour 0 in Fig. 1, such that the
appropriate soak time is achieved.
Scheduled outages for each unit, determined from historical
experience [22], are inputted in time-series format to the Wilmar
model. In this case, CCGT units with the capability to operate
in open-cycle mode are considered to be available to run in
open-cycle mode for a portion of their scheduled outage. Given
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
4
IEEE TRANSACTIONS ON POWER SYSTEMS
TABLE I
GENERATION MIX OF TEST SYSTEM
TABLE III
CHARACTERISTICS OF CCGT UNITS (CAPABLE OF MULTI-MODE
OPERATION) IN COMBINED- AND OPEN-CYCLE MODES
TABLE II
FUEL PRICES BY FUEL TYPE
that gas turbine equipment is more accessible and compact in
comparison with the steam turbine equipment, it was assumed
that one third of the maintenance period was sufficient for the
gas turbine.
III. TEST SYSTEM
The test system used is the Irish 2020 system, based on
portfolio 5 from the All Island Grid Study [22], [30]. Four
103.5-MW OCGT units were removed from the original grid
study portfolio as recent generation adequacy reports would
indicate they are unlikely to be built by 2020 [31]. Table I shows
the number of units, installed capacity, and average operating
cost (fuel) by generation type. (The multi-mode capable CCGT
units in open-cycle mode are shown on the last row.) Three
different levels of installed wind power were examined: 2000,
4000, and 6000 MW, which supply 15%, 29%, and 44% of
the total energy demand, respectively. Fuel prices are as given
in Table II. Base-load gas generators (i.e., CCGTs and CHP)
are assumed to have long-term fuel contracts and therefore
pay a cheaper fuel price compared to mid-merit gas generators
(i.e., OCGTs, ADGTs, and legacy CCGTs). Differences in the
fuel price for coal and gas oil in the Republic of Ireland and
Northern Ireland reflect varying delivery costs. The original
demand profile from [22] with a 9.6-GW peak and 54-TWh
total demand was scaled down to a profile with a 7.55-GW peak
and 42-TWh total demand to reflect a reduction in predicted
demand, seen in recent long-term forecasts [31].
The test system assumes that there is 1000 MW of HVDC
interconnection in place between Ireland and Great Britain and
it is scheduled on an intra-day basis, i.e., it can be rescheduled
in every 3-h rolling planning period. A simplified model of the
British power system is included, with aggregated units, no integer variables for generators and where wind generation and
load are assumed to be perfectly forecast. The total demand in
Britain is assumed to be 370 TWh with a peak of 63 GW and
the installed wind capacity is assumed to be 14 GW. A carbon
was assumed.
price of
Five (of the ten) CCGT units on the Irish system include bypass stacks and therefore can run in open-cycle mode. Each
of these units is currently installed and operational. The characteristics of these units in combined-cycle mode are given in
Table III. Limited data was available for these units in opencycle mode so each was given characteristics similar to a typical open-cycle gas turbine (OCGT) unit, as shown in Table III.
As CCGT 2 and CCGT 5 are comprised of two gas turbines
configuration), these units
connected to one steam turbine (
were modeled as having two identical open-cycle units available
for dispatch when the CCGT is operated in open-cycle mode.
CCGTs 2 and 3, located in Northern Ireland and CCGTs 1, 4,
and 5, located in the Republic of Ireland, contribute to the minimum units online constraint in their respective regions.
IV. RESULTS
A number of model runs were conducted to investigate the
potential for multi-mode operation of CCGT units. The Wilmar
model was run in deterministic mode as this is more representative of current scheduling practice. A year-long dispatch was
produced for each of the three wind power penetrations outlined
in Section III, when 1) multi-mode operation of CCGT units is
not allowed and 2) when multi-mode operation of CCGT units
is allowed.
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
TROY et al.: MULTI-MODE OPERATION OF COMBINED-CYCLE GAS TURBINES WITH INCREASING WIND PENETRATION
5
TABLE V
OCGT PRODUCTION (GWh) WITH INCREASING WIND PENETRATION
TABLE VI
DIFFERENCE IN OPEN-CYCLE PRODUCTION (GWh) FROM MULTI-MODE
UNITS WITH NO REPLACEMENT RESERVE TARGET ENFORCED
Fig. 2. Average production from a CCGT in open-cycle mode (line) and
average number of instances generators utilized open-cycle operation (grey
column), shown for various levels of installed wind capacity.
TABLE VII
AVERAGE HOURLY SURPLUS SPINNING RESERVE (MW) AVAILABLE
AND REPLACEMENT RESERVE TARGET (MW)
TABLE IV
AVERAGE UTILIZATION FACTORS WITH INCREASING WIND PENETRATION
A. Usage of the Multi-Mode Function
The average number of times a CCGT unit with multi-mode
capability was run in open-cycle mode and the average production from a CCGT in open-cycle mode over the year, at each
of the wind penetrations examined, is shown in Fig. 2. Despite
increasing wind penetration being correlated with an increased
demand for flexibility, be it fast starting or ramping, Fig. 2 shows
the multi-mode function is used less frequently as wind penetration on the system increases.
As more wind power, with an almost zero marginal cost, is
added to a system, the production from thermal plant is increasingly displaced and as such there is an increased likelihood of
generators operating at part-load. To illustrate, Table IV gives
the annual utilization factor (ratio of actual generation to maximum possible generation during hours of operation) averaged
for the coal, CCGT, and peat units on the system with 2000-,
4000-, and 6000-MW wind power. Therefore, as wind penetration increases, online part-loaded units are more often available
to ramp up their output to meet unexpected shortfalls in production, avoiding the need to switch on fast-starting units, such as
the CCGTs in open-cycle mode.
The trend seen in Fig. 2 is consistent with the production from
peaking plants as wind penetration increases. Table V shows the
drop in production from the most utilized OCGT unit, with increasing wind penetration when multi-mode operation is and is
not allowed. Reduced production from peaking plants due to increased wind penetration has also been observed in other wind
integration studies such as [17]; however, it is also likely that
systems with base-load units that have slower ramp rates than
those examined in this study will rely on fast-starting units (such
as CCGTs in open-cycle mode) more often as wind penetration
increases. (All units on the test system are assumed to be capable of ramping from minimum to maximum output in one
hour or less.) The average production from the CCGT units in
open-cycle mode, as seen in Fig. 2, is comparable with average
production levels from dedicated OCGT peaking plants on the
system when multi-mode operation of CCGTs is not enabled.
As wind penetration increases, so too will the demand for replacement reserve, due to the increased forecast error. The replacement reserve target can be met by fast-starting offline units
or from excess spinning reserve if available. If sufficient excess
spinning reserve is not available to meet the replacement reserve
target, the model must ensure a number of fast-starting units are
offline and available for operation to maintain a secure system.
Consequently, as a result of maintaining the replacement reserve
target, production from fast-start units (such as the multi-mode
units in open-cycle mode) is reduced. Additional simulations
were conducted for the various wind penetrations with no replacement reserve target, to investigate the extent that maintaining replacement reserve suppressed the multi-mode units
from running in open-cycle mode. For many systems, such as
the Irish system, this is more representative of current practice,
where no replacement reserve target formally exists. Table VI
shows the difference in the average open-cycle production from
multi-mode units that results when no replacement reserve targets are enforced.
As seen, in the absence of a target for replacement reserve,
open-cycle production from the multi-mode units is utilized
substantially more for the 2000-MW and 4000-MW wind
power scenarios. However, with 6000-MW wind power, due to
more frequent part-loading of units, there is more frequently
an excess of spinning reserve on the system, as well as offline
fast-starting units (as per Table V) which can contribute to the
replacement reserve target. Thus, with 6000-MW wind power,
the replacement reserve target has little effect on the open-cycle
operation of multi-mode units. Table VII shows the average
surplus spinning reserve available and the average replacement
reserve target per hour for each of the wind cases examined.
Fig. 3 shows the capacity factor for each CCGT in combined-cycle mode and its production over the year in open-cycle
mode for the 2000-MW wind power scenario. An inverse relationship is evident between the open-cycle production from a
CCGT and the capacity factor of the CCGT, which indicates
that usage of the multi-mode function is related to the amount
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
6
IEEE TRANSACTIONS ON POWER SYSTEMS
Fig. 3. Combined-cycle capacity factor (dashed line) and open-cycle production (solid line) for each CCGT with multi-mode capability for the 2000-MW
wind power system.
