Comments on Issue Paper

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Boston Energy is supports the CAISO continued efforts evolve its market structure to ensure energy storage can fully participate in the market and the attributes that energy storage bring to the market are fully valued.  In light of the CAISO recent FERC filing proposing to impose a minimum state of charge constraint for energy storage resources, should that proposal be approved at FERC implementing a market based solution within two-years should be given the highest priority by the ISO in this initiative. 

Boston Energy further asks the ISO to prioritize a review of its tariffs and business practice manuals to address charging restrictions utilities are imposing on energy storage resources.  These restrictions are real and active in the market today.  CAISO must ensure that energy storage resources can utilize the ISO’s tools such as OMS and real-time telemetry to properly inform the ISO and its market systems of the restrictions being imposed on the resource by the utility.  These restrictions are dynamic and can change every 5-seconds in some instances.  Having the ability to utilize the ISO’s current set of tools to communicate these restrictions to the ISO is crucial to ensuring feasible market schedules and dispatch. 

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

At this time Boston Energy doesn't have a position on the relative priority of the items listed above.  We encourage the ISO to continue to explore with stakeholders the options for expanding the multi-interval optimization for all resources participating in the ISO’s market.  We understand the complexity of this issue but it has been discussed multiple times over the past few years and supported by a wide range of stakeholders.  More transparency from the ISO on the challenges of making such changes would be helpful.  In the spirit of transparency, the ISO should start posting immediately the advisory in advance of the bid submission deadline so scheduling coordinators can take these prices into account when developing their bidding strategy.

With regards to bid cost recovery rules for storage resources, Boston Energy doesn’t have a position at this time.  However, this issue has been raised multiple times over the past few years by energy storage resources as an important topic for the ISO to review.  To support this issue Boston Energy asks the ISO to quantify the magnitude and frequency of advisory prices causing uneconomic binding schedules for energy storage resources.  Have this information will help in the prioritization process. 

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Should FERC approve the ISO minimum state of charge proposal eliminating the artificial and non-market based constraint within two-years this item should be the highest priority of this initiative.  CAISO has requested in its FERC filing a two-year sunset and CAISO should hold true to that statement and work with stakeholders to implement a market based solution in that timeframe.  Generally, we believe that the market is best served by giving resources the ability to respond to real-time prices, which may be different than those published in the day-ahead market. This is one of the true bedrock elements of LMP market designs and the CAISO should not lose sight of the true benefits of the three settlement market design it has chosen to implement.  Any market design change intended to help manage the energy storage resources in the real-time market must be market-based and incentives rather than use artificially constraints that lead to inefficient outcomes that could potentially result in reliability issues should the ISO’s day-ahead market solution not match real-time market conditions.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

Boston Energy has no comments on this item at this time.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

Boston Energy believes strongly proper price signals that reflect actual market conditions should nearly eliminate the need for the ISO to use exceptional dispatch.  In instances where exceptional dispatches are needed the ISO should look to determine if a market based solution can be implemented to  eliminate recurring exceptional dispatches.  Should the ISO use exceptional dispatches to manage an energy storage resources state of charge, the energy storage resource should be held financially harmless if the ISO’s action lead to real-time market revenue lost opportunity.  Lost opportunity costs are widely used in ISO markets and should be considered if the ISO intends to utilize exceptional dispatches to manage an resources state of charge.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

As highlighted above Boston Energy asks the ISO to prioritize a review of its tariffs and business practice manuals to address charging restrictions utilities are imposing on energy storage resources.  These restrictions are dynamic and can change every 5-seconds in some cases.  We ask the ISO allow scheduling coordinators to utilize the ISO existing tools (OMS and Telemetry) to communicate to the ISO the charge restriction being imposed on the resource.  The ability to utilize these tools will allow for proper market solutions and schedules.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

CalCCA thanks the California Independent System Operator (CAISO) for taking on this initiative.  As the capacity from storage continues to increase, it will be important to ensure that the energy markets function as efficiently as possible in light of the technological differences presented by storage resources.  CalCCA generally believes that the CAISO has appropriately scoped this initiative and provides the following comments for consideration as the CAISO continues its policy development.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Real Time Market Enhancements

Marginal Cost

The accurate reflection of marginal cost is critical to enabling efficient market operation.  The evolution of storage is continuing, and more is now known about costs, including cycling costs, as the market expands.  CalCCA supports the CAISO recommendation to continue investigating cycling costs, including the relative cycling cost dependent on the state of charge as outlined in the Issue Paper at pages 6-7, in the discussion of bid submission timelines.  In addition, if the CAISO cannot extend the real-time market optimization horizon sufficiently forward to fully capture a charge/discharge cycle, the CAISO should consider an end of horizon opportunity cost as suggested by the Market Surveillance Committee.  There is merit in such an approach; however, further evaluation is necessary to determine if such a mechanism is appropriate and if it can sufficiently address the need to constrain resources to ensure their state of charge.

