To support its information gathering, the Committee on the Future of the Electric Power System in the United States convened a workshop on February 3, 2020, titled Models to Inform Planning for the Future of Electric Power in the U.S. More than 370 participants registered for the workshop to discuss available models for long-range, transmission, and distribution planning, as well as the broader context of how these models are used and future opportunities and needs.
The workshop featured four panels and two keynote addresses, each followed by open discussions among speakers, committee members, and workshop attendees. Participants from industry, government, and academia explored strengths and weaknesses of available models and modeling approaches, discussed how they have been used to guide utility planning and policy, and considered how they may be improved to account for emerging developments and prepare for a rapidly changing landscape in the coming decades.
The workshop was unclassified and open to the public. The workshop proceedings offers a condensed summary of the proceedings based on recordings, slides, and transcripts from the workshop, which can be accessed at www.nas.edu/gridmod.
Providing a reliable and resilient supply of electric power to communities across the United States has always posed a complex challenge. Utilities must support daily operations to serve a diverse array of customers across a heterogeneous landscape while simultaneously investing in infrastructure to meet future needs, all while juggling an enormous array of competing priorities influenced by costs, capabilities, environmental and social impacts, regulatory requirements, and consumer preferences. A rapid pace of change in technologies, policies and priorities, and consumer needs and behaviors has further compounded this challenge in recent years.
The Committee on the Future of Electric Power in the U.S. was convened by the National Academies of Sciences, Engineering, and Medicine to evaluate strategies for incorporating new technologies, planning and operating strategies, business models, and architectures in the U.S. electric power system. To support its information gathering, the committee convened a workshop on February 3, 2020, titled “Models to Inform Planning for the Future of Electric Power in the United States.” Speakers and participants from industry, government, and academia discussed available models for long-term transmission and distribution planning, as well as the broader context of how these models are used and future opportunities and needs.
Participants explored strengths and weaknesses of available models and modeling approaches, discussed how they have been used to guide utility planning and policy, and considered how they may be improved to account for emerging developments and prepare for a rapidly changing landscape in the coming decades.
THE PURPOSE OF MODELING
John Weyant, Stanford University and Energy Modeling Forum, set the stage with an overview of the use of modeling in the electric power system. Electric system modeling has always been difficult, he argued, but the challenge has become much greater over the past 30 years in the context of rapid changes in the technologies, institutions, and regulations. He said that modelers struggle to keep up with these developments while attempting to understand a future that is likely to have little resemblance to the past.
Despite these challenges, Weyant and other participants stressed that models are well worth the effort when created and used appropriately—that is, when the right model is applied to the right question, with appropriate transparency and awareness of uncertainties and limitations. “Often, it’s not the models themselves, it’s how they’re used that tends to be the constraint on them being more used and useful,” Weyant said.
Models are particularly useful as tools for understanding how policy changes or infrastructure investments might impact long-term outcomes. Weyant and other participants described how models provide context for anticipating the costs, market dynamics, environmental impacts, and distributional impacts of different technologies or policies, thus laying the groundwork for planners to make informed decisions.
“They’re not a crystal ball,” said Karen Palmer, Resources for the Future. “Models are a useful way to represent complex systems, tweak inputs, and gain insights.” While energy predictions for future decades can be imprecise, by continually improving models’ inputs and assumptions, it is possible to further improve their usefulness.
David Daniels, U.S. Energy Information Administration, described models as calculators that simulate the effects of decisions. “They don’t replicate the decisions of the real world, … [where] people are making decisions for all kinds of reasons, competitive pressures, all kinds of things,” Daniels said. “But models do simulate the effects of these decisions in the real world.”
Human behaviors and politics may influence how models and their outputs are interpreted and used. Many discussions focused on the imperative to appropriately communicate about models, their outputs, and their limitations with decision makers and other end users. Several participants noted that, for people who are not themselves modeling experts, it can be difficult to grasp the nuances of models and what they can and cannot provide. A few participants suggested ways experts and model creators can translate or explain models to make their insights accessible to these audiences.
MODELING AN INCREASINGLY COMPLEX LANDSCAPE
A pervasive challenge to modeling in the electric power system is the enormous number of variables involved. Modeling electric power systems can require accounting for a wide range of power sources, local geographies, and climate considerations; multiple and overlapping jurisdictions, such as federal, state, or local governments; multiple stakeholder interests, such as producers, customers, and regulators; and the vast uncertainties endemic to system operation, such as policy changes and unexpected outages. New risks, such as cybersecurity threats, are constantly emerging, further complicating the picture.
The real world is far too large and complex for a model to fully represent. Instead, modelers choose dimensions to focus on—such as region, time, power sector, fuel, or technology. Much as a telescope can be pointed at only one part of the sky, these choices create trade-offs between breadth and depth and affect the precision of model outputs.
A BROAD SPECTRUM OF MODELS AND APPROACHES
Given that no single model can capture all of the dimensions of the electric power system and related infrastructures at all temporal and spatial scales, a few participants stressed the need for a combination of models and
approaches. “Individually, no model can tell you the answer,” Daniels said. “Collectively, though, if you get an ensemble of models that are heterogeneous enough, you can cover the whole phase space.”
Throughout the workshop, participants discussed the design, merits, and applications of a variety of existing models such as the energy-economic model U.S. Regional Economy, Greenhouse Gas, and Energy (US-REGEN), developed by the Electric Power Research Institute; the Program on Coupled Human and Earth Systems (PCHES), developed by the Energy Modeling Forum; the Engineering, Economic, and Environmental Electricity Simulation Tool (E4ST), developed by Resources for the Future, Cornell University, and Arizona State University; and the Regional Energy Deployment System (ReEDS), the flagship model of the National Renewable Energy Laboratory, among many others. While many widely used computer modeling tools are discussed, this workshop is not an exhaustive review of all existing tools.
In addition to using multiple models in a complementary or an integrated way, several participants illustrated how the same models can be used for different purposes. For example, one user might focus on assessing costs and values of infrastructure investments, while another might use the same model to study decarbonization, guide the adoption of renewables, or inform the location of distributed energy resources.
KEEPING PACE WITH EVOLVING NEEDS
Participants considered how new modeling methods, computational capabilities, and integration approaches might provide insights on today’s most pressing electric power system needs.
Dealing with the interplay between different systems and models is an overarching challenge, several participants observed. Jason Fuller, Pacific Northwest National Laboratory, noted that distribution planning overlaps to a much greater extent with other aspects of the electric power system and indeed with other sectors than ever before. “It’s not just the grid anymore,” he said. “As power engineers, we have a tendency to think about transmission lines and power lines and transformers. … But our system is becoming something that’s much more intertwined with the fabric of society. It’s becoming tied into communications and other elements of the system.”
Efforts such as the Grid Modernization Laboratory Consortium (GMLC), the Grid Modernization Initiative (GMI), and the North American Energy Resilience Model (NAERM) are advancing the design of next-generation tools in order to help decision makers better understand energy, communications, and natural gas system interdependencies; rapidly predict national consequences of an extreme event; and model new solutions. Weyant argued that the increasing impact of multisector, dynamic interactions on the electric system necessitates the inclusion of as many interactions as possible in new models.
Tom Overbye, Texas A&M University and PowerWorld Corporation, spoke of the need to bridge academia and industry to address key challenges in electric power system planning. “How do we effectively leverage the broad community of researchers and practitioners?” he asked. He suggested systemic changes to align incentives and rewards, as well as practical steps to match the right experts and models with the right questions. He and other speakers explored ways to improve access to data and models to answer critical questions and train the next generation of practitioners, including through the development of synthetic grids.
Several participants stressed the importance of thinking broadly and creatively about what the future may bring. While it is difficult to model extreme and unpredictable events, such events do happen, and improving models to account for these events can help planners build more resilience into the nation’s critical electric power systems.
In addition, several participants noted the importance of keeping models flexible, where possible, in order to extend their use across different policy scenarios, market changes, or technologies, especially in light of the rapid pace of change in the electric power system.
MOVING TOWARD A MORE INTEGRATED APPROACH
Accurate models are valuable for all aspects of electric system operations and planning, from generation to transmission to distribution, and across multiple time scales, from microseconds to decades. A few participants discussed the increasing need to integrate models across these previously siloed facets. “It used to be almost completely separate,” reflected Anjan Bose, Washington State University. “Transmission planners never had to actually
talk to distribution planners, and distribution planners never had to talk to transmission planners, and we did just fine. And now, of course, everything is changing.”
Amos Ang, Southern California Edison, summarized the task for transmission planners as follows: “The two key issues that we are tasked [with], from the transmission planning perspective, is to be able to identify the issues before they happen, and [to] create solutions and fixes in place before these issues arise,” he said. Joseph Eto, Lawrence Berkeley National Laboratory, added that in his view, the most significant challenges for transmission planning are the changing resource mix, the complexities associated with incorporating new inverter-based generation technologies, modeling for resource variability, changing market structure and behavior, and understanding long-term impacts of new public policies and technologies.
On the distribution side, several speakers pointed out that many utilities are seeing a growing need for models that integrate and account for bidirectional flows, distributed energy resources, batteries, and other emerging technologies. Branden Sudduth, Western Electricity Coordinating Council, described how the changing resource mix creates the potential for stress conditions to occur at any time of day, not just during periods of extremely high or extremely low loads, as was assumed in the past. For distribution planning, several speakers suggested a need to move from deterministic models to probabilistic models and to work toward capabilities for dynamics analyses.
Many communities are setting ambitious carbon reduction and renewables adoption targets, and modeling is used to assess options and guide investments to achieve these goals. A session focusing on the experience of the Los Angeles Department of Water and Power (LADWP) highlighted valuable lessons for such efforts. The LADWP effort demonstrates how a continuous—rather than a “one-shot”—planning process has helped some utilities stay abreast of rapidly changing technologies, policies, and consumer perspectives. In describing the LADWP case, several speakers underscored the value of a holistic, integrated approach to address the multiple technological, environmental, economic, and social facets of energy planning.
Whether one is examining the details of daily operations or creating a 30-year forecast, a persistent challenge across all electric system modeling is the need to understand and mitigate trade-offs while being poised to react to challenges and developments as they occur. “This real-time balancing of supply and demand really drives everything that’s going on in this industry,” Eto said. “Things happen in the blink of an eye.”
February 3, 2020
Beckman Center, Huntington Room
Granger Morgan,* Carnegie Mellon University and Committee Chair
|8:35 AM||Overview of Models Used in Electric System Analysis and Planning
John Weyant, Stanford University Energy Modeling Forum
|9:15 AM||Long-Term Electric System Modeling
Moderator: Karen Palmer,* Resources for the Future
|11:00 AM||Models Used for Transmission Planning
Moderator: Anjan Bose,* Washington State University
|1:30 PM||Models Used for Distribution System Planning
Moderator: Susan Tierney,* Analysis Group
|3:20 PM||Case Study—Modeling to Support LADWP’s IRP and Stakeholder Engagement
Moderator: Reiko Kerr,* Los Angeles Department of Water and Power
|4:40 PM||Closing Keynote
John Grosh, Lawrence Livermore National Laboratory
* Denotes member of National Academies Committee on the Future of Electric Power in the U.S.