Hawaiian Electric, Hawaii’s main investor-owned utility, has been facing the challenge of incorporating renewable energy into its island grids for years now, from the weather-induced shifts in utility-scale wind and solar farms, to the backflows from rooftop solar on its distribution circuits.
But that’s nothing compared to the 100 percent renewable grids it will have to operate someday. That’s because those grids will have to operate without the stability of synchronous generators — the spinning steel turbines, wrapped in wire and magnets, and mostly powered by fossil fuels, that serve as the anchors of almost all large-scale power grids today.
Last week, HECO filed a plan with state regulators that represents its first steps toward a future grid that doesn’t rely on these massive spinning generators.
By 2024, it hopes to replace two set-to-close power plants — the 203-megawatt AES Hawaii coal-fired power plant on Oahu and the 212-megawatt Kahului oil-fired power plant on Maui — with a mix of energy storage, grid services and “Renewable Dispatchable Generation power-purchase agreements.”
In another bold move driven by state policy, almost all of this mix of replacements — about 240 megawatts of standalone storage, 200 megawatts of grid services such as frequency regulation and demand-side capacity, and up to 900 megawatts of renewables-plus-storage projects — will be built and owned not by the utility, but by third parties through an open request-for-proposals process.
These are the fundamentals of the “Proposed Final Variable Renewable Dispatchable Generation RFPs” plan that HECO filed on July 17 with the Hawaii Public Utilities Commission.
On first glance, the new RFPs appear to be a continuation of HECO’s previous big solar and energy storage procurement from earlier this year. As we covered at the time, these Phase 1 contracts, for 262 megawatts of solar and 1,048 megawatt-hours of storage distributed over three islands, included some “jaw-dropping” low prices for solar-plus-storage systems, ranging from 12 cents to 8 cents per kilowatt-hour. They were formally approved in March.
The projects announced earlier this year did come with some novel structures to support Hawaii’s long-range grid needs. For example, AES’ 25-year, 30-megawatt solar and 120-megawatt-hour storage PPA uses a monthly “lump sum payment” to the developer based on net energy potential and facility availability, rather than energy delivered. That reduces curtailment risk for the solar developer and gives HECO some dispatchability of the storage system.
But the new RFPs are fundamentally different from the first batch, according to Ravi Manghani, energy storage head at Wood Mackenzie Power & Renewables.
“The Phase 2 RFP takes a more technically advanced approach toward resource planning, with a variety of new resources,” he said. That includes fully dispatchable storage, grid services including demand-side capacity and frequency response, and renewable-plus-storage contracts that encourage turning over at least some of the associated battery capacity to serving grid needs.
Hawaii’s island-by-island grids are highly unusual, compared to the networked and transmission-served grids of the mainland. That’s forced Hawaii to spearhead its own efforts to create a regulatory framework, called Integrated Grid Planning, that’s akin to an island-scaled version of the integrated resource plans developed by mainland U.S. utilities to manage their long-range generation and transmission needs.
“It’s a pretty interesting approach to resource planning, and perhaps an indication of how utilities elsewhere might do resource planning” as they start to encounter the same challenges as HECO is facing, Manghani said.
A resource mix to replace an island grid’s retiring generation
HECO is still heavily reliant on oil, which made up 63 percent of the utility’s electricity fuel mix for Oahu and Hawaii island and 68 percent for Maui in 2018. The AES Hawaii plant, the state’s only coal-fired power plant, made up an additional 17.5 percent of Oahu’s fuel mix last year — a gap that will need to be filled by alternatives by 2022, when it is scheduled to close after its current PPA expires.
The Kahului plant, set to close in 2023, is much smaller than Maui’s main oil-fired power plant, and is used only intermittently to avoid curtailment of wind farms and other renewable energy resources. Still, its location on the wind-rich western part of Maui makes it a critical grid resource for the island.
“There are unique challenges to developing a procurement plan that combines third-party participation via competitive bidding with the resources required to replace two key spinning generation resources for Oahu and Maui,” Jim Alberts, Hawaiian Electric senior vice president for business development and strategic planning, wrote in an email this week.
First, HECO must ensure that projects are built and running in time to meet the need of the facility being replaced. HECO has set guaranteed commercial operation dates “as early as possible to provide some cushion in event of project delays,” and provided for liquidated damages if schedule milestones are missed to incentivize projects to meet their commercial operations dates.
Second, HECO needs its new resources to be “available 24 hours a day, similar to the spinning generation resources being replaced, which is not typical of variable generation,” Alberts wrote. That’s something of an understatement, of course. Solar and wind power is weather-dependent, and solar power can’t generate power at night.
But HECO’s new procurements are meant to provide round-the-clock coverage, in the case of its standalone storage and grid services components, or a much broader than usual range of operating conditions, in the case of its new “Variable Renewable Dispatchable Generation and Energy Storage” RFPs.