The discussion of marginal cost would benefit from a minor expansion: the inclusion of physical resource characteristics within the system optimization.  While some storage technologies can rapidly change from charging to discharging, this is not the case for all storage technologies and therefore should not be the universal expectation (e.g., some storage resources must “rest” for a period of time after fully charging or discharging).  Modeling of such resources with inappropriate operational assumptions will place risk on the bidder that can and should be avoided.  The CAISO should thus ensure through this process not only that the marginal cost of the resource is accurately reflected, but that the system algorithms and the master file data accurately represent the resource’s capability.

Advisory Intervals and Spread Bidding

It is important that the CAISO recognize the shortcomings of using real-time market advisory intervals when the time horizon of the real-time market is too short to be able to optimize storage charging and discharging.  There is general consensus that storage will play an important role in energy arbitrage in the future.  In a 65-minute real-time environment, the forward look is simply not long enough to enable meaningful energy arbitrage.  If given only these short optimization horizons and associated price differentials within them, the dispatch of such resources may come to look like that of regulating and spinning resources, rather than efficient optimization of storage resource charging and discharging.  Similarly, spread bidding may make little sense in a real-time environment where the spread may not be realized in any 65-minute look, but with an expanded look may be a viable method such as in the day-ahead market.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Ensuring State of Charge

The CAISO must examine closely the pros and cons of a long-term constraint on storage resources that is based upon elements other than price.  Significant concern has been expressed about the potential for the day-ahead market to optimize resources and schedule storage to meet net load peaks only to have the real-time market change this optimization and result in reliability concerns.  This problem is a theoretical concern that tends to ignore the realities of prices and profit maximization by storage resources.  It also ignores the potential for not being able to address another reliability event if constraints are put in place that limit the flexibility and availability of storage resources.

Storage is not so dissimilar to a use-limited resource that must make choices between honoring its day-ahead award or deviating when real-time prices do not reflect the day-ahead expectations.  Economically, a storage resource will be faced with the following decisions when the real-time price is trending significantly higher than the day-ahead prices and awards when considering deviating from their day-ahead award:

1. Is there enough time and an expectation of prices becoming low enough to deviate now (assuming a discharging real-time event) while still enabling a sufficient state of charge to honor their day-ahead award?
2. Is the real-time price that is presently higher than the day-ahead price going to be sustained or is it a temporary spike that can be expected to subside later?

Clearly if the answer to #1 is yes, then the resource should follow the real-time price signal as doing so would make incremental profit and not prevent the resource from honoring their day-ahead award.  If the answer to #1 is no, then the resource must address the second question.  It is possible that the elevated prices are systematic and will remain higher than the day-ahead prices in all hours.  If that is the case, then it may not be profit maximizing to discharge now and deviate from the day-ahead awards as the buyback of the day-ahead award in the real-time may result in less total profit for the resource.  On the other hand, if the price spike is due to a local issue resulting in momentary high prices, signifying a reliability need in that area, it may not only be profitable to follow the real-time price but may also be the best option for reliability. 

It is easy to conceive of a case in which a storage device is located in an area where prices are high due to an outage that is temporary in nature but is driving high prices.  It is further possible that aside from the storage device, the next available resource has a lengthy start-up time.  In such a situation, the storage device could address the immediate reliability need while the CAISO starts the other resource that can then provide for the later evening peak.  Choosing which reliability event the CAISO should protect for is not an easy decision to make and should be investigated closely.

