A battle is brewing in southern California between a utility planning to build a new natural gas plant and clean energy advocates who say solar and storage facilities could provide the power for cheaper. The issue could come to a boil this week.
At issue is the Puente Power Project, a 262 MW natural gas plant to serve Southern California Edison that was approved last year to replace older generation set to be pulled offline in 2020.
The plant would cost about $299 million and be constructed by NRG Energy. It is designed, along with renewables and storage resources, to help replace about 2,000 MW of older once-thru cooling units at the Mandalay and Ormond Beach power plants, which are both part of the Moorpark Subarea in California.
The region includes the cities of Oxnard, Santa Barbara, and Goleta, and since 2012 the utility has been eyeing a a potential shortfall when the older units are retired. Additionally, upgrades are being considered at the Ellwood Generating Facility, a 54 MW peaking plant.
At the urging of several state lawmakers and following a Los Angeles Times report that suggested the state was overbuilding power plants, the California ISO agreed to look into alternatives to the Puente plant this summer.
The results, released last month, stung clean energy advocates.
CAISO determined that by procuring distributed resources that include solar power, demand response and energy storage, the Puente gas plant could be avoided — but at a higher cost.
Stacked up against the $299 million gas plant, CAISO's study found three clean energy alternatives would cost anywhere between $309 million and $1.1 billion.
Renewable energy advocates say the grid operator is mistaken.
"We took a look at that $1.1 billion number and recognized that it was not anywhere close," said Clean Coalition Policy Director Doug Karpa, who has been working on the issue.
“They just muddled everything in, and said 'we're just gonna throw in a bunch of batteries.' From an engineering modeling standpoint it's perfectly fine, a megawatt is a megawatt," said Karpa. "But from a cost perspective, if you're actually gonna go out and build the thing, you would never use all batteries. You could talk to any developer out there, and they would say that's not the cost effective way to do that."
Clean Coalition has since done its own modeling to show solar-plus-storage facilities can be cheaper than potential replacements examined by CAISO. A hearing is set for Sept. 14 at the California Energy Commission to discuss the cost estimates.
Until then, CAISO representatives say they can't discuss specific aspects of the cost studies. But the difference in modeling highlights the challenge in evaluating investment opportunities for power generation in an age of climate change. Costs are falling for several different types of carbon-free technologies, often faster than accounted for by resource planning processes.
Each of CAISO's three alternatives includes 80 MW of energy storage-enabled demand response resources, 25 MW of incremental solar/energy storage hybrid resources, and approximately 30 MW of existing slow responding demand response resources coupled with incremental energy storage.
But that 135 MW is insufficient to meet the Moorpark needs. Incremental distributed resources plus grid connected battery storage would cost $805 million, CAISO determined. If the Ellwood Generating Station is also retired, incremental distributed resources plus grid connected battery storage could cost up to $1.1 billion. Other configurations without the retirement would cost between $309 million and $359 million.
"If you were doing this for real, you would use a combination of solar and storage. It makes sense," said Karpa.
One problem, Karpa said, is that the towns involved and CAISO are not project developers. And the study used battery costs from 2014, when they've been falling by about 11% annually. By 2018, he said, battery costs could be about half the estimates used.
"The component costs weren't right," said Karpa, "and they didn't design a system that could take advantage of the federal investment tax credit for solar. That means their project is already 50% too expensive."
"You have to bring a financial awareness to the project, which CAISO just didn't. That's not their expertise," he said.
That's something the ISO concedes in its assessment.
"As the ISO does not conduct resource procurement, the ISO’s access to cost information is limited to publicly available sources," the report says. "While the ISO initially did not include development of cost information in the original scope, very high level estimates were subsequently developed based on publicly available information to be helpful and provide a starting point for subsequent discussion in the CEC proceeding."
In addition to using outdated battery cost numbers, Greentech Media also notes the study assumed a capital cost for solar in 2020 that is significantly above California prices today.
According to Karpa, when you model more realistic alternatives, they show that a solar-plus-storage solution is achievable at $267 million to just replace the development of Puente, and solar-plus-storage could replace both Puente and Ellwood for approximately $406 million.
Among the major differences, Clean Coalition said its model more appropriately sizes the storage required by modeling the real generation and dispatch capabilities of solar-plus-storage.
"CAISO’s unrealistic profile of solar output and storage dispatch resulted in underestimating the energy generation of solar by nearly half and oversizing of storage," the group said in a statement.
SCE, NRG still support Puente
Heading into next week's hearing, NRG and Southern California Edison are still supportive of the Puente project. They note that resource planning is a long and difficult process — decisions cannot constantly be revisited, or nothing would get built.
In a statement, the utility said it "supports all of the projects that it awarded in 2014 and which have since been approved by the CPUC ... the Puente project was competitively selected, in addition to energy efficiency and distributed generation projects."
The Moorpark local capacity requirements process has been underway for five years, and SCE stressed that the selected projects need to be developed in just a few years.
"It is important to remember that the approval for new resources is a very lengthy and costly process, and it is not practical to continually revisit development decisions, as this can make it more difficult to address reliability needs on the system," SCE said.
The utility said it is "speculative" to assume that carbon-free resources "can be developed on the scale and at the cost needed to competitively replace the Puente project by 2021."
NRG, which would construct the power plant, said CAISO's analysis continues to show Puente "is the optimal project for reliability and the rate payers."
NRG spokesman David Knox said the company "understands the advantages, limitations, constraints, and costs of implementing preferred resources, including solar-plus-storage."
"This knowledge and experience allows us to confidently say that the Puente Power Project is the best project to meet the local capacity resource needs in the Moorpark Sub Area," he said.
The evidentiary hearing on Puente begins at 9 a.m. PST on Thursday, Sept. 14. The Energy Commission has also scheduled committee meetings on the project for Tuesday, Sept. 12 and Monday, Sept. 18, accessible online at the CEC website.