An RFP built to incentivize grid services from solar-plus-storage
These solar-plus-storage projects are expected to make up the lion’s share of Phase 2’s new development, with HECO projecting the equivalent of 594 megawatts of solar for Oahu, 135 megawatts for Maui, and 32 to 203 megawatts for Hawaii Island, depending on whether other renewable energy projects become available.
HECO proposes that all of these solar projects must be paired with storage, and that all of that storage has the ability to be charged from the grid, rather than only from the associated solar. That’s an important point for solar-plus-storage projects, because batteries must use solar power for the vast majority of their charging needs to earn the federal Investment Tax Credit (ITC), and charging them with grid power could jeopardize that tax treatment.
To gain dispatch control over these assets, HECO will use a Renewable Dispatchable Generation power-purchase agreement, which will “provide complete dispatch rights to [HECO] provided the resource and/or storage is available,” Alberts wrote.
And to encourage developers to increase the proportion of their battery capacity open to grid storage, HECO’s filing proposes linking the RFP valuation of the associated storage with how frequently developers allow the utility to charge and discharge it from the grid. Batteries limited to solar charging only will get 10 percent of their megawatt rating counted toward meeting the RFP.
But batteries that HECO is allowed to grid-charge up to the maximum allowed by the ITC — 25 percent of total charging over the five-year ITC recapture period — will be allowed “up to 100 percent” of their megawatt rating to be counted for the RFP, “depending on the amount of grid charging.”
Standalone storage, the must-have resource
HECO’s standalone storage portion of its proposal, though smaller than its “renewable dispatchable” RFP, is more critical to making sure Oahu and Maui’s grids are ready for the coming plant closures, WoodMac’s Manghani noted.
“With standalone storage, you can end up balancing the grid in a much better way than with solar-plus-storage,” with no complications regarding ITC compliance or other competing interests.
HECO’s proposal notes that its 200-megawatt solicitation on Oahu and its 40-megawatt solicitation on Maui are meant to “provide capacity-like features to the respective island electric grids, by enabling energy from excess capacity and energy in low demand periods to be used during anticipated high demand periods.”
It would also require completion of these projects by 2022 and 2023, respectively, as opposed to a more relaxed 2025 completion date for its renewable projects.
One point of contention among Hawaii energy industry stakeholders was HECO’s decision to include a “utility self-build team” in its plans for standalone storage. HECO’s proposal noted that it “received comments from stakeholders that reflect a sentiment of concern regarding the participation and treatment of the self-build team,” most likely from storage developers concerned that it may be a precursor to the utility rejecting their bids in favor of doing the job itself.
HECO noted that the Hawaii Public Utilities Commission (PUC) is “inclined to prohibit any [self-build option] proposals until the Companies can detail their processes and methodology to evaluate these bids on a fair basis,” and lays out its plan to evaluate “all self-build, affiliate, and independent power producer proposals as equally as possible on an ‘apples to apples’ basis through comparative analysis.”
The utility is also under PUC directive to “ensure that standalone energy storage charged with fossil fuels is the last resort in meeting any capacity needs,” a nod to concerns that grid-charging requirements may lead to displacement of cleaner alternatives.
Grid services: A novel approach
Finally, HECO would seek out more than 140 megawatts of demand-side reduction capacity and 65 megawatts of frequency regulation capacity across Oahu, Maui and Hawaii, the majority of it on Oahu, under its Grid Services RFP proposal.
As WoodMac’s Manghani noted, this portion of the proposal combines well-known concepts of using demand response to reduce peak loads and tapping fast-responding grid assets to help balance grid frequency, but it applies them in novel ways to meet Hawaii’s unique needs.
For example, frequency regulation markets are a part of the portfolio of every U.S. grid operator, he said. “But at the distribution level, I believe this is the first one to go out and ask for regulation services from third-party resources.”
Hawaii has no transmission systems to create markets for frequency regulation, leaving it up to state policy to create new structures to allow third parties to play a role alongside utility investments to solve the problem.
The categories of HECO’s grid services procurement come from its Integrated Demand Response Portfolio Plan, a long-running proceeding meant to create markets to bring demand-side resources to bear as grid assets, and support the rollout of technology and operational infrastructure to make them as reliable as traditional reserves of last resort.
HECO’s proposal notes that the landscape for procuring the fast frequency response for contingency (FFR1) services it needs to replace the AES and Kahului plants has changed in recent months, due to other PUC decisions. For example, earlier this month the PUC issued Order 36410 denying HECO’s request for about $104 million for a “Contingency and Regulating Reserve Battery Energy Storage System,” forcing HECO to review whether it needs to change its Grid Services plans to adjust.
Another PUC decision in July, Order 36406, has led HECO to rewrite its RFP rules for Hawaii Island to allow solar-plus-storage systems to bid to supply FFR1 services, it noted.
Pilot projects in California and elsewhere have shown that solar farms can use their inverters to mitigate frequency and voltage disruptions, and batteries have been doing the job in mid-Atlantic grid operator PJM’s frequency regulation market for almost a decade.