This investigation should consider that market participants are savvy or will become so quickly in the energy market and are likely very capable of make decisions to the two questions above that are in their best interest.  As long as market prices depict the reliability needs on the grid, then the resource providers can make the best decisions on how to address such issues. If resource providers make the wrong choices about their offer prices, they will face financial consequences depending on the level of the future interval prices in comparison to the short-term interval prices.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

CalCCA offers no opinion at this time.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

CalCCA offers no opinion at this time.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

Policy Direction

The CAISO listed in the Issue Paper three potential policy issues for consideration.  The expansion of the real-time market to include a further look ahead, while it could avoid the need for workarounds to address the shortcomings of a limited real-time market optimization, appears to have been dismissed as not viable with current optimization tools.  The CAISO should provide additional information within this initiative of the nature of the issues that make a longer look ahead infeasible as well as what would need to be done to make such a look ahead feasible.  Such a mechanism would not only make market optimization easier (e.g., limit exceptional dispatch, ensure sufficient state of charge, etc.) but would also provide information to resources to better enable them to offer their resources in the real-time market.  The other two elements involve the potential for a new product and the potential for constraining resources in the real-time market to preserve their day-ahead award.

If the CAISO pursues a new product, the first question that must be asked is what entity(ies) creates the demand for the product?  This is not a causation question but one of bidding.  Will it be those serving load that submit demand bids or will the demand be created by the CAISO based upon its own criteria?  The former is necessary for a properly functioning market while the latter may by chance reflect the willing interaction of buyers and sellers but is equally likely not to.  This issue is raised in a number of environments (e.g., Residual Unit Commitment, Flexible Ramping, corrective capacity, etc.).  In each of these cases, there is always a concern over the ability of the CAISO to accurately reflect the true needs of the system rather than having buyers reflect their own needs such as is done in the energy market. 

With regard to constraints on resources based upon day-ahead expectations, as noted above, such a mechanism (i) may work in some cases, (ii) may not be constraining to what the resource owner would have done based on their own decision, or (iii) could stand in the way of the CAISO using a resource for reliability purposes due to an artificial constraint preventing such an action.  Consideration of a constraint should therefore be an element that is used sparingly to protect against events that have a high probability of reliability impacts if not enacted.

The CAISO does note in the Issue Paper that the CAISO can exceptionally dispatch resources including an instruction for a storage device not to discharge.  In such an event, there will be foregone profit by the resource that cannot be ignored as well as the fact that the exceptional dispatch is not reflected in the market prices. 

The Issue Paper also discusses the potential for different charging rates depending upon the state of charge of the resource.  This is an element that also deserves further examination.

Finally, the CAISO should consider whether there is additional information that could be made available to the market to allow resources to make better decisions without providing information that would present the potential for the exercise of market power.  For example, if the CAISO is concerned that the fleet of storage resources will not be available for the net load peak, then perhaps examining the potential to publish the aggregate state of charge will help storage resources to make decisions about how likely the current real-time price is to continue to remain high.  This of course should be tempered by consideration of the ability of any individual entity or small group of entities to use the information to manipulate the market.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

The California Energy Storage Alliance (CESA) recognizes the leadership of the CAISO with regard to revising market design elements that currently overlook the specific operational characteristics of energy storage assets. As the ISO notes within the Issue Paper, energy storage is set to be a key resource type to meet our peak net loads, particularly considering California’s energy and climate targets.[1] Thus, addressing issues related to energy storage participation is timely and necessary in order to fully integrate these resources into the market, leverage the flexibility of these assets to maintain grid reliability, support an efficient marketplace that delivers savings to ratepayers, and maximize their use to achieve the State’s clean energy goals.[2]

The present effort is the result of several months of active engagement between ISO staff and the energy storage stakeholder community within the Resource Adequacy (RA) Enhancements Initiative. In RA Enhancements, ISO staff proposed a means to ensure RA energy storage assets with a Day-Ahead (DA) schedule would have sufficient state-of-charge (SOC) to meet said schedule: the minimum state-of-charge (MSOC) requirement. In said initiative, energy storage stakeholders noted a series of necessary improvements to the CAISO’s treatment of storage assets that could, jointly, minimize the need for a tool akin the MSOC and ensure proper compensation of opportunity costs in the case the MSOC was applied.[3] In response, ISO staff recommended adopting an MSOC requirement designed to be applied only in critical days (i.e. those with a residual unit commitment (RUC) process unfeasibility), paired with a two-year sunset clause. Notably, the adopted MSOC proposal does not include a means to compensate energy storage assets for potentially forgone revenues resultant from its application. In the RA Enhancements Phase 1 Final Proposal, the ISO noted that a stakeholder initiative would be launched to identify a successor mechanism to the MSOC, and address other issues identified by storage stakeholders.

