Search form
Search
FERC Order 809
Solutions Using Data Science
February 8, 2016
Perspective on Operating Challenges
By: Rick Smead
Perspective on Data Science Solutions
By: George Danner
2
The Situation
 U.S. is making strategic shift to Natural Gas Fired
Electricity Generation
 Success of shift depends on better coordination among
gas supplies, transport services, ISOs / RTOs, and power
generators
 FERC Order 809 seeks to improve alignment of gas
supplies with needs of power generators through:
 Changes to the pipeline nomination process
 Finding ways to compress the timeframe between nom
submissions and schedule confirmation
3
Anticipate Operating Complexities
 Need to anticipate significant increase in Gas-Fired Plants to
replace Coal Plants, at the same time that Renewable Sources,
such as wind and solar are moving into the same space
 This evolution has two major implications


Winter generation use of gas is increasing, competing for
capacity with heating loads
Intermittent nature of renewables means that gas-fired
generation operation will be more volatile
 Service interruptions will create problems for entire supply
chain
 In spite of challenges, growth in natural gas-fired power
generation can provide win-win opportunity for entire natural
gas supply chain
4
Operating Challenges
Changes to proven operating procedures could
introduce new operating risks
 Less time to validate nomination data could create
errors and potential legal risks
 Tighter deadlines could hamper Gas Controllers’
ability to prepare pipeline for shifts in volume
 Insufficient lead-time to manage line-pack
 Impaired ability to ensure delivery of scheduled volumes
5
Current Perspective
 Strong consensus that more computers will not
resolve risks
 Existing processes are critical to ensuring correct
data is used for confirmation and scheduling
processes
 Gas Control can respond to rapidly changing
situations, but only by anticipating and having
enough lead time to move resources such as line
pack where they are needed
6
Technology Can Be a Part of Solution
 The solution is in:
 Using new technology to anticipate, validate, and alert
 Developing capabilities that help schedulers and controllers:
 On both sides of the transaction
 Prepare them for potential events, then prompt them with the
solution when the event occurs
 The solution could look something like how pilots practice
responding to situations in a simulator and then respond to
alerts in the cockpit
7
Visualization Makes a Difference
 It is insufficient simply to find the right policies or operating
rules
 Many times, the challenge is gaining buy-in from skeptical
groups of stakeholders
 Experience has shown that helping stakeholders visualize “a
day in the life” can break-down significant communications
barriers
 By letting stakeholders “play with the rules” they can:
 See what is happening
 Understand how the policies work
 Build consensus
8
Proposition: A GEH Simulation
 Scope can range from:
 Combinations of Gas Contracts
 Pipeline ecosystem
 Linked gas/electric ecosystem
 Provide ability to conduct “experiments” with a variety of
policies
“What if this step was added to the coordination protocol?”
 Can also stress test the policies against a wide range of future
scenarios, both “normal” and “exceptional”
 Use Subject Matter Experts from both domains to construct the
model
 Make it a transparent process
9
Proposition: Create a Flight Simulator
 It is also possible to create a game-like interface on top of
the model to turn it into an immersive exercise
 Companies do this to “role play” various upcoming
scenarios (competitive strike, new product line, …)
 Could the same thing work for GEH?
10
Proposition: Compress Confirmation Process
 Help Pipeline Schedulers expedite verification of
nomination
 Issue: Incomplete or flawed nomination data could impair
service of accepted nominations
 Solution: Use analytics tools to identify errors and alert
scheduler to problems
11
Presenters’ Bio
George Danner
Chief Science Officer
George is a leader in the application of Data Science tools to solve some of business’ most complex
business problems. He has deep experience in combining analytical tools such as agent-based simulation
modeling, Monte Carlo simulation, System Dynamics, Game Theory, Visualization, and Real Options with
visualization tools to help clients gain better insight into the impact their strategic and operational
decisions. He is the author of Profit From Science, published by Macmillan. George has 29 years of
experience in corporate strategy, specifically operational and financial analysis, across a wide variety of
industries: manufacturing, energy, telecommunications, transportation and financial services. He holds a
Bachelor of Science in Mechanical Engineering from Texas A&M University and a Master of Science in
Management from Massachusetts Institute of Technology.
Rich Smead
Managing Director of
Advisory Services
Rick specializes primarily in the natural gas sector, offering expert policy analysis and advice, litigation
support, and strategic advice with respect to gas pipelines, potential supplies, and market initiatives. His
background includes over nine years as a Director with Navigant Consulting, Inc., and over three decades
in the natural gas industry. That experience included over 20 years as the chief regulatory officer for
major interstate pipeline systems, with multiple experiences advising non-U.S. governments in the
creation of a regulatory structure. His consulting practice has spanned the LNG industry, both for project
developers and for a large national oil company, and the U.S. shale gas boom and its interaction with LNG,
the pipeline industry, power generation, and other gas markets. At Navigant, he managed and co-authored
the first major quantification of the U.S. shale potential in 2008, the pivotal North American Natural Gas
Supply Assessment. Most recently, he has been deeply involved in the opportunities for the use of the
nation’s natural gas abundance, including power generation, LNG exports, and gas-to-liquids technology.
He holds a Bachelor of Science in Mechanical Engineering from the University of Maryland and a law
degree from George Washington University.
12
Appendix
13
Proven Technology for FERC 809 Objectives
Data Science can address FERC 809 Objectives:
FERC
Objectives
Risks
Compress
Nomination –
Schedule
Approval Process
Incomplete or flawed
nomination data could
impair service of accepted
nominations
Machine Learning to learn when confirmation data is
not valid, analyze such data, and alert schedulers to
potential errors
Unanticipated volume
changes adversely impact
pipeline reliability
Simulation tools for gas controllers to practice
responding to potential situations and pre-plan
responses for when the situation develops
Data Science Solutions
Real-time Event Monitoring by embedding simulation
logic into appropriate operational systems, which
monitor on-going transactions and alert controllers
when specific scenarios are identified
14
Proven Technology for FERC 809 Objectives
Data Science can address FERC 809 Objectives:
FERC Objectives
Create contracting
flexibility to help
entities meet their
collective load
obligations in a more
efficient manner
Risks
Improper or
unpredictable use of
contractual capacity
impairs rights of other
shippers
Data Science Solutions
Simulation tools for agents responsible for
managing a pool of contracts, which allow its
schedulers to practice responding to potential
situations and pre-plan responses for when the
situation develops
Real-time Event Monitoring by embedding
simulation logic into appropriate operational
systems, which monitor on-going transactions
and alert schedulers when specific scenarios are
identified