by Jeff St. John
December 19, 2017

Everybody talks about how energy storage needs to be freed to stack different use cases and revenue streams to meet its full potential. But who’s actually doing something about it?

In California, home of the country’s biggest energy storage mandate, state regulators are on the verge of putting together a set of rules  -- 12 of them, to be precise -- that could bring this value-stacking dream closer to reality. 

Over the coming weeks, the California Public Utilities Commission will be considering a vote on a proposed decision that would set these 12 rules in place, along with key definitions of the “multiple benefits and services to the electricity system” that it’s hoping to enable for the state’s growing portfolio of energy storage. 

The proposed decision (PDF) from CPUC Commissioner Carla Peterman represents a final step in a four-year-long proceeding, launched in 2013 to enable the goals of Assembly Bill 2514, the 2010 state law which set the state’s 1.325-gigawatt energy storage mandate in motion. Since then, the state’s three big investor-owned utilities have engaged in multiple rounds of storage procurements, and contracted for hundreds of megawatts' worth of projects ranging in scope from multi-megawatt grid-connected systems to aggregations of distributed, behind-the-meter batteries. 

But these projects have been built around single use cases -- deferring investment in distribution grid infrastructure, providing capacity for utility resource adequacy needs, or bidding its ability to reduce load into state grid operator CAISO’s energy markets. While utilities, storage developers and CAISO have been talking about expanding into multi-use operations, the ability to do so has been limited by the lack of rules in place to allow it. 

Commissioner Peterman’s latest proposed decision, reviewed in last Thursday’s CPUC meeting, would break open those limitations, with implications both for the remaining rounds of procurements called for under AB 2514 starting in 2018, and for broader storage applications in years to come. 

Those possibilities could also extend to already-built storage projects. While the proposal does leave several more complicated issues to future working groups, “there are still a number of different combinations of multi-use storage applications able to occur today using the rules adopted in this decision.” 

This set of rules could open up long-awaited opportunities for utilities to seek out wholesale market opportunities alongside the distribution-level or resource adequacy services they’re getting from storage today. As the proposed decision notes, “most tariffs, contract provisions, and rules assume that a resource will only provide one service. Multiple-use applications present a completely different paradigm.” 

At the same time, the proposal noted: “utilities shall not unreasonably restrict a storage provider’s ability to provide multiple services so long as those services do not conflict with the framework adopted today or the specific contractual terms of the utility’s solicitation." Storage developers in the state have been looking forward to exploring that freedom for some time. The list of parties involved in this CPUC proceeding, including Advanced Microgrid Solutions, Stem, Sunrun, Sunverge and Tesla, indicates the range of interest. 

So what are the 12 rules, and how are they going to change the possibilities for energy storage in California? Let’s take them in rough order, starting with some key definitions. 

Defining storage realms: Customer vs. distribution vs. transmission

There are three types of storage delineated in the proposed decision’s first four rules:  

  • Rule 1. Resources interconnected in the customer domain may provide services in any domain.
  • Rule 2. Resources interconnected in the distribution domain may provide services in all domains except the customer domain, with the possible exception of community storage resources, per Ordering Paragraph 11 of D.17-04-039.
  • Rule 3. Resources interconnected in the transmission domain may provide services in all domains except the customer or distribution domains.

In simple terms, these three rules state that “storage resources may only provide services within their own grid domain, or a higher-level grid domain but not in reverse.” Specifically, customer-sited storage can play across all three domains; distribution-connected can play in its own domain, as well as transmission, but not behind the meter; and transmission-connected can only play in its own domain. 

This is a common-sense distinction. Aggregated behind-the-meter batteries are already playing a role in aiding systemwide services, as with the state's Demand Response Auction Mechanism. But there are far fewer ways in which a transmission system-connected battery could serve the needs of customers interested in energy storage, like managing demand charges or future time-of-use rates, or participating in demand response programs. 

As the decision notes, “these rules are based on current law and practice, and no party opposed these rules.” The one exception being contemplated is for community storage resources -- batteries connected at the distribution level, but serving a customer-facing purpose.

Rule 4 -- “Resources interconnected in any grid domain may provide resource adequacy, transmission and wholesale market services” -- more or less summarizes the previous three rules by noting that resources interconnected in any grid domain may provide resource adequacy and wholesale market services, since they’re all transmission-related. 