As a result of its procedural background, the Energy Storage Enhancements Initiative encompasses a wide array of issues: some of them are related to energy storage participation broadly, while others are directly related to the replacement of the MSOC. While CESA generally agrees with the issues put forth by the ISO within the Issue Paper, the establishment of a form of prioritization is necessary to ensure timely and actionable outcomes.

CESA supports prioritizing elements of the CAISO’s market design that, due to their formulation, currently hinder the economic dispatch and utilization of storage assets; namely, the multi-interval optimization (MIO) tool, the application of “spread bidding” and bid-cost recovery (BCR) to energy storage, and the ability to represent marginal costs related to cycling and SOC. Revising these elements can ensure energy storage is dispatched feasibly and economically, and that all the services rendered by it are properly compensated. Moreover, the revision of these elements could influence the scope and formulation of any MSOC successor. Once these elements are considered, CESA recommends assessing potential replacements for the MSOC requirement, which include the extension of the Real-Time (RT) market’s optimization horizon, the establishment of an energy shifting product, or the creation of a biddable stored energy product. As such, CESA proposes dividing this initiative into two phases: Phases 1 and 2. CESA’s prioritization proposal is summarized in the following procedural calendar and justified in the answers for Questions 2-5.

• Phase 1: Energy Storage Management, Operation, and Compensation (targeting December 2021 Board of Governors meeting)
  • Reflection of SOC and cycling within marginal costs
  • Revision of the MIO tool
  • Spread bidding revision
  • Revision of the BCR mechanism
  • Revision of variable charging rates
  • Revisions to exceptional dispatch
• Phase 2: Ensuring State-of-Charge (targeting Q1 2022 Board of Governors meeting)
  • Understanding the challenges of extending the RT market’s optimization horizon
  • Scoping the energy shifting product

[1] See Issue Paper at 10.

[2] See Issue Paper, at 3.

[3] See CESA, “Comments of the California Energy Storage Alliance on the RA Enhancements Initiative Phase 1 Draft Final Proposal.”

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

CESA strongly supports the inclusion of issues considered as “related to representing marginal costs”. This collection of issues, which include spread bidding, modifying the MIO tool and revising the BCR mechanism in addition to improving the representation of marginal costs, could be better understood as market elements that require modifications to properly utilize storage assets available to the ISO. As such, addressing these issues will have a substantial effect on the ability of asset owners to properly manage their resource, and of the ISO to maximize their utilization. In this context, CESA deems these topics should have the highest priority. In the following paragraphs, CESA elaborates on the urgency to address each of the topics contained within this section and shares some preliminary solutions for the ISO to consider.

• Spread bidding: The ISO must fundamentally revise its RT market structure to properly represent the bid curves submitted by asset operators. Currently, the ISO’s RT market captures the bid curves supplied by storage operators and oftentimes dispatches them according to the implied spread between charge and discharge bids. CESA notes that this method carries potential risks as some resources might receive uneconomic instructions regardless of the expected spread. Moreover, this limits the owner’s management of their asset and introduces substantial financial and contractual risks, both key components to secure financing. Consider a storage asset that has a revenue obligation to discharge every time prices hit a particular threshold; this cannot be guaranteed given the current spread bidding practice. This practice, in turn, increases the likelihood RA-providing resources participating in the RT market would find themselves unable to comply with DA schedules, the precise situation the ISO has sought to minimize with the MSOC. Hence, CESA urges the ISO to reform the RT market optimization tools to act based on specific bid points and not expected spreads.
• The Multi-Interval Optimization (MIO) Tool: The current MIO tool does not adequately process the bid curves submitted by storage assets. This is due to the tool’s focus on conventional resources. This focus can result in high power charge commands when the storage asset is reaching full SOC, consecutive charging commands despite the asset being at 100% SOC, and discharge directives despite the resource having a discharge schedule just ahead of the MIO horizon, among other operational issues. As a result, the MIO software’s operation can lead to undesired discharge in intervals prior to the evening peak, potentially causing reliability concerns similar to the ones the ISO sought to mitigate with the MSOC. In order to address the limitations of MIO, CESA supports the proposals made by LS Power within the RA Enhancements Initiative. LS Power offered two solutions for this issue: (1) link real-time dispatch (RTD) instructions directly to the binding interval and not the advisory intervals; or, (2) reduce the number of advisory intervals for NGRs from 13 to two or three. CESA agrees with these recommendations as resource owners already face strong incentives to align their behavior with reliability-driven outcomes (i.e. abide to their DA schedules given the penalties associated with them).
• The Bid-Cost Recovery (BCR) Mechanism: Currently, BCR is calculated using settled cost and revenue values from the DA and RT markets. These values are netted across the day (24 hours) for the RT market. This mechanism is incapable of identifying circumstances when a storage asset would “fall short” on its revenue requirements since, compounded with the practice of spread bidding, the netting of the BCR over a 24 hour period would invariably result in no need for compensation. Given this limitation, energy storage stakeholders have proposed to the ISO to either net costs and revenues for BCR over a charge/discharge cycle (e.g. 8-9 hours for a 4-hour battery) or netting all costs to charge the resource with the revenue from discharging to ensure the bid spread for the resource is covered. Prima facie, CESA considers the latter solution would be the most adequate to properly compensate all types of energy storage assets, regardless of their duration.