Fig. 4. Average production from OCGT peaking units in each wind power scenario, with multi-mode operation of CCGTs not allowed (light grey) and allowed (dark grey).
TABLE VIII
PERCENTAGE CHANGE IN TOTAL PRODUCTION WHEN MULTI-MODE IS
ENABLED, SHOWN FOR EACH WIND PENETRATION
TABLE IX
MAGNITUDE AND FREQUENCY OF REPLACEMENT RESERVE SHORTFALL,
SHOWN FOR VARIOUS LEVELS OF INSTALLED WIND
of time the CCGT is offline. The more often a CCGT is not in
operation but available for dispatch, the more opportunities it
has to run in open-cycle mode and this relationship would be
expected regardless of the plant portfolio.
The percentage change in total production (combined-cycle
plus open-cycle) that results when multi-mode operation of
CCGTs is enabled is shown in Table VIII, for each of the
wind penetrations examined. Multi-mode operation increased
production for CCGT5, the lowest merit CCGT which was
seen to utilize the function most frequently, across all the
wind penetrations examined. Total production from CCGT3
and CCGT4, which are mid-merit CCGTs, is reduced in all
cases but one. There is a risk (particularly for CCGTs that are
frequently the marginal unit on the system such as CCGT3
and CCGT4), when offering open-cycle operation, of being
dispatched from combined-cycle to open-cycle operation at
times of low net demand (demand minus wind generation) to
alleviate minimum load issues and then losing out to another
generator that can come online faster/cheaper, when the net
demand increases again. However, it is also likely that in a
market environment, generators would strategize when they
would offer this multi-mode capability to avoid losing out on
production. CCGT1, the highest merit CCGT, benefits from
increased production when multi-mode operation is enabled
on the system with 2000-MW and 4000-MW installed wind
power. This is due to increased exports and reduced production
from the other CCGTs, as opposed to increased production in
open-cycle mode.
contracts. Their open-cycle capacity (as seen in Table III) is
also larger than the capacity of the OCGTs (103.5 MW each)
and they benefit from avoided start-up costs when transitioning
from combined-cycle mode. Thus, when multi-mode operation
of CCGTs was enabled, production from OCGT peaking plant
tended to be substituted by production from the CCGTs in opencycle mode. Fig. 4, which shows the average production from
OCGTs for each wind penetration level when multi-mode operation of CCGTs is allowed and not allowed, illustrates this
point. Assuming open-cycle production from CCGTs is more
economic than production from OCGTs, as is the case here, it is
possible that by enabling multi-mode operation of CCGTs sufficient flexibility could be extracted from a systems portfolio of
plant to avoid building additional peaking units, or equally that
OCGT units would no longer be able to cover their costs and so
would be forced to retire from service. Both situations may then
lead to increased production from CCGTs in open-cycle mode.
Table IX shows the total shortfall in replacement reserve over
the year and the number of hours in which this occurred, for each
of the wind penetrations examined, when multi-mode operation
of CCGTs is and is not allowed. The additional fast-starting
generation available to the system when multi-mode operation
of CCGT units is allowed significantly reduces the shortfall in
replacement reserve. This contributes to a more secure system
by preventing capacity shortfalls when wind forecasts prove to
be overly optimistic and also indicates that, depending on the
market structure, the generators may benefit from an additional
revenue stream, via ancillary services payments for the replacement reserve provided.
In addition to enhanced system security, the additional flexibility available to the system when multi-mode operation of
CCGT units is allowed will also yield production costs savings.
Table X shows the total system operating cost savings achieved
by enabling multi-mode operation of CCGTs. The total system
B. Benefits Arising From Multi-Mode Operation
The efficiencies of the OCGT peaking units on the system are
comparable with the CCGT units in open-cycle mode. However,
the CCGT units running in open-cycle operation are assumed
to have a lower gas price, to reflect the advantage of long-term
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
TROY et al.: MULTI-MODE OPERATION OF COMBINED-CYCLE GAS TURBINES WITH INCREASING WIND PENETRATION
7
TABLE X
TOTAL SYSTEM COST SAVING (MC) RESULTING
FROM MULTI-MODE OPERATION OF CCGTS
cost is made up of fuel, carbon, and start-up costs for the Irish
and British system combined, as they are co-optimized. In this
case, these savings were achieved at no additional cost as each
of the CCGTs is currently capable of multi-mode operation.
A modest reduction in plant start-ups for multi-mode units
averaged
(in combined-cycle mode) was also observed (
over the three wind power scenarios), relative to the case when
multi-mode operation is not allowed, which would indicate benefits for the steam equipment via avoided wear-and-tear.
Fig. 5. Average production from a CCGT in open-cycle mode (line) and
average number of instances generators utilized open-cycle operation (grey
column), with interconnector scheduled day-ahead and intra-day on 2000-MW
wind system.
C. Sensitivity Studies
Usage of the multi-mode function is dependent on many
factors, particularly the amount of flexibility already present
in the system. A sensitivity study was conducted to examine
the usage of the multi-mode function when the system was less
flexible to meeting demand. This involved running the model
with 2000-MW wind power (as this level of wind generation
greatest usage of CCGTs in open-cycle mode) and power
exchange across the interconnector fixed day-ahead as opposed
to intra-day. Examining the usage of the multi-mode function when the interconnector is scheduled day-ahead versus
intra-day illustrates how a less flexible system will utilize this
flexible resource more frequently. Fig. 5 shows the average
production from a CCGT in open-cycle mode and the average
number of instances CCGTs utilized open-cycle operation,
with the interconnector scheduled day-ahead and intra-day on
the 2000-MW wind power system. The average production
from CCGTs in open-cycle mode on the system with day-ahead
scheduling of the interconnector is seen to be more than three
times greater than the system with intra-day scheduling of
the interconnector. By fixing the power exchange between
the Irish and British systems day-ahead, when there is greater
uncertainty in the expected wind generation and demand, the
system is forced to dispatch generators such as the multi-mode
CCGT units, as opposed to reschedule imports/exports, to
compensate for wind and load forecast errors. Likewise, systems with seasonal hydro restrictions may see greater usage
of multi-mode CCGT operation during these periods when the
operating flexibility of the system is reduced.
In addition, the type of wind and load forecasts employed by
a system will also determine the usage of the multi-mode function. Additional simulations were completed running the model
in stochastic and perfect foresight mode. These represent different means of including load and wind forecasts in the scheduling process; whereby stochastic optimization can be considered to represent a system employing ensemble forecasts, deterministic optimization is representative of a system utilizing a
single forecast, and the perfect forecast scenario is a hypothetical case where no forecast error exists. The robust solutions
obtained by stochastic optimization showed less deployment of
the multi-mode function compared with the deterministic results. The stochastic solution, optimized for several wind and
Fig. 6. Average production from CCGT in open-cycle mode (GWh), shown for
different methods of optimization with 2000-MW wind power.
load scenarios, typically has more units online to cover all scenarios and therefore is more prepared to deal with unforseen
shortfalls in wind generation or increases in demand without
the need for starting peaking plant. The capacity factors of the
CCGT units are also higher for the stochastic case compared
to the deterministic case, indicating that there was also less opportunity for these units to run in open-cycle mode when the
system is optimized stochastically. Running the Wilmar model
with perfect foresight of the system demand and wind profile
also reveals even less open-cycle operation from CCGTs as in
this case, with no forecast errors on the system (except forced
outages of generators), fast starting units are in less demand relative to the deterministically optimized solution. Fig. 6 compares
the average open-cycle operation from the multi-mode CCGTs,
on the system with 2000-MW wind power, when optimized with
perfect foresight, stochastically and deterministically. The average open-cycle production from a CCGT unit is seen to be
11% less on the stochastically optimized system and 35% less
on the system with perfect forecast compared to the deterministic case.