Defining storage realms: Reliability vs. non-reliability

Rules 5 through 8 organize storage on a different measure -- whether they’re classified as “reliability services” or as those “crucial to the reliable operation of the electricity system,” versus “non-reliability services.” This is one of the most important issues at stake in this proceeding, because one of its chief goals is to clearly define what services must be provided before others, which will set significant limits on a storage system’s freedom to operate. 

At the same time, it’s a simpler task than it might seem. In the latest revisions to the proposed decision, the CPUC has classified all customer-facing applications as “non-reliability,” and nearly all distribution and transmission-facing applications as "reliability." The sole exception is the ability to provide “imbalance energy” to the CAISO’s real-time imbalance market. 

Rule 5 sets out the fundamental distinction between the two classes of service: “If one of the services provided by a storage resource is a reliability service, then that service must have priority.”

Because all of the non-reliability services in the proposed decision lie in the customer-located domain, this rule is essentially saying that those storage systems or aggregations will have to prioritize distribution or transmission grid-level services over their own customer-facing values. 

But what does “priority” mean for storage that wants to provide more than one reliability service? That definition is provided in Rule 6: “Priority means that a single storage resource may not contract for two or more different reliability services from the same capacity in a single, or multiple, domains, over the same or overlapping time interval for which the resource is committed to perform or be available.” In other words, no storage system can “double-book” its capacity or capability for more than one reliability service at a time. 

The one exception to Rule 6 comes in Rule 7: “A single storage resource may contract for resource adequacy capacity and provide wholesale market reliability services using the same capacity, and over the same time interval.”

This exception lines up with how many of California’s existing energy storage implementations operate today, including much of the storage contracted for by Southern California Edison and San Diego Gas & Electric to mitigate the impacts of the San Onofre nuclear power plant closure, as well as aggregated storage serving in the state’s Demand Response Auction Mechanism.

It also aligns with rules that allow utility-contracted resource adequacy capacity to provide any wholesale market service to fulfill its must-offer obligations: "To restrict this ability, and only for storage, would be unnecessarily restrictive and discriminatory.”  

Many of the energy storage providers commenting on this proceeding would have liked a looser set of rules on these matters. “In place of adopting specific rules prioritizing reliability services, the storage industry generally supports a system of penalties and incentives, or else argue that existing rules and penalties for reliability services are sufficient,” the proposed decision notes. 

However, “because we are just beginning to understand the different types of services that storage can provide, as well as the real-time economic decisions and tradeoffs that come from providing multiple services, we find it prudent to be more restrictive at this time to ensure that safety and reliability are not compromised,” it states. 

At the same time, the revised rules are more flexible than they were in previous versions, the proposal added. “We note that Rule 6 has been refined from the original Report, which stated that no storage resource could provide more than one reliability service, regardless of whether or not the obligation to perform the services occurred over the same time interval,” it wrote. 

The revised proposal does this by breaking out three categories of multiple-use applications: time-differentiated, capacity-differentiated and simultaneous. “This change modifies the structure significantly, as the first type would allow for a single resource to provide multiple reliability services that occur in different time intervals, using the same capacity, thus maintaining reliability.”

Adding this flexibility, however, calls for additional rules to define how it can and can’t be used. That’s one of the goals of Rule 8, which states that “storage providers must clearly demonstrate when contracting for services both the total capacity of the resource, with a guarantee that a certain, distinct capacity be dedicated and available to the reliability service, whether or not the individual devices within an aggregated resource will always be used to provide it.” This rule could also apply to discrete storage systems or aggregations that wish to split up their capacity between different reliability services. 

How to enforce compliance, examine costs, and keep playing fields level

Rules 9 and 11 are about compliance, enforcement and penalties for breaking the other rules.

Rule 9 requires that the contracting utility or entity make its enforcement and penalties known “upfront so that storage providers fully understand, and quantify, any risk in providing services.”

Rule 11 sets a bright line for meeting all availability and performance requirements “to avoid or defer a transmission or distribution asset upgrade.” This doesn’t mean that meeting other service requirements “are simply optional or that any penalty would or should be waived,” only that those two grid deferral use cases are “two categories “singled out because of their importance to reliability.”

Rule 10 is a bit trickier, because it deals with how energy storage providers, contracting utilities, and everyone else involved must share data on the cost of providing service.