Reflection of SOC and cycling within marginal costs: CESA appreciates the ISO’s consideration of means to improve the reflection of several costs incurred by energy storage resources. In the Issue Paper, the ISO notes that several storage stakeholders have advocated for a modification that would allow them to alter their bids based on their resource’s SOC. CESA, within the RA Enhancements Initiative, proposed to adapt the policy adopted by the Electric Reliability Council of Texas (ERCOT) which allows storage to update its bid curves at every five-minute interval, rather than submitting bid curves for all five-minute intervals for the following hour. CESA recognizes the ISO’s consideration of this proposal; however, given the urgency to address this issue and that the ISO notes this does not represent a feasible near-term solution, CESA favors evaluating the proposal to allow storage assets to submit multiple RT market bid curves that are dependent on SOC. Regarding the issue of cycling, CESA considers this could be incorporated in the same fashion: resources willing to cycle more than once could submit a series of bids curves dependent on SOC and the current cycle. In addition, CESA recommends that CAISO explore adding the reporting of CAISO’s modeled SOC value for NGR battery assets. This would enable asset operators to understand the energy budget CASIO believes an asset will have.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

As noted in CESA’s responses to Question 1, CESA appreciates the ISO’s commitment to timely find a successor for the MSOC requirement. Given the ISO’s intention to revise market elements that conflict with the operational nature of energy storage resources, the revision of these elements could influence the scope and formulation of any MSOC successor. As such, CESA recommends that the ISO assess potential MSOC replacements once revisions to other market elements are considered. Within the Issue Paper, the ISO puts forth four alternatives to ensuring SOC: the extension of the Real-Time (RT) market’s optimization horizon, the establishment of an energy shifting product, or the creation of a biddable stored energy product.

Within the Issue Paper, the ISO notes that the first potential solution, extending the RT market, is not technologically feasible. While CESA respects the ISO’s conclusions, more clarity regarding the costs and challenges of expanding the RT optimization horizon would enable stakeholders to better evaluate this proposal and provide substantive feedback on if and when the ISO should pursue this option. In light of the complexities linked with extending the RT optimization horizon and the need to replace the MSOC requirement, CESA recommends the ISO focus its attention on the second proposed solution: creating an energy shifting product.

In the RA Enhancements Initiative, CESA advocated for the creation of an energy shifting product as it would directly address the issue that ISO has tried to solve with its MSOC: incenting storage assets to engage in daily diurnal energy arbitrage.^[1] This product could complement market signals sent by energy and ancillary services pricing to incent shifting behavior while properly compensating storage assets for the services they render. CESA supports this potential policy direction, as its effects could ease storage development, not just operation, and would welcome further engagement around the design of this product. In particular:

• How could it affect existing market products (energy and ancillary services)?
• How could it support co-optimization across products and time horizons?
• How could it create efficiencies for the market and economic value for ratepayers?

It is also worth noting that the CAISO’s Day-Ahead Market Enhancements (DAME) considers new market products and other modifications that could potentially affect the development of an energy shifting product. As such, CESA requests the ISO considers the merits of evaluating the development of this product within or in parallel of the DAME initiative, in order to minimize divergent policy development and enhance the synergies across proposed solutions. In addition, CESA recommends that CAISO explore adding regulation throughput considerations to its SOC management. This could enable CAISO to consider the impact of regulation dispatches on schedule feasibility.