A sensitivity analysis was also conducted using a higher level
of demand on the system. In this case, the original demand profile from [22] with a 9.6-GW peak, discussed in Section III,
was run for each wind scenario. The average production from
a CCGT in open-cycle mode over the year is shown in Fig. 7
to be six to eight times greater on the 9.6-GW peak demand
system, where peaking capacity is in greater demand, compared
to the 7.55-GW peak demand system, at each of the wind power
penetrations examined. In addition to the increased demand resulting in increased open-cycle production from the multi-mode
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
8
IEEE TRANSACTIONS ON POWER SYSTEMS
TABLE XI
TOTAL SYSTEM COST, REPLACEMENT RESERVE SHORTFALL AND TOP-UP PAYMENT, SHOWN FOR VARIOUS MULTI-MODE CONFIGURATIONS
Fig. 7. Average production from a CCGT in open-cycle mode on the 7.55-GW
peak demand system (light grey) and the 9.6-GW peak demand system (dark
grey), shown for various levels of installed wind power.
CCGTs (as well as combined-cycle production), the other main
difference between the scenarios is the predominant direction of
power transfer on the interconnector. With 2000-MW installed
wind capacity, the Irish system is a net importer of power from
Britain, at both levels of demand examined. However, as more
wind power is installed on the 7.55-GW peak demand system,
the marginal electricity price is reduced sufficiently with respect
to the British system such that Ireland becomes a net exporter
of power. Although increasing wind power penetration on the
9.6-GW peak demand system also reduces the marginal price,
it is still a net importer with 6000-MW installed wind power.
Thus, on occasions when forecast wind is overestimated and the
system is in need of fast-starting plant, the 7.55-GW peak demand system, being a net exporter, can more frequently choose
to curtail exports or start up a unit to compensate. In contrast, the
9.6-GW peak demand system, being a net importer, more often
only has the option to turn on fast-starting plant. Hence, this implies that a system which tends to be a net exporter is inherently
more flexible, and has more options for dealing with variable
wind power than a system that is a net importer of power. In this
scenario with higher demand, each of the multi-mode CCGT
units experienced increased total production (combined-cycle
plus open-cycle) when multi-mode operation was allowed, suggesting that offering multi-mode capability may prove more
profitable on a system with a smaller capacity margin.
Given the low deployment of the multi-mode functionality
and the high capacity factor in combined-cycle mode for CCGT
1 and 2, as seen in Fig. 3, it would appear that there is insufficient incentive for all CCGTs capable of multi-mode operation
to offer this flexible capability. Thus, given that CCGTs 3, 4, and
5 have low capacity factors in combined-cycle mode, additional
simulations were conducted to investigate the benefits yielded
if these units alone, and if CCGT 5 alone, offered multi-mode
capability. Table XI shows the total system cost (for Ireland and
Britain) and the magnitude of the replacement reserve shortfall
over the year for these configurations (in addition to other configurations examined in the paper). Examining the shortfall in
the replacement reserve target for the different configurations
of the reduction in replacereveals that the majority
ment reserve shortfall due to multi-mode capability is attributable to CCGT 5, while CCGTs 1 and 2 are seen to have no
impact on the replacement reserve shortfall. Thus, CCGTs capable of open-cycle operation, which have very low output in
combined-cycle mode, have value in providing replacement reserve.
As seen in Table VIII, the multi-mode CCGTs may experience a reduction in total production as a result of offering
multi-mode capability to the market. This was also observed
to be the case for CCGTs 3 and 4, when only three units offered multi-mode operation. This indicates that a system seeking
to increase its flexibility via multi-mode operation of CCGTs,
possibly to facilitate integration of variable renewables, may
need to reward these units either through ancillary service payments or another market mechanism to restore their revenue
to original levels (i.e., when multi-mode operation was not allowed). The subsidy or “top-up payment” required to restore the
revenue of these units to their original level is estimated here
as the loss in total production multiplied by the average electricity price. The average “top-up payment” required is shown in
Table XI with the number of units requiring this payment shown
in parenthesis. However, it should be noted that this represents
the worst-case figure given that the multi-mode CCGT unit offered this capability in all time periods, rather than when it was
profitable for them to do so, as would likely be the case in reality.
V. CONCLUSIONS
This paper examines if allowing CCGT units to operate in
open-cycle mode, when this is technically feasible and cost
optimal, could deliver benefits to a system with a high wind
penetration or to the generators themselves. It is shown that the
This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination.
TROY et al.: MULTI-MODE OPERATION OF COMBINED-CYCLE GAS TURBINES WITH INCREASING WIND PENETRATION
extra fast-starting capacity available from multi-mode operation of CCGTs can reduce the replacement reserve shortfall,
indicating an opportunity for increasing system reliability.
Low-merit CCGTs will utilize the multi-mode function more
as they are frequently offline and available for dispatch, while
the increased competition among generators, typical at higher
levels of wind generation, results in multi-mode operation of
CCGTs being utilized less frequently. Peaking production from
CCGTs in open-cycle mode can displace peaking production
from OCGTs, potentially reducing the need for such units to
be built. Sensitivity studies reveal that usage of the multi-mode
function is dependent on the level of flexibility inherent in
a system. Optimizing the system stochastically or allowing
intra-day trading on interconnectors reduces the need for flexibility to be extracted from generators and consequently results
in less frequent deployment of the multi-mode function.
ACKNOWLEDGMENT
The authors would like to thank A. Mahon and A. Barnes of
ESB for their helpful contributions.
REFERENCES
[1] R. Kehlhoffer, Combined-Cycle Gas & Steam Turbine Power Plants,
2nd ed. Tulsa, OK: PennWell, 1999.
[2] W. J. Watson, “The success of the combined cycle gas turbine,” in
Proc. IEEE Conf. Opportunities and Advances in International Electric Power Generation, 1996, pp. 87–92.
[3] U. C. Colpier and D. Cornland, “The economics of the combined cycle
gas turbine—An experience curve analysis,” Energy Policy, vol. 30, no.
4, pp. 309–316, 2002.
[4] A. Shibli and F. Starr, “Some aspects of plant and research experience
in the use of new high strength martensitic steel P91,” Int. J. Pressure
Vessels and Piping, vol. 84, pp. 114–122, 2007.
[5] F. Starr, “Background to the design of HRSG systems and implications
for CCGT plant cycling,” Oper. Mainten. Mater. Issues, vol. 2, no. 1,
Apr. 2003.
[6] R. Anderson and H. van Ballegooyen, “Steam turbine bypass systems,”
Combined Cycle J., 2003.
[7] Winning With European Wind, European Wind Energy Association,
2009. [Online]. Available: http://www.ewea.org.
[8] H. Chandler, Empowering Variable Renewables, Options for Flexible
Electricity Systems. Paris, France: International Energy Agency,
2008.
[9] F. Van Hulle and P. Gardner, Wind Energy—The Facts, Part 2
Grid Integration, 2008. [Online]. Available: http://www.wind-energy-the-facts.org/.
[10] P. Brown, J. Lopes, and M. Matos, “Optimization of pumped storage
capacity in an isolated power system with large renewable penetration,”
IEEE Trans. Power Syst., vol. 23, no. 2, pp. 523–531, May 2008.
[11] V. Hamidi and F. Robinson, “Responsive demand in networks with
high penetration of wind power,” in Proc. IEEE/PES Transmission and
Distribution Conf. Expo., 2008.
[12] L. Göransson, “Wind power in thermal power systems—Large-scale
integration,” Licentiate thesis, Dept. of Energy and Environment,
Chalmers Univ. Technology, Goteburg, Sweeden, 2008.
[13] B. Lu and M. Shahidehpour, “Short-term scheduling of combined-cycle units,” IEEE Trans. Power Syst., vol. 19, no. 3, pp.
1616–1625, Aug. 2004.
[14] B. Blevins, Combined-Cycle Unit Modeling in the Nodal Design.
Taylor, TX: ERCOT, 2007.
[15] Multi-Stage Generating (MSG) Unit Modeling, CAISO, 2010. [Online]. Available: http://www.caiso.com/2078/2078908392d0.html.
9
[16] N. Troy, E. Denny, and M. O’Malley, “Base-load cycling on a system
with significant wind penetration,” IEEE Trans. Power Syst., vol. 25,
no. 2, pp. 1088–1097, May 2010.
[17] Growing Wind—Final Report of the NYISO Wind Integration Study,
NYISO, 2010. [Online]. Available: http://www.nyiso.com.
[18] Integration of Renewable Resources—Operational Requirements and
Generation Fleet Capability at 20% RPS, California ISO, 2010. [Online]. Available: http://www.caiso.com/2804/2804d036401f0ex.html.
[19] Western Wind and Solar Integration Study, National Renewable Energy
Laboratory, 2010. [Online]. Available: http://www.nrel.gov/wind/systemsintegration/wwsis.html.
