Big Data Analytics: Research Needs
Ali Ghassemian
April 28, 2016
Plan
DOE’s Grid Modernization Initiative (GMI) represent a comprehensive effort to help shape the
future of our nation’s grid and solve the challenges of integrating conventional and renewable
sources with energy storage and smart buildings, while ensuring that the grid is resilient and secure
to withstand growing cybersecurity and climate challenges.
DOE’s comprehensive Grid Modernization Multi-Year Program Plan (MYPP), is the blueprint for
DOE’s R&D agenda and the modernized grid. The plan lays out the agenda and priorities for the
Department’s research, development, and demonstration efforts to enable a modernized grid,
building on concepts and recommendations from DOE’s Quadrennial Energy Review (QER) and
Quadrennial Technology Review (QTR). The plan describes the R&D activities that DOE will focus on
over the next five years, including continued opportunities for public-private partnerships and
enhanced collaboration among DOE offices.
Through the GMI, MYPP, QER, and QTR the Department will help frame new grid architecture design
elements, develop new planning and real-time operations platforms, provide metrics and analytics to
improve grid performance, and enhance government and industry capabilities for designing the
infrastructure and regulatory models needed for successful grid modernization.
Successful transformation requires a coordinated, whole-grid approach designed to research,
develop, demonstrate, and assist in the deployment of key technologies.
2
Challenges
• Modern-day demands on the U.S. electricity grid have
required the introduction of smart devices to improve
resilience and reliability in the face of changing grid
characteristics and dynamics such as an increase in
renewables, electric vehicle integration, changing load
patterns, changing flow, and faster performance speeds.
• As a result, today’s grid produces an enormous amount
of data and a significant challenge is how to enable grid
operators to make sense of such a large quantity of grid
state and customer data in near-real time.
• Today’s technologies, tools, and techniques are not
presently up to the challenge.
3
Big Data Challenges
• Numerous sensors of different types placed in networks
provides massive amount of data
• Data from different sources need to be validated and
integrated.
• Data needs to be processed into actionable information to be
useful for system operators and engineers
• As data volumes increases, scalable solutions to handle huge
amounts of data become a necessity.
4
What is Needed and how to Achieve it?
• Develop advanced computational (software) and control technologies
(hardware) to improve the reliability, resilience, and efficiency of the
nation’s electricity systems
• Help prevent blackouts and improve reliability by providing wide-area
real-time visibility into the conditions of the grid
‐ Track and expand the use of real-time data in grid control rooms and
energy management systems
• Build the technology to manage and control future grid operations
‐ Improve the performance of modeling tools and computations that are
the basis of grid operations
• Assess and mitigate near- and long-term risks to energy infrastructures
and systems
‐ Track and expand the use of quantitative risk and uncertainty methods
in Federal and state-level energy system decisions regarding energy
infrastructure investment
5
Advanced Grid Modeling Program
AGM Develops the Core Operational and Planning
Components Supporting the Future Grid
Advances the computational and mathematical methods
underpinning operator tools
Seeks to develop “faster than real time” analytical tools
through work in three main areas
•
•
•
6
Data management and analytics
Mathematical methods and computation
Models and simulations
Main Areas for Analytical Tools
• Data management and analytics (DM&A):
‐ These activities focus on the way data is collected, used,
stored, and archived to improve applicability of large, multisource datasets for real-time operations and off-line planning
studies.
• Mathematical Methods & Computation (MM&C):
‐ Effort addresses emerging mathematical and computational
challenges arising in power systems, developing new
algorithms and software libraries.
• Models & Simulations (M&S):
‐ Research on a new class of fast, high fidelity capabilities that
underpin better grid operations and planning in a large-scale,
dynamic and stochastic environment.
7
Projects in AGM
•
Developing State Estimation that can run at a
unprecedented 0.5 s speed.
•
Developing Dynamic Contingency Analysis tools (DCAT)
•
Integrating Planning Dynamics and system Protection
simulations models in Computer aided protection
engineering tool (CAPE) that allows performing a more
accurate analysis of the behavior of protection equipment
during the first cycles after a fault condition.
8
Projects in AGM
•
Developing probabilistic Methods for Electric Grid
Operations
•
Refining the MMWG models by modeling governor
deadband and adjusting active governor ratio and load
composition in order to match up measured EI frequency
responses. This will help with validating power system
dynamic model that is used to perform contingency
analysis.