Currently, the CPUC estimates that around 11 GW of energy storage will be needed by 2030 in order to meet California’s ambitious energy goals. Given the growth of energy storage is directly related to the State’s commitment to decarbonization, the ISO has the opportunity to clearly define a product that could further facilitate the market integration of storage resources set to provide daily solar energy shifting. As such, the establishment of an energy shifting product could both ease the ISO’s concerns regarding ensuring SOC and facilitate the development and contracting of resources that will be certainly required to maintain reliability and achieve California’s energy and environmental goals.

[1] See CESA, “Comments of the California Energy Storage Alliance on the RA Enhancements Initiative Phase 1 Draft Final Proposal.”

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

CESA supports the inclusion of the variable charging rates issue within the scope of this Initiative. As noted in our response to Question 2, currently, the ISO may issue dispatch instructions that direct a battery storage resource to aggressively charge in periods where the SOC is nearing 100%. These instructions are often unfeasible since some storage technologies like lithium-ion, alike most electrochemical storage technologies, have diminishing charge acceptance rates. In essence, one can think of the charge rate as a function with an inverse relationship with the SOC. Given this physical consideration, the ISO is correct in including this issue in the scope.

In order to address this challenge, the ISO asks stakeholders if options currently available on the resource’s side could be used. CESA considers this could be resolved by allowing resources to communicate a charging rate curve by SOC as part of the Masterfile. This would allow resources to communicate the fact that its charging rate varies depending on the SOC range it is at.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

CESA appreciates the ISO’s consideration of exceptional dispatch within the Issue Paper. As noted by the ISO there are two potential types of exceptional dispatch that merit revision with regards to their application for energy storage. First, the case when an asset is exceptionally dispatched at 0 MW, essentially retaining SOC. CESA considers that compensation for opportunity costs under this scenario is warranted.

The second case describes a situation in which operators specifically procure SOC rather than a target MW amount. While CESA can appreciate the potential for this case, the development of an energy shifting product could substantially decrease the need for such a form of exceptional dispatch (ED). Hence, CESA recommends the ISO focus this initiative’s efforts on developing an ED compensation framework that properly characterizes the opportunity costs forgone by storage asset that has been directed to hold SOC.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

CESA notes that the grid-scale storage resources being deployed onto the CAISO grid also include long duration and emerging technologies beyond lithium-ion batteries. As the ISO considers how to best integrate storage resources through this initiative and others, it should ensure that market changes consider the impact to all types of storage resources.   

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Please see DMM comments at following location: 

http://www.caiso.com/Documents/DMM-Comments-on-Storage-Enhancements-Issue-Paper-May-20-2021.pdf

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Please see the link to DMM comments in response to question 1.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Please see the link to DMM comments in response to question 1.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

Please see the link to DMM comments in response to question 1.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

Please see the link to DMM comments in response to question 1.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

Please see the link to DMM comments in response to question 1.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

The California Large Energy Consumers Association (CLECA) appreciates the fact that the CAISO is examining the best use of storage to meet capacity needs during times of system stress.  During the August and September 2020 heatwaves, there were multiple occasions when storage was charging during Warnings, and even during Stage 2 and Stage 3 Emergencies; these occasions were shown in the Final Root Cause Analysis.[1]  It is our expectation that storage will be best positioned to assist in meeting the energy needs during a period of system stress, not contribute to system stress. 

One item missing from the current scope for this stakeholder initiative is the interaction between the utilization of storage and the dispatch of reliability demand response programs.  On Saturday, August 15, 2020, during a grid Warning, the Base Interruptible Program was called at 15:00 which is about same time grid connected batteries started charging. We would encourage consideration of whether and under what conditions it would be appropriate or inappropriate to utilize reliability demand response to facilitate the charging of storage.  

[1] Final Root Cause Analysis, Figure B.27 and Figure B.28, at 112. 

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

The CAISO’s effort to start this initiative is appropriate and timely. CPUC staff particularly note the importance of the Exceptional Dispatch and Bid Cost Recovery (BCR) efforts as very positive steps that can lead to improvement in the market. Staff anticipate significant cooperation with the CAISO on those issues.