[20] L. Göransson and F. Johnsson, “Dispatch modeling of a regional power
generation system—Integrating wind power,” Renew. Energy, vol. 34,
no. 4, pp. 1040–1049, 2009.
[21] N. Troy and M. O’Malley, “Multi-mode operation of combined cycle
gas turbines with increasing wind penetration,” in Proc. IEEE Power
and Energy Soc. General Meeting, 2010.
[22] Wind Variability Management Studies, All Island Renewable Grid
Study—Workstream 2B, 2008. [Online]. Available: http://www.dcenr.
gov.ie.
[23] P. Meibom, R. Barth, B. Hasche, H. Brand, and M. O’Malley, “Stochastic optimization model to study the operational impacts of high
wind penetrations in Ireland,” IEEE Trans. Power Syst., vol. 26, no. 3,
pp. 1367–1379, Aug. 2011.
[24] A. Tuohy, P. Meibom, E. Denny, and M. O’Malley, “Unit commitment
for systems with significant wind penetration,” IEEE Trans. Power
Syst., vol. 24, no. 2, pp. 592–601, May 2009.
[25] P. Meibom, WILMAR—Wind Power Integration in Liberalised
Electricity Markets, 2006. [Online]. Available: http://www.wilmar.
risoe.dk/Results.htm.
[26] L. Söder, “Simulation of wind speed forecast errors for operation planning of multiarea power systems,” in Proc. Int. Conf. Probabilistic
Methods Applied to Power Systems, 2004.
[27] J. Dupacova, N. Growe-Kuska, and W. Romisch, “Scenario reduction
in stochastic programming: An approach using probability metrics,”
Math. Program., vol. 95, no. 3, pp. 493–511, 2003.
[28] R. Doherty and M. O’Malley, “A new approach to quantify reserve demand in systems with significant installed wind capacity,” IEEE Trans.
Power Syst., vol. 20, no. 2, pp. 587–595, May 2005.
[29] J. M. Arroyo and A. J. Conejo, “Modeling of start-up and shut-down
power trajectories of thermal units,” IEEE Trans. Power Syst., vol. 19,
no. 3, pp. 1562–1568, Aug. 2004.
[30] Redpoint Validated Forecast Model and PLEXOS Validation Report
2010, Commission for Energy Regulation, 2010. [Online]. Available:
http://www.allislandproject.org.
[31] Generation Adequacy Report 2010–2016, EirGrid, 2009. [Online].
Available: http://www.eirgrid.com.
Niamh Troy (M’11) received the B.Sc. degree in applied physics from the University of Limerick, Limerick, Ireland. She is currently pursuing the Ph.D. degree at the Electricity Research Centre in University College Dublin, Dublin,
Ireland.
Damian Flynn (SM’11) is a senior lecturer in power engineering at University
College Dublin, Dublin, Ireland. His research interests involve an investigation
of the effects of embedded generation sources, especially renewables, on the
operation of power systems.
Mark O’Malley (F’07) received the B.E. and Ph.D. degrees from University
College Dublin, Dublin, Ireland, in 1983 and 1987, respectively.
He is a Professor of electrical engineering in University College Dublin and
is director of the Electricity Research Centre with research interests in grid integration of renewables.
Prof. O’Malley is a member of the Royal Irish Academy.
1
Unit Commitment with Dynamic Cycling Costs
Niamh Troy, Student Member, IEEE, Damian Flynn, Senior Member, IEEE, Michael Milligan, Senior
Member, IEEE, and Mark O’Malley, Fellow, IEEE
Abstract—Increased competition in the electricity sector and
the integration of variable renewable energy sources is resulting
in more frequent cycling of thermal plant. Thus, the wearand-tear to generator components and the related costs are a
growing concern for plant owners and system operators alike.
This paper presents a formulation that can be implemented
in a MIP dispatch model to dynamically model cycling costs
based on unit operation. When implemented for a test system
the results show that dynamically modeling cycling costs reduces
cycling operation and tends to change the merit order over time.
This leads to the burden of cycling operation being more evenly
distributed over the plant portfolio and a reduces the total system
costs relative to the case when cycling costs are not modeled.
Index Terms—Thermal Power Generation, power system modeling
N OMENCLATURE
Indices/Sets
t, T
g, G
i, I
j, J
l, L
Time step, set of time steps
Units, set of units
Interval of cycling cost function, set of intervals of cycling cost function
Level of ramp, set of all ramp levels
Segment of the piecewise linearization of the
variable cost function, set of all segments of
the piecewise, linearization of the variable
cost function
Constants
ag , bg , cg
costSg
ThSg (i)
costSg (i)
Rg
Rg (j)
Coefficients of the quadratic production cost
function for unit g
Cycling cost increment for each additional
start-up for unit g
ith threshold corresponding to cumulative
start-ups for unit g
Cycling cost increment for each additional
start-up, while NSg (t,i) < ThSg (i+1) for unit g
production change (MW) over time period t
deemed damaging for unit g
j th production change (MW) over time period t deemed damaging for unit g
N. Troy ([email protected]), D. Flynn ([email protected]) and M.
O’Malley ([email protected]) are with the School of Electrical, Electronic and Communications Engineering, University College Dublin, Ireland.
Michael Milligan is with the National Renewable Energy Laboratory, Golden,
CO 8041 USA (email: [email protected]).
This work was conducted in the Electricity Research Centre, University
College Dublin, Ireland, which is supported by the Commission for Energy
Regulation, Bord Gais Energy, Bord na Mona Energy, Cylon Controls, EirGrid, the Electric Power Research Institute (EPRI), ESB Energy International,
ESB Energy Solutions, ESB Networks, Gaelectric, SSE Renewables, and
Viridian Power & Energy. This publication has emanated from research
conducted with the financial support of Science Foundation Ireland under
Grant Number 06/CP/E005.
costR
g
ThR
g (i)
costR
g (i)
Ig
j̄g
P̄g
Pg
Ag
NLg
Flg
Tlg
U Tg
DTg
T̄
Tgcold
ccg
hcg
hup
hdown
M
α, β, γ
Cycling cost increment for unit g for each
additional ramp > Rg
ith threshold corresponding to cumulative
ramps for unit g
Cycling cost increment for unit g for each
R
additional ramp, while NR
g (t,i) < Thg (i+1)
Total number of intervals in cycling cost
function for unit g
Number of ramp levels defined for unit g
Maximum capacity for unit g
Minimum capacity for unit g
Fixed cost for unit g ($/h)
Number of segments in piecewise linearization of the variable cost function for unit g
Slope of segment l of the variable cost
function for unit g
Upper limit of block l of the piecewise linear
production cost function of unit j (MW)
Minimum up time for unit g
Minimum down time for unit g
Number of hours in the planning period
Number of hours unit g must be offline,
beyond its minimum downtime, before it is
considered to be in a cold state
Cold start-up cost for unit g
Hot start-up cost for unit g
Number of hours unit g has been online for
at start of planning period (h)
Number of hours unit g has been offline for
at start of planning period (h)
Large number
Scaling factors
Binary Variables
sg (t)
zg (t)
vg (t)
stepSg (t, i)
rg (t)
rg (t, j)
stepR
g (t, i)
equal to 1 when a unit starts up at time t
equal to 1 when a unit shuts down at time t
equal to 1 when a unit is online at time t
equal to 1 when NS (t,1) ≥ ThS (i) at time t
equal to 1 when a unit undergoes ramp >
Rg between time t − 1 and t
equal to 1 when a unit undergoes ramp >
Rg (j) between time t − 1 and t
equal to 1 when NS (t,1) ≥ ThR (i) at time t
Positive Variables
NSg (t)
NSg (t,i)
Cumulative start-ups for unit g
Cumulative start-ups for unit g beyond
threshold ThSg (i)
2
CSg (t)
NR
g (t)
NR
g (t,i)
CR
g (t)
cpg (t)
csg (t)
pg (t)
D(t)
δl (g,t)
Total cycling cost attributed to start-ups for
unit g
Cumulative ramps > Rg for unit g
Cumulative ramps > Rg beyond threshold
ThR (i) for unit g
Total cycling cost attributed to ramping for
unit g
Production cost for unit g at time t
Start-up fuel cost for unit g at time t
Output (MW) for unit g at time t
System demand (MW) at time t
Power produced in block l of the piecewise
linear production cost function of unit g at
time t (MW)
I. I NTRODUCTION
I
NCREASED competition in the electricity generation sector coupled with the large-scale deployment of variable
renewable energy sources, particularly wind power, has led
to increased plant cycling in power systems worldwide [1],
[2]. Cycling may be defined as frequent start-ups or ramping
of units. Some generation types (such as hydro or even opencycle gas turbines) are more suited to frequent cycling, but
for others, particularly units designed for base-load operation,
cycling can accrue large levels of damage within the plant’s
components leading to increased maintenance requirements
and forced outage rates. Thermal shock, metal fatigue, corrosion, erosion and heat decay are common damage mechanisms
that result from cycling operation [3]. The wear-and-tear which
arises incurs increased maintenance costs for generators, but
in addition to this, loss of revenue due to more frequent and
longer outages, increased fuel costs due to more frequent startups and reduced plant efficiency, as well as additional capital
costs due to component replacement can also be expected.