Rule 10: “In response to a utility request for offer, the storage provider is required to list any additional services it currently serves, or intends to serve, outside of the solicitation, and take the revenue from these services into consideration when pricing offers to the utilities’ solicitation. In its evaluation of bids, the utilities must take these services into account in evaluating and ranking bid prices.” 

This is a critical new measurement that needs to be made in multi-use applications, the proposed decision notes. “When all providers are providing a single service, cost discipline is typically achieved through competition via the request for offers process for that service,” it wrote. “When providers are providing multiple services, as the rules are designed to allow, there is no guarantee that the ratepayers are getting a fair deal on the cost without requiring reporting or analysis of those costs.”

At the same time, this rule “does not propose or require any CPUC oversight of the earnings of storage developers, nor do we assign that authority to the utilities,” it noted. Nor does it extend to actually setting the rules for how much a storage system is actually compensated for these disparate services. 

Indeed, there’s a world of “double-counting issues, such as double payments, overlapping value streams, or redundant compensation,” that remain to be worked out -- which is why Rule 12 is still an “interim rule:” 

Rule 12: In paying for performance of services, compensation and credit may only be permitted for those services which are incremental or distinct. Services provided must be measurable, and the same service only counted and compensated once to avoid double compensation.

Putting the meat on the bones of this “no double compensation” rule is just one of the tasks before a working group that this proposed decision would create. Another pretty important one will be determining “appropriate metering, measurement, and accounting for Multiple-Use Applications,” starting with the time-differentiated variety, and then moving into capacity-differentiated and simultaneous applications. 

A multi-use wish list for utilities and storage developers

One of the top items on the energy storage industry’s wish list is determining how to account for behind-the-meter storage that also participates in CAISO’s wholesale market. Today, almost all distributed energy storage in the state is participating in the same way that demand response and other “load-modifying” resources do, through programs such as CAISO’s Proxy Demand Resource (PDR).

“No additional metering is required for this type of participation, and BTM storage providers are already participating as a PDR today,” the proposed decision notes. 

The second option, “direct participation in the wholesale market as a non-generator resource (NGR),” is more complicated, because it raises the issue of retail versus wholesale compensation. “Many parties, including but not limited to PG&E, SDG&E and Stem, recommended further discussion for behind-the-meter resources to participate as a NGR, and specifically the accounting methodologies necessary to determine incrementally and to determine the split between wholesale and retail on an on-going operational basis.” 

One of the most pressing issues utilities face as they seek out their 2018 storage mandate procurements is setting up guidelines for “a subset of a distributed energy resource aggregation to provide distribution-level services.”

Deferring or augmenting traditional distribution grid investments is expected to become a major part of the value stack for energy storage and other distributed energy resources (DERs). But CPUC’s framework for valuing DERs for distribution grids, its Distribution Resources Plan and Integrated Distributed Energy Resources proceedings, is still being worked on. 

Both “CPUC and CAISO staff recommend exploring this issue in the working group, based on their assessment that this use case is likely in the very near term,” the proposed decision notes. “The working group should draw from experience in the [Integrated Distributed Energy Resources] pilot and related experience.” 

In the meantime, to help define what the state’s utilities should be seeking out on this front, the proposed decision lays out the following categories of services: 

  • Distribution capacity services are load-modifying or supply services that distributed energy device/resources provide via the dispatch of power output for generators or reduction in load that is capable of reliably and consistently reducing net loading on desired distribution infrastructure.
  • Voltage support services are substation and/or feeder-level dynamic voltage management services provided by an individual device/resource and/or aggregated device/resources capable of dynamically correcting excursions outside voltage limits, as well as supporting conservation voltage reduction strategies in coordination with utility voltage/reactive power control systems.
  • Reliability (back-tie) services are load-modifying or supply services capable of improving local distribution reliability and/or resiliency. Specifically, this service provides a fast reconnection and availability of excess reserves to reduce demand when restoring customers during abnormal configurations.
  • Resiliency (microgrid) services are load-modifying or supply services capable of improving local distribution reliability and/or resiliency. This service provides a fast reconnection and availability of excess reserves to reduce demand when restoring customers during abnormal configurations.
California is the state that's gotten closest to figuring out how to value stack storage. Keep tuned to follow the evolution of these rules.