•
Developing Dynamic Models for Year 2030 Eastern
Interconnection
9
Projects in AGM
•
Developing GridPACK which is an open source HPC library that
includes functionality such as Power Flow, Dynamic Simulation,
Power Flow Contingency Analysis, and Dynamic Security
Assessment
•
Power System Parallel Dynamic Simulation Framework for RealTime Wide-Area Protection and Control
•
Developed Management & Optimization of VARs for Future
Transmission Infrastructure with High Penetration of
Renewable Generation
10
Main areas of concentration in AGM projects
applicable to the Big Data
• Parallel computing for big data analysis
• integration of multiple data sources
• PMU, SCADA, Weather, etc.
• Improved event detection and data mining algorithms
• Improved processed and actionable information
11
Data Analytics Research @ CURENT (Center for Untra-Wide-Area
Resilient Electric Energy Transmission Networks)
Data Processing –
Data compression
and reconstruction;
using synchro-phasor
data; use load data to
identify potential
demand response
12
Data Integration –
Overcoming
challenges of
integrating data from
different information
systems
Data Mining and
Statistics –
Correlating data
from IEDs, weather,
power systems to
identify
vulnerabilities and
locate faults
Cyber and Data
Security Produced a new
method to detect
coordinated cyberattacks in power grid
information systems
GMLC
DOE announced funding in January, 2016 of up to $220
million over three years for DOE’s National Labs and
partners.
The Grid Modernization Laboratory Consortium (GMLC)
funding will support critical research and development in
the AGM Subprogram. The concentration is in:
1.
2.
3.
13
Load Modeling
Protection
Solvers
GMLC AGM Projects
• Load Modeling:
• Contribute towards development and validation of
mathematical model structure capturing emerging load
dynamic behaviors.
•Protection:
• Enhance protection system (misoperation) modeling
capabilities, as a platform for the study and coordination of
protection devices and approaches.
•Solvers:
• Create, maintain, and enhance a scalable math solver library for
grid planning and operations tools that work on a variety of
computational platforms
14
National Academies of Science, Engineering, and
Medicine
DOE commissioned National Research Council (NRC) to engage in a
study with the following main charge:
•
What are the critical areas of mathematical and computational
research that much be addressed for the next-generation
electric transmission and distribution (grid) system?
•
Identify future needs.
•
In what ways, if any, do current research efforts in these areas
(including non-U.S. efforts) need to be adjusted or augmented?
15
Findings
1. Analyze the vast amount of that data coming to control
centers and represent the results in a way suitable for
timely decision making.
2. Improve modeling of grid operation to increase efficiency.
3. Advances in the mathematical and computational
algorithm are needed to address the challenges arise from
the integration of more alternative energy sources into the
system as well as to help reduce the risk of voltage
collapse and enable lines to be used within the broader
limits, and flexible ac transmission systems and storage
technology can be used for eliminating stability-related
line limits.
16
Findings
4. Establish a more complete models that include the dynamic
interaction between the transmission and distribution systems.
5. Establish a better planning models for accurate forecasting to deal
with uncertainties (such as distributed-generation technologies,
climate change, shifting rainfall, frequency of intense weather
events, changes in policies to reduce emissions of carbon dioxide)
6. Establish a modeling and mitigation plans for high-impact, lowfrequency events (including coordinated physical or cyberattack;
pandemics; high-altitude electromagnetic pulses; and large-scale
geomagnetic disturbances) by doing fundamental research in
mathematics and computer science that could yield dividends for
predicting the consequences of such events and limiting their
damage.
17
NAS Recommendations
The future grid will rely on integrating advanced computation and
massive data to create a better understanding that supports
decision making.
To complement this research specifically focused on tools for the
next-generation grid, a range of potentially applicable research
exists. Examples include research into the general topics of:
• uncertainty quantification
• simulation and analysis of complex adaptive systems
• simulation and analysis of multi-time-scale systems
• methods for characterizing and controlling resilience and
reliability.
18
NAS Recommendations
Make data available for testing
Create synthetic data for development, testing and validation
Advance the ACOPF
Research on Data driven approaches applied to the operations,
planning and maintenance of power systems.
Improve the nonlinear, nonconvex optimization algorithms
DOE and NSF should sponsor the development of new-open
source software for research community
Promote collaboration between national labs and universities
Integration of theory and computational methods should be a
high-priority research area
DOE should support research to extend dynamical systems
theory and associated numerical methods to encompass
classes of systems that include electric grids
•
•
•
•
•
•
•
•
•
19
Summary
AGM program intends to enhance reliability, resiliency, security,
and efficiency of the electric power and enable advanced
mitigation and recovery strategies, by:
• Accelerating performance
• Developing predictive/corrective decision-support capabilities
• Integrating model platforms
20
Question
?
21