Studies indicate that the magnitude of these cycling related
costs are high, but accurately quantifying them is challenging
[4], [5]. The level of wear-and-tear for a unit that undergoes
cycling operation will be dependent on many factors including
the operating history of the plant (i.e. how much creep damage
it has accumulated), and the engineering design of the plant.
It is also typical to see a time lag of several years from when
cycling occurs to when the damage manifests itself [6].
Research related to the cost of generation cycling has been
undertaken by EPRI and Intertek Aptech and the approaches
employed can be categorized as top-down (statistical analysis)
or bottom-up (component modeling). EPRI carried out a
top-down study utilizing multivariate regression models to
analyze the operating regimes of 158 units from NERC (North
American Electric Reliability Corporation) GADS (Generating
Availability Data System) and CEMS (Continuous Emission
Monitoring) data in an attempt to identify patterns relating
plant operation to capital expenditure. However, the inconsistency in accounting practices between the units complicated
the modeling and no correlation was found [7], [8]. Intertek
Aptech employ a combination of top-down models based on
historical operations, forced outage and cost data as well as
bottom-up methods which calculate operational stresses and
the life expenditure of critical components to determine cycling costs for individual generating units [4]. Intertek Aptech
have analyzed cycling costs for over 300 generating units and
found that the cost of cycling a conventional fossil-fired power
plant can be as much as $2,500-500,000 per start/stop cycle
depending on unit age, operating history and design features,
and are often grossly underestimated by utilities [4], [6].
Not considering these costs, however, will result in an uneconomic plant dispatch, yet markets currently do not include
specific cycling cost components in their bidding mechanisms,
or at best cycling costs are bundled into a generator’s startup or operating costs. Depending on the operating regime of
a plant, these cycling related costs can accumulate rapidly
and are therefore dissimilar to plant characteristics such as
heat rate, which typically vary over a much longer time-scale.
Therefore, to examine the impact of these costs accurately,
they should be modeled in a dynamic manner such that they
accumulate within the optimization process based on how
the unit is being operated and thereby can influence dispatch
decisions.
This paper presents a novel formulation to dynamically
model these cycling costs, which can be integrated into a MIP
(mixed integer programming) unit commitment and economic
dispatch model. This facilitates more accurate modeling of
these costs and examination of how they accumulate in line
with the operating regime of the plant. The formulation defines
a cycling cost which increments with each additional plant
start-up or ramp with the resulting cost function being linear,
piecewise linear or step-shaped. A case study is included to
determine how implementing dynamic cycling costs for a test
system over a period of one year will affect the resulting
dispatch, relative to a scenario where cycling costs are not
considered. This new approach to modeling cycling costs is
particularly suitable for long-term planning studies where it
can be used to reflect the ageing effect on a plant over time.
It may also have applications for real-world market models
where it can discourage the same unit from being repeatedly
dispatched to cycle by incurring an incremental cost to reflect
the wear-and-tear to that unit, which can consequently alter
its position in the merit order.
The paper is organized as follows: Section II details the
formulation of dynamic cycling costs, Section III describes a
unit commitment model and economic dispatch model used
to implement the dynamic cycling cost formulation and also
describes the test system, Section IV details the results of the
case study and Section V summarizes the findings.
II. F ORMULATION OF DYNAMIC C YCLING C OSTS
A detailed formulation for implementing dynamic cycling
costs which increase in line with unit operation is presented.
Cycling costs are subdivided into costs for (A) start-ups and
(B) ramps. The formulation utilizes three main steps: (i) a
binary variable is set to indicate that damaging operation has
occurred at time step t, (ii) a counter tracks how much of that
type of operation has occurred up to that point, and (iii) an
incrementing cycling cost is incurred at that time step. Linear,
piecewise linear and step-shaped cost functions for both starts
and ramps are detailed here.
3
A. Cycling costs related to start-ups
Linear: Constraints 1-3 allow a dynamic, linearly incrementing cost for wear-and-tear related to start-ups to be modeled. Based on the online binary variable, vg (t), constraint 1
sets the start-up, sg (t), and shut-down, zg (t), binary variables
equal to 1 appropriately, when unit g is started up or shut
down at time t. Constraint 2 increments a counter, NgS (t),
to track how many start-ups have been performed by that
unit. Constraint 3 determines the start-up related cycling cost,
CgS (t), with the final term ensuring that a cost is only incurred
when the decision is made to start the unit at time t (i.e. sg (t)
= 1). Figure 1 provides an example of this linearly increasing
cost function, where the cycling cost increment costSg is set
equal to 100. (It is also possible to initialize the counter NgS (t)
with the number of starts that have been carried out previously
if this is known).
sg (t) − zg (t) = vg (t) − vg (t − 1), ∀ t ∈ T, ∀ g ∈ G (1)
NgS (t) ≥ NgS (t − 1) + sg (t), ∀ t ∈ T, ∀ g ∈ G
(2)
¡
¢
CgS (t) ≥ NgS (t).costSg − M. 1 − sg (t) , ∀ t ∈ T, ∀ g ∈ G
(3)
CgS (t) ≥
¶
Ig µ
X
¡
¢
NgS (t, i). costSg (i) − costSg (i − 1)
i
Fig. 2.
¡
¢
− 1 − sg (t) .M, ∀ t ∈ T, ∀ g ∈ G
(5)
Piecewise linearly increasing start-up related cycling cost
Step Function: Alternatively, if less information is known
regarding the shape of the cost function an appropriate simplification may be to define a step function, where CgS (t) does
not increment until T hSg (i) is reached. Again, it is required
that T hSg (1) is equal to 1. NgS (t, i) is determined by constraint
6 and in this case can be greater than or less than 0 (it was
previously defined as a positive variable only). Constraint 7
sets the binary variable stepSg (t, i) equal to 1 when NgS (t, i)
has exceeded T hSg (i), and constraint 8 determines the cycling
cost. Figure 3 provides an example of this incrementing, stepshaped cost function, where costSg (t, 1) is set equal to 100,
costSg (t, 2) is set equal to 150 and T hSg (2) equals 4.
NgS (t, i) =
µ
¶
NgS (t − 1, 1) + sg (t) + 1 − T hSg (i),
(6)
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
Fig. 1.
NgS (t, i) − stepSg (t, i).M ≤ 0,
Linearly increasing start-up related cycling cost
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
Piecewise Linear: By defining i thresholds, T hSg (i), each
corresponding to a cumulative number of plant start-ups, at
which point the start-up related cycling cost, CgS (t), will
increase by incremental cost costSg (i) for each additional
start, a piecewise linear incremental cost function can be
modeled. Constraint 4 is a modified form of constraint 2
which counts the cumulative number of start-ups. For i >
1, the start-up counter, NgS (t, i), will not have a positive
value until NgS (t, 1) has reached T hSg (i). T hSg (1) must equal
1. Constraint 5 determines the total cycling cost. Figure 2
provides an example of a piecewise linearly increasing cost
function, where costSg (1) is set equal to 100, costSg (2) is set
equal to 150 and T hSg (2) equals 4.
NgS (t, i) ≥
µ
¶
NgS (t − 1, 1) + sg (t) + 1 − T hSg (i),
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
(4)
CS (t) ≥ costSg (i).stepSg (t, i) −
¡
¢
1 − sg (t) .M,
∀ t ∈ T, ∀ g ∈ G, ∀ i ≤ Ig
Fig. 3.
