Even when a utility uses one of the best resource planning models available, executives can still short-circuit the model’s capabilities.
In the case of Dominion Virginia Energy, the utility’s high-end PLEXOS model can readily determine how much solar, storage and wind power would save customers the most money. But Dominion excluded storage when it ran the model, contending that storage is “more costly” than other options in testimony elicited by the Sierra Club, through legal discovery.
The PLEXOS model has the capacity to evaluate battery resources, but the utility “simply failed” to use the model to do so, testified expert witness Rachel Wilson, of Synapse Energy Economics, for the Sierra Club. She said the utility should be required to model solar + storage as “selectable resources” at reasonable costs in its future modeling work.
Having bypassed solar + storage—which is increasingly seen as the lowest-cost option—Dominion Virginia plans to meet peak demand in Virginia with up to 3.2 GW of additional gas-fired combustion turbines by 2033. The utility’s parent company owns gas pipelines. (The utility did include a state-mandated 30 MW storage “pilot” project in its plans.)
Dominion’s plan also handicaps solar by:
- Not considering flexible solar generation—whereas in Georgia, 4 GW of flexibly operated solar now could save customers money.
- Allowing competitive bids for power purchase agreements on only 25% of new solar, while reserving 75% of new solar for the utility to build. That’s contrary to a finding that “the solar power purchase agreement option is substantially less costly than the company-build solar option,” according to testimony by Gregory Abbott, a Deputy Director with the State Corporation Commission’s Division of Public Utility Regulation.
- Assuming constant costs for solar, rather than continued declining costs as panel efficiencies improve and soft costs decline. That constant-cost assumption was uncovered by Ms. Wilson of Synapse Energy Economics, through a legal discovery process. The unreasonableness of using constant solar costs is shown by testimony in a Georgia proceeding that even NREL’s mid-case projections for solar costs “have historically underestimated real-world cost declines,” according to Energy and Environmental Economics Senior Partner Arne Olson.
- Limiting solar additions to 480 MW per year—except in 2028, when 1.2 GW of solar would be added to reach the state’s legislated goal of 5 GW of solar.
Solar has been similarly short-changed in resource plans recently filed by utilities in North Carolina, Georgia and Tennessee. Regulated utilities in 33 states must file “integrated resource plans” (IRPs) every few years, for review by state regulators.
In a rare point of agreement, a solar industry group and Dominion agreed that the state’s requirement to use a 23% solar capacity factor for modeling purposes was too low, though they differed on how much. The 23% value used over the next 15 years was “artificially low,” testified Mid-Atlantic Renewable Energy Coalition’s expert witness Michael Volpe, a vice president at Open Road renewables. He predicted “capacity factors nearing 30% over the next few years” as the use of single-axis trackers and bifacial panels become the norm. Dominion maintained that a 25.4% capacity factor, which it used in developing its previous plan, was appropriate.
Noting that Dominion announced $17 billion in additional capital spending in Virginia at a March 25 investor presentation, regulatory staff had requested that Dominion re-run the PLEXOS model to reflect the full $17 billion in planned spending, but a hearing examiner denied the request.
Staff of the State Corporation Commission (SCC) are now reviewing Dominion’s plan and all testimony. Last year, regulators rejected Dominion’s initial IRP. SCC Deputy Director Gregory Abbott testified that the new plan’s lowest-cost option costs nearly $8 billion less than that of the plan regulators rejected last year.