Until recently, the electric power industry was an exception to the old adage, “It is difficult to make predictions—especially about the future.” Most electric utilities were able to forecast power supply and demand with relative confidence.
“In most places until about 2008, electric demand growth mirrored the trajectory of gross domestic product. Load was forecastable, generation was dispatchable, and electrons flowed in one direction,” said Bill Gould, EPRI director of strategic analysis and technology assessments. “Resource and system planning was much more predictable than it is today.”
Utility resource planners now face many more variables. For instance, the U.S. Energy Information Administration expects non-hydro utility-scale renewable generation to account for nearly 9% of total U.S. power supply next year, led by intermittent wind power. Non-hydro utility-scale renewables accounted for 7.1% in 2015 and 3.1% in 2008. States such as California, Vermont, and Hawaii are calling for 50–100% renewable energy in power portfolios. Increasing variable generation from customer-owned photovoltaic systems and other distributed resources tests the limits of local distribution circuits. The U.S. Environmental Protection Agency’s Clean Power Plan mandates states to limit electric sector carbon emissions.
These and other developments introduce uncertainties into utility planning and operations. How will more intermittent generation impact system reliability? How will conventional power plants, designed for baseload operation, perform as they operate more flexibly to accommodate solar and wind? How will environmental impacts of solar and wind power plants affect their operation? How will consumers interact with the new generation mix?
Traditional forecasting models are based largely on one-way electricity flow, dispatchable generation, and economic growth estimates—and they cannot account for these new developments.
“The grid is more dynamic than ever before, and the modeling methodology needs to reflect this,” Gould said.
Modeling: A Research Imperative
Gould is leading EPRI’s “research imperative” to improve understanding of the changing power system and identify modeling gaps.
The effort is detailing key certainties and uncertainties integral to grid planning and operations, then addressing them in EPRI’s models, and providing key insights to enhance modeling industry-wide. EPRI is creating a roadmap for incremental changes to existing models and for developing methodologies that factor nontraditional variables such as customer behavior.
For instance, EPRI is modifying its U.S. Regional Economy, Greenhouse Gas, and Energy (US-REGEN) model to help better understand how changing supply, new transmission and distribution technologies, and more active customers could interact with various economic and policy scenarios. The electric demand side of US-REGEN will examine in greater detail the interactions between end-use efficiency, demand response, and the competitiveness of electricity in the broader energy system. On the electric supply side of the model, EPRI is investigating ways to incorporate more operational realities—the moment-by-moment technical options for balancing electricity supply and demand—when creating multidecadal scenarios.
Another major step will be to link planning for transmission and distribution, which have historically been loosely connected. As part of this, EPRI is working with the New York Power Authority and the Advanced Grid Innovation Laboratory for Energy to develop, test, and validate new modeling and assessment practices.
Aligned with EPRI’s Integrated Grid concept, these efforts to improve, enhance, and integrate industry forecasting models can help equip the industry to tap the full potential of distributed energy resources by incorporating them into grid planning and operations.
“Rather than build one supermodel, we’re going to assemble a connected set of tools to address the many new uncertainties,” Gould said. “Each tool will be useful on its own and even more useful as we link them in a consistent approach for thinking about the future.”
EPRI Technical Experts:
Bill Gould, Tom Wilson