Step increasing start-up related cycling cost
(7)
(8)
4
TABLE I
A NALOGOUS T ERMS
Hot and Cold Starts: Either the linear, piecewise linear or
step formulations can be extended to differentiate between hot
and cold start-ups for units. Constraint 9 will set the binary
variable scold
(t) equal to 1 only if unit g is started at time
g
t, having been offline for Tgcold plus its minimum downtime,
DTg . In constraints 2, 4 and 6 ‘+ sg (t)’ is replaced with ‘+
sg (t) + α.scold
(t)’. A scaling factor, α, is chosen based on
g
the ratio of cycling damage caused by a hot start relative to a
cold start, and thus normalizes NgS (t, i) to count in terms of
hot starts.
X
Piecewise
Linear &
Step
vg (t − n), ∀ t ∈ T, ∀ g ∈ G
n=1
(9)
B. Cycling costs related to ramping
1) Define one ramp level: The simplest form of incurring
cycling costs related to ramping duty is to define a change
in output, Rg , between consecutive time periods, greater
than which, damaging transients will occur within unit g.
Constraints 10 and 11 ensure that the binary variable r(t)
is set to 1 when a change in output exceeding Rg occurs.
To avoid double counting cycling costs when large ramps are
experienced in the start-up or shut-down process, the final term
ensures that the constraints are non-binding when the unit
is in the start-up or shut-down process. If the ramp-related
cycling costs are likely to exceed the start-up or shut-down
cost, constraint 12 is needed to prevent the model setting s(t)
and z(t) both equal to 1 in constraint 1, in order to make
constraints 10 and 11 non-binding.
¡
¢
pg (t) − pg (t − 1) − M.rg (t) ≤ Rg + M.sg (t),
∀ t ∈ T, ∀ g ∈ G
¡
¢
pg (t − 1) − pg (t) − M.rg (t) ≤ Rg + M.zg (t),
∀ t ∈ T, ∀ g ∈ G
sg (t) + zg (t) ≤ 1, ∀ t ∈ T, ∀ g ∈ G
(10)
(12)
Utilizing the binary variable, rg (t), a counter NgR (t) is
defined, as before, to incur an incrementing, ramp-related
cycling cost, CgR (t). Using the formulation from Section II.A,
the ramp-related cycling cost function may be linear, piecewise
linear or step-shaped. Constraints 2 and 3 are replaced with
the analogous ramp terms shown in Table I to implement a
linearly incrementing cost. Constraints 4 and 5, or 6 to 8, are
replaced with the analogous ramp terms as shown in Table I to
define a piecewise linear, or step shaped, incrementing ramp
related cycling cost respectively.
2) Define multiple ramp levels: The previous formulation,
where one level Rg is set to define a ramp, can be expanded
to incur a dynamic ramp-related cycling cost, for j ramps of
different magnitudes, Rg (j). Constraint 13 ensures that for a
ramp less than Rg (1), the binary variable rg (t, j) will equal
zero for all j. A ramp greater than Rg (1), but less than Rg (2),
will set rg (t, 1) equal to one, and so forth. The final term
rg (t)
costR
g (t)
NR
g (t)
CR
g (t)
sg (t)
costS
g (t,i)
NS
g (t,i)
ThS
g (t,i)
CS
g (t)
stepS
g (t)
rg (t)
costR
g (t,i)
NR
g (t,i)
ThR
g (t,i)
CR
g (t)
stepR
g (t)
ensures that the constraint is non-binding when the unit is
starting up. A¡ corresponding constraint
is needed for down
¢
ramps,
where
p
(t)
−
p
(t
−
1)
in
constraint
13 is replaced
g
g
¡
¢
with pg (t−1)−pg (t) and M.sg (t) is replaced with M.zg (t).
Constraint 14 ensures that the binary variable, rg (t, j), which
indicates that a ramp ≥ Rg (j) has occurred, can only have a
value of 1 for one ramp level j, at any given time. As before,
constraint 12 is required to prevent sg (t) and zg (t) both being
set to 1, to make constraint 13 and its corresponding down
ramping constraint non-binding.
j
X
¢
¡
¢
¡
rg (t, j)
pg (t) − pg (t − 1) < Rg (1). 1 −
j=1
+ Rg (2).rg (t, 1) + ... + Rg (j).rg (t, j − 1)
+ P¯g .rg (t, j) + M.sg (t),
where Rg (1) < Rg (2) < Rg (j)... < P¯g ,
∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
j
X
(11)
Ramps
sg (t)
costS
g (t)
NS
g (t)
CS
g (t)
Linear
Tgcold +DTg
(t) ≥ vg (t) −
scold
g
Starts
rg (t, j) ≤ 1, ∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
(13)
(14)
j=1
As with hot and cold starts, scaling factors are used to
normalize NgR (t) to count in terms of one ramp level, as shown
in constraint 15, where r(t, j) is expressed in terms of r(t, 1).
Constraint 16 determines the total ramp-related cycling cost,
shown here with a constant cost increment, costR
g , with the
final term ensuring that the cost is only incurred in a time
period when a ramp (> Rg (1)) occurs.
NgR (t) = NgR (t − 1) + rg (t, 1) + β.rg (t, 2)
+.... + γ.rg (t, j), ∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
CgR (t)
≥
NgR (t).costR
g
j
X
¡
¢
− 1−
rg (t, j) .M
j=1
(15)
(16)
∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
To combine this formulation of j ramp levels with i cost
thresholds (i.e piecewise linear) constraints 15 and 16 are
replaced by constraints 17 and 18, such that once NgR (t, i)
R
R
reaches T hR
g (i), Cg (t, i) will begin incrementing by costg (i).
5
NgR (t, i)
=
¡
NgR (t
− 1, 1) + rg (t, 1) + β.rg (t, 2)
¢
+.... + γ.rg (t, j) + 1 − T hR
g (i) (17)
∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g , ∀ i ≤ Ig
CgR (t) ≥
−
¶
Ig µ
X
¡
¢
R
NgR (t, i). costR
(i)
−
cost
(i
−
1)
g
g
i
j
X
Start-up costs which were dependent on the period of time
the unit had been offline were modeled as follows:
¡
¢
csg (t) ≥ vg (t) − vg (t − 1) .hcg ∀ t ∈ T, ∀ g ∈ G (29)
csg (t) ≥ vg (t) −
X
¢
vg (t − n) .ccg ,
n=1
(18)
To include a step-shaped ramp related cycling cost function,
constraints 6-8 are replaced with the analogous terms for
ramping from Table 1.
Minimum up time constraints were formulated by constraints 31, 32 and 33. Equation 31 is only included if the
number of hours a unit must remain online to satisfy its
minimum up time, Bg , is greater than or equal to 1.
t≤Bg
X ¡
III. D ISPATCH M ODEL AND T EST S YSTEM
¢
1 − vg (t) = 0, ∀ g ∈ G
cpg (t) + csg (t) + CgS (t) + CgR (t) (19)
t∈T g∈G
t+U Tg −1
X
vg (n) ≥ U Tg .sg (t), ∀ g ∈ G
n=t
pg (t) = D(t), ∀ t ∈ T
(20)
pg (t) ≤ P̄g .vg (t), ∀ t ∈ T
(21)
pg (t) ≥ P g .vg (t), ∀ t ∈ T
(22)
(32)
∀ t = Bg + 1...T̄ − U Tg + 1
T̄
X
¡
¢
vg (n) − sg (t) ≥ 0, ∀ g ∈ G
n=t
subject to
X
(31)
t
To examine how cycling costs, modeled dynamically, will
impact plant dispatch the new formulation was implemented
in a conventional MIP unit commitment model based on [9],
[10]. The unit commitment problem was formulated as
XX
(30)
∀ t ∈ T, ∀ g ∈ G
rg (t, j).M, ∀ t ∈ T, ∀ g ∈ G, ∀ j ≤ j̄g
j=1
M inimize
Tgcold +DTg
¡
(33)
∀ t = T̄ − U T + 2...T̄
¢
where Bg = max 0, vg (T)U Tg -hup
g +vg (T)
¡
g∈G
As per [9], a piecewise linear approximation of a quadratic
production cost function for each unit was adopted, as represented by:
Minimum down time constraints were formulated using
constraints 34, 35 and 36. Equation 31 is only included if
Lg ≥ 1.