CPUC staff believe CAISO’s first priority should be ensuring efficient dispatch of storage resources in the real time market. The largest issue for energy storage resources in the CAISO markets is the inability of the current real time market to effectively manage both charge and discharge sides of the storage transaction. The current real time market design relies on market participants guessing future price outcomes to formulate bids for storage resources long before market run time. Market operations could be more efficient if instead the market scheduled dispatch based on actual costs and on the market’s own up to date, centrally forecast prices.

The majority of other problems (including Bid Cost Recovery issues and shortcomings of the current multi-interval optimization sited by some market participants) stem from this limitation of the real time market. Accordingly, we continue to recommend that CAISO focus on improvements to the real time market. In particular, CPUC staff would like to hear more on the following subjects:

-CAISO has said that extending the real time market look ahead to 6 hours is technologically infeasible.  Is there some amount that the look ahead could be extended? Could additional time periods be added to the optimization in a way that is similar to how EIM areas are added? If not, how is this expansion different?

-Does the CAISO have a plan as to when the forward look out will be feasible to extend? Given the state’s decarbonization goals, energy storage resources, in conjunction with renewable energy production, are likely to be a large and essential part of the California Grid in the near future. In the last year, CPUC ordered procurements have resulted in more than 1000 MW of oncoming energy storage. Significantly more storage procurement is expected to result from an upcoming CPUC medium-term integrated resource planning decision.  Has the CAISO started to plan for this by investing in operational capability consistent with the vision of optimizing storage resources in real time?

-CPUC staff suggest that in addition to the focus on ensuring that resources can provide power for the net peak, CAISO also examine whether any of the proposals will help the market to account for both charging and discharging of storage simultaneously.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Marginal costs:

CPUC staff support the inclusion of all verifiable marginal costs in the default bids and market bids of energy storage resources. To the extent that this may require additional development of Default energy bid parameters, staff support that development.

Bid Cost Recovery:

CPUC staff thank the CAISO for including this important issue in this initiative. With thoughtful cooperation Staff believe that significant improvements to Bid Cost Recovery and Energy Storage participation can be achieved.

In general principle, staff believe that the key to Bid Cost Recovery (BCR) is keeping costs and revenue together in the same netting account. For example, CAISO’s current BCR framework attempts to keep real time costs and revenue separate from day ahead costs and revenue. In the case of energy storage, it would make sense to keep the costs to charge an energy storage resource together with the revenue earned from discharging that same energy. One possibility would be to keep the daily framework and count the energy that had been discharged during the day, and the revenue earned from that discharge. Then to count costs, the calculation would have to count backwards until it had accounted for the right amount of energy (including efficiency losses) and the costs to consume that energy into the resource. This may involve stretching the costs into the previous day, or even further. CPUC staff look forward to continuing to work on BCR for energy storage resources in cooperation with CAISO staff throughout this initiative.

Multi interval optimization and spread bidding:

CPUC staff do not recommend that CAISO pursue changes to multi-interval optimization. Instead, we recommend CAISO pursue including a more explicit spread bidding option in the real time market.

CPUC staff continue to view storage resource operations as being best represented in the market by a spread bid, or a bid that offers to charge and discharge energy based the difference between costs to charge and earnings for discharge. This view has been discussed previously, for example in our June 15, 2020 comments on the Energy Storage and Distributed Energy Resources Phase 4 initiative. At least one market participant has expressed interest in not having a spread bid and not being scheduled by the CAISO’s advanced multi-interval optimization, but instead being scheduled for each interval independently based solely on their bid for that interval.  To date, market participants that express interest in single interval dispatch have not provided an explanation of how single interval dispatch would be more advantageous to the system. CPUC staff believe that working to improve accuracy and compensation for storage resources within the current multi-interval optimization would be much more effective than to forego the advantages represented in the CAISO optimization by ending use of the forward outlook for dispatching storage resources.

Additionally, the desire to pick a strike price rather than a spread suggests that resource operators may not be bidding their actual marginal cost at times. Because the resource operators may not know the price that they will pay for energy that they later discharge, it could be difficult to represent real marginal costs. Whenever the market normalizes bidding that is not a realistic representation of marginal cost, concerns about manipulation should arise. For example, it may be harder to detect or track bids that are made in an attempt to impact CRR prices or to hedge risks taken in trades that occur outside of the CAISO markets. CPUC staff understand that parties involved in this stakeholder initiative may have many different business plans but emphasize that CAISO’s market design should focus on facilitating competitive market behavior while ensuring reliability.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