t≤Lg
X¡
cpg (t)
= Ag vg (t) +
t+DTg −1
X
Flg δl g(t), ∀ t ∈ T, ∀ g ∈ G (23)
δl g(t) + P g vg (t), ∀ t ∈ T, ∀ g ∈ G
(24)
l=1
T̄
X
¡
1 − vg (n) − zg (t)
¢
≥ 0, ∀ g ∈ G
n=t
δ1 (g, t) ≤ T1g − P g , ∀ t ∈ T, ∀ g ∈ G
(25)
δl (g, t) ≤ Tlg − Tl−1g , ∀ t ∈ T, ∀ g ∈ G, ∀ l = 2..N Lg − 1
(26)
δN L (g, t) ≤ P̄g − TN Lg −1 − Tl−1g , ∀ t ∈ T, ∀ g ∈ G (27)
δl (g, t) ≥ 0, ∀ t ∈ T, ∀ g ∈ G, ∀ l = 1..N Lg
where Ag = ag + bg P g +
(35)
∀ t = Lg + 1...T̄ − DTg + 1
N Lg
X
vg (n) ≥ DTg .zg (t), ∀ g ∈ G
n=t
l=1
pg (t) =
(34)
t
N Lg
X
¢
vg (t) = 0, ∀ g ∈ G
cg P 2g .
(28)
¡
(36)
∀ t = T̄ − DT + 2...T̄
¢
where Lg = max 0, (1 − vg (T)).DTg − hdown
+(1 − vg (T))
g
The formulation was applied to the 10 unit test system used
in [9], [11], which was duplicated to give a 20 unit system, thus
facilitating a larger case study. The peak demand (1500 MW)
was doubled (3000 MW) and a historical year-long hourly
demand profile for the Irish system was scaled to produce a
demand profile with a 3000 MW peak. The model was run
for the test year, optimizing each day at an hourly resolution.
Generator cycling costs are difficult to determine and largely
uncertain, as discussed in Section I. The figures used here,
6
shown in Table II, to implement dynamic cycling costs for the
test system, are conservatively based on those in [12] and are
intended to illustrate how dynamic cycling costs could impact
system operation, rather than provide an accurate estimate of
such costs. Piecewise linear costs for starts and ramps were implemented with the incremental cost (costSg (i) or costR
g (i)) increasing by 10% and 20% when the start counter (NgS (t, 1)), or
ramp counter (NgR (t, 1)), exceeded 100 (T hSg (2) or T hR
g (2))
and 200 (T hSg (3) or T hR
g (3)) respectively. The scaling factor,
α, was chosen to be 2, i.e. each cold start incremented
NgS (t, 1) by 2 (while a hot start incremented NgS (t, 1) by 1).
Two ramp levels, Rg (1) and Rg (2) corresponding to 20% and
40% of the difference between maximum and minimum output
for a unit, were modeled. Scaling factors were chosen such that
ramps greater than Rg (1) or Rg (2) incremented NgR (t, 1) by
1 or 2 respectively.
related to plant start-ups was also found to have the knock
on effect of increasing generator ramping. Over the year a
22% increase in ramping (NgR (t, 1)) was observed relative to
the case when no cycling costs were modeled as generators
were more frequently ramped down to minimum output, rather
than shut-down, in an effort to avoid incurring cycling costs
for starting up.
TABLE III
I MPACT OF DYNAMIC CYCLING COSTS FOR START- UPS ON TOTAL ANNUAL
STARTS
Units
Base-load (Units 1-4)
Mid-merit (Units 5-10)
Peaking (Units 11-20)
Total
No cycling
costs modeled
Cycling cost for
starts modeled
34
1372
577
1983
12
1005
838
1855
TABLE II
I NCREMENTAL CYCLING COSTS $, ( I =1)
Units
costS
g (i)
costR
g (i)
1-4
5-10
11-20
300
60
30
15
3
1.5
IV. R ESULTS
This section examines how plant dispatches for the test
system are affected over one year when (i) a cycling cost
related to start-ups is implemented, (ii) a cycling cost related
to ramping is implemented, and (iii) cycling costs related to
start-ups and ramping are implemented simultaneously.
TABLE IV
I MPACT OF DYNAMIC CYCLING COSTS FOR START- UPS ON AVERAGE
PLANT CAPACITY FACTORS (%)
Units
Base-load (Units 1-4)
Mid-merit (Units 5-10)
Peaking (Units 11-20)
No cycling
costs modeled
Cycling cost for
starts modeled
92.59
27.82
0.85
92.73
25.42
2.23
A. Start-up Related Cycling Costs Results
Implementing a dynamic cycling cost for plant start-ups, as
shown in Table II, was seen to result in an overall reduction in
plant start-ups. This is seen in Table III, which reveals reducing
starts for base-load and mid-merit units. For base-load units,
the reduction in starts was correlated with increased production
as, having the largest incremental cycling costs, these units
avoided shut-downs and their online hours increased. This is
evident through the average capacity factor shown in Table
IV. Mid-merit units, however, which also had reduced starts,
saw reduced production indicating that they were utilized less
often. As these units were started up and shut down, and
subsequently incurred cycling costs, it became more economical after some point to dispatch peaking units. Thus, starts
and production increased for peaking units when a dynamic
cycling cost for start-ups was modeled. Figure 4 illustrates
the cumulative start-ups for the mid-merit and peaking units
over the year when (i) cycling costs were modeled and (ii)
when cycling costs were not modeled. Starts are seen to
accumulate rapidly between 0 and 2000 hours and for hours
greater than 7000, as these are the winter months and thus
have higher demand, requiring more plant start-ups. Beyond
1000 hours the cycling costs which are accumulated by midmerit plant begin to have an effect on their position in the
merit order and consequently peaking plant are seen to be
dispatched more frequently. Modeling dynamic cycling costs
Fig. 4. Cumulative plant start-ups over the year, shown when dynamic cycling
costs for starts were (i) modeled and (ii) not modeled
Units within the same class, i.e. base-load, mid-merit or
peaking, were also seen to converge to a similar number of
annual start-ups, as indicated by the reduced standard deviation
of annual start-ups seen in Table V. This indicates that once
a unit has been cycled and its cycling cost is incremented,
the next time a unit needs to be cycled the costs will have
now changed such that a different unit (most likely the next
in the merit order) may be scheduled. This leads to the burden
of cycling operation being more evenly distributed across the
units. Over a long horizon, i.e. several years, this effect can
lead to a shift in the merit order, a trend which can be seen
in Figure 4.
To facilitate a sensitivity analysis, multiples of the initial
incremental cycling costs, costSg (1), shown in Table II, were
also examined. As the incremental cost was increased the
reduction in start-stop cycling that is achieved by modeling
7
TABLE V
I MPACT OF DYNAMIC CYCLING COSTS FOR START- UPS ON TOTAL ANNUAL
STARTS
Units
Base-load (Units 1-4)
Mid-merit (Units 5-10)
Peaking (Units 11-20)
No cycling
cost modeled
Avg.
Std. Dev.
8.5
9.9
228.7
75.7
57.7
73.1
Cycling cost for
starts modeled
Avg.
Std. Dev.
3
3.6
167.5
26.1
83.8
27.5
that all units reflect their cycling costs, or do not, to avoid
the situation where only some generators are bidding cycling
costs as this leads to inefficient operation and excessive costs.
TABLE VII
C HANGE IN STARTS WHEN A SUBSET OF UNITS BID CYCLING COSTS FOR
START- UPS
∆ Starts
Units 1, 2, 3, 4, 9, 10
All other units
dynamic cycling costs quickly saturated as seen in Figure 5,
thus indicating that the majority of plant cycling is unavoidable. Table VI shows a breakdown of the total number of
starts by unit group, which again reveals that increasing starts
for peaking units are correlated with increasing incremental
cycling cost, as it becomes more favorable to dispatch these
units due to the relatively larger cycling costs associated with
the mid-merit units. (The ripples in the curve shown in Figure
5 result from the increasing starts for peaking units, as seen
in Table VI.)
-86
+256
B. Ramping Related Cycling Costs Results
Fig. 5. Impact of dynamic cycling cost on total start-ups, shown for various
multiples of costS
g (i)
Implementing a dynamic cycling cost for plant ramping
(shown in Table II) resulted in a 90% reduction in ramping
overall, as seen in Table VIII. As described previously, assuming a ramp greater than 20% or 40% of the difference between
a unit’s maximum and minimum output increments the ramp
counter, NgR (t), by a value of 1 or 2 respectively. The total
value of NgR (t) at the end of the test year, summed for all
units, is shown in Table VIII. Base-load units which carried
out the greatest amount of ramping when cycling costs were
not modeled, saw the greatest reduction in ramping operation
when cycling costs for ramps were implemented. The dramatic
reduction in ramping that was achieved by implementing
dynamic ramping costs, however, led to increased start-stop
cycling as might be expected, although only by 3.3% over
the year. The most notable change to the overall dispatch that
resulted from the introduction of dynamic ramping costs was
a slight reduction in production from base-load plant allowing
for increased production from mid-merit and peaking units as
seen in Table IX, thereby spreading the ramping requirement
over more units. Thus, including the ramping cost was also
seen to result in a slightly greater number of units online
(5.94 per hour on average when dynamic ramping costs were
modeled, versus 5.92 when no cycling costs were modeled).
TABLE VI
I MPACT OF DYNAMIC CYCLING COSTS FOR STARTS ON TOTAL PLANT
START- UPS , SHOWN FOR VARIOUS MULTIPLES OF costS
g (i)
TABLE VIII
I MPACT OF DYNAMIC CYCLING COSTS FOR RAMPING ON TOTAL ANNUAL
RAMPING (NgR (t, 1))
No cycling cost
costS
g (i)*0.5
costS
g (i)*1
costS
g (i)*2
costS
g (i)*3
costS
g (i)*10
Base-load
(Units 1-4)
Mid-merit
(Units 5-10)
Peaking
(Units 11-20)
34
13
12
13
13
13
1372
1104
1005
941
907
869
577
781
838
896
948
992
A scenario where cycling costs were only modeled for a
subset of the total fleet was also examined. The 6 largest units
on the system (units 1, 2, 3, 4, 9, 10) were chosen based
on the assumption that these units would be most impacted
by cycling operation and thus most likely to bid a wear-andtear cost into the market if such an option was available. The
results showed that although the number of annual start-ups
was reduced for these units, the start-ups for the other units
increased by a much greater amount as seen in Table VII. This
would indicate the need for a uniform policy relating to the
bidding of cycling costs to be implemented in markets, such
Units
Base-load (Units 1-4)
Mid-merit (Units 5-10)
Peaking (Units 11-20)
Total ramping
No cycling
costs modeled
Cycling cost for
ramps modeled
3717
2214
795
6726
120
1224
623
1967
TABLE IX
I MPACT OF DYNAMIC CYCLING COSTS FOR RAMPING ON AVERAGE PLANT
CAPACITY FACTORS (%)
Units
Base-load (Units 1-4)
Mid merit (Units 5-10)
Peaking (Units 11-20)
No cycling
costs modeled
Cycling cost for
ramps modeled
92.59
27.82
0.85
92.21
28.61
1.02
C. Start-up and Ramping Cycling Costs Results
Implementing dynamic cycling costs (as shown in Table
II) for starts and ramping simultaneously, reduced both types
8
of cycling operation relative to the case when no cycling
costs were modeled, as shown in Table X. Base-load units,
having the largest cycling costs, see the greatest reductions
in cycling operation. Nonetheless, neither total starts nor total
ramps were reduced in this scenario as much as starts alone
or ramps alone were reduced when cycling costs for starts
or ramps were modeled individually. However, when cycling
costs for start-ups only were modeled, ramping operation
increased and likewise when cycling costs for ramping only
were modeled, starts increased. Thus when the cycling costs
that would have been incurred due to both start-ups and
ramping are examined (assuming the costs given in Table
II), the case in which cycling costs for start-ups and ramping
were modeled simultaneously had the lowest overall cycling
costs, as shown in Figure 6. This would indicate that modeling
cycling costs for starts and ramping simultaneously is the most
cost effective way to reduce cycling and as such one should
not be considered without the other.
TABLE X
I MPACT ON TOTAL ANNUAL STARTS AND RAMPS WHEN DYNAMIC
CYCLING COSTS FOR BOTH START- UPS AND RAMPING WERE MODELED
Units
Base-load (Units 1-4)
Mid merit (Units 5-10)
Peaking (Units 11-20)
Total
No cycling costs
modeled
Starts
Ramps
34
3717
1372
2214
577
795
1983
6726
Cycling cost for starts
and ramps modeled
Starts
Ramps
12
144
1003
2069
855
1456
1870
3669
Fig. 7.
Total system costs shown for various scenarios
being determined by the level of knowledge of the generator’s
cycling costs.
The formulation for piecewise linear incremental cycling
costs related to plant start-ups and ramps was implemented
for a test system. Although the incremental costs chosen are
approximations, the results reveal certain trends that are likely
for power systems where generators undergo regular cycling
and reflect the resulting wear-and-tear costs in their bids. For
example, dynamically modeling cycling costs for generator
starts was seen to reduce the number of starts, but caused
ramping operation to be increased (and vice-versa), whilst
modeling cycling costs for only a subset of the generation
fleet was seen to induce much higher levels of cycling in the
remaining generation. It was also seen that as cycling costs
accumulated over time changes in the merit order occurred,
and that modeling cycling costs led to an overall saving for
the system as cycling operation was subsequently reduced.
R EFERENCES
Fig. 6. Cycling costs (that would have been incurred) shown for various
scenarios
Finally, when total system costs are examined for the
scenario including cycling costs and compared to the total
system cost for the scenario in which cycling costs were not
modeled, but were calculated and added afterwards, it can be
seen that modeling cycling costs leads to lower system costs
overall. This is shown in Figure 7. In this example, the cost
saving seen is considerable i.e. 14%.
V. C ONCLUSIONS
Interest concerning cycling costs is growing and this paper sets out a formulation that can utilize knowledge of
incremental wear-and-tear costs related to plant start-ups or
ramping, to implement a dynamic incrementing cycling cost.
The formulation covers linear, piecewise linear and stepshaped cycling cost functions, the appropriate choice for a user
[1] L. Göransson, and F. Johnsson, “Large scale integration of wind power:
moderating thermal power plant cycling”, Wind Energy, vol. 14, no. 1,
pp. 91-105, 2011.
[2] N. Troy, E. Denny and M. O’Malley, “Base-load cycling on a system with
significant wind penetration”, IEEE Transactions on Power Systems, vol.
25, issue 2, pp. 1088 - 1097, 2010.
[3] “Damage to Power Plants Due to Cycling”, EPRI, Palo Alto, CA, 2001.
[4] Editorial, “Profitable Operation Requires Knowing How Much it Costs
to Cycle your Unit,” Combined Cycle Journal [online], Spring 2004,
available: http://www.combinedcyclejournal.com/
[5] S. Lefton, “Profitable Operation Requires Knowing How Much it Costs to
Cycle your Unit”, Combined Cycle Journal, pp. 49-52, Second Quarter,
2004.
[6] S. Lefton, P. Besuner, P. Grimsrud, A. Bissel and G. Norman, “Optimizing
power plant cycling operations while reducing generating plant damage
and costs at the Irish Electricity Supply Board”, Aptech Engineering
Service, Tech. Rep. 123, Sunnyvale, CA, 1998.
[7] “Determining the Cost of Cycling and Varied Load Operations: Methodology”, EPRI, Palo Alto, CA: 2002. 1004412.
[8] “Correlating cycle duty with cost at fossil fuel power plants”, EPRI, Palo
Alto, CA: 2001. 1004010.
[9] M. Carrion and J.M. Arroyo, “A computationally efficient mixed-integer
linear formulation for the unit commitment problem”, IEEE Transactions
on Power Systems, vol. 21, no. 3, pp. 1371 - 1378, 2006.
[10] J.M. Arroyo and A.J. Conejo, “Optimal response of a thermal unit to
an electricity spot market”, IEEE Transactions on Power Systems, vol.
15, no. 3, pp. 1098 - 1104, 2000.
[11] S.A. Kazarlis, A.G. Bakirtzis and V. Petridis, “A genetic algorithm
solution to the unit commitment problem”, IEEE Transactions on Power
Systems, vol. 11, no. 1, pp. 83-92, 1996.
[12] S. Lefton, P. Besuner, D.D. Agan, “The real cost of on/off cycling”,
Modern power systems, vol. 26, no. 10, 2006.