CPUC staff is unable to provide a well-informed answer to this question at this time. Staff currently understand that a tool like this is under development as part of the ESDER 4 implementation. As such, staff are unclear on how this suggestion is different or would constitute additional functions for the end of hour state of charge tool.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

CPUC staff support the inclusion of this topic as an item for discussion and investigation to determine whether it represents a real physical constraint or cost for storage resources. Efficient dispatch will be best achieved when all costs and abilities of energy storage resources are represented in the market. To the extent that charging rates or wear related to charging rates constitute real constraints or costs on energy storage resources they should be represented in CAISO’s optimization. Staff would like to learn more about this, particularly about how it applies to different types of energy storage resources, and whether cooling systems or other capital can impact the cost or constraints represented in variable charging rates.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

Yes, staff believe CAISO should include modifications to current Exceptional Dispatch practices for storage resources as part of this initiative. As storage becomes a larger part of the market, it will be essential for CAISO to effectively manage all parts of energy storage resources participation, including exceptional dispatches.

Just like the market dispatches, exceptional dispatches can work most efficiently if they envision both sides of the storage transaction. In this case, issuing exceptional dispatches to charge, without a corresponding discharge, can lead to increased generation usage and GHG emissions, and inefficient pricing outcomes. We suggest the following topics as a start of discussion for Exceptional Dispatches for storage:

-should exceptional dispatches include charge and discharge?

-how should charge and discharge exceptional dispatches impact market pricing? For example, if we go with one sided exceptional dispatches for charging then doesn’t this have a potentially downward impact on prices later if the charge is not taken on bid? This may be different than regular exceptional dispatches for energy as the price impact could come at a time when there is not a significant need on the grid.

-should storage exceptional dispatches be specified as charging MW or state of charge? Or both? Would it be best if operators had multiple options?

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

Scarcity pricing should be out of scope:

CPUC staff do not believe that scarcity pricing is a relevant tool for solving any issues with real time dispatch of energy storage resources. The problem is not related to scarcity, but to the limited look ahead timeframe of the CAISO optimization. If the CAISO optimization were able to schedule both charging and discharging of storage resources simultaneously, then it would be more likely to charge resources for optimal discharge in the real time market. Scarcity pricing could increase costs for load but would not impact storage resources with day ahead schedules because of the CAISO’s reasonable and necessary bid cost recovery rules. It is not clear that increased scarcity pricing will lead to better dispatch, as the dispatch issue is not caused by bidding behavior necessarily but by the short forward look out. Accordingly, CPUC recommend that the CAISO does not pursue scarcity pricing in this initiative.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

The Public Advocates Office at the California Public Utilities Commission (Cal Advocates) supports the objectives of the California Independent System Operator’s (CAISO) Energy Storage Enhancements Initiative to ensure that energy storage will provide needed system reliability in the most efficient and cost-effective manner.  Cal Advocates also supports the CAISO’s objective to “maximize [energy storage’s] use and effectiveness to achieve clean energy goals.”[1]

In general, Cal Advocates believes that market-based mechanisms such as the proposed energy shift and biddable state-of-charge (SOC) products should be prioritized over market constraints such as the minimum SOC.  Sufficient SOC from energy storage in the real-time market is necessary for reliability, especially when the grid is stressed.  To address this concern, the CAISO has proposed a minimum SOC requirement for storage during stressed conditions.[2]   In contrast to a constraint on SOC, biddable market-based products will allow the CAISO to procure needed energy or SOC by conducting price discovery and managing the market in the most efficient and cost-effective manner, to the benefit of both ratepayers and resource owners.  Cal Advocates also agrees that more accurate representation of the marginal costs of energy storage is necessary to achieve efficient market outcomes.

Cal Advocates agrees with other stakeholders that re-formulating the real-time market optimization would be the best solution to address the issues identified in the initiative.  The reformulation would incorporate sufficient look-ahead to capture both the charging and discharging cycles for energy storage and to manage the storage fleet’s SOC going into critical peak hours.  However, Cal Advocates agrees with the CAISO that this approach is not technically feasible at this time and cannot be achieved simply with more computing power.  Cal Advocates does not support the recommendation made by other stakeholders to conduct a cost-benefit analysis for implementing expanded real-time look-ahead, as such an analysis would provide limited value given the technical barriers involved.

Cal Advocates recommends that the CAISO conduct quantitative market-simulation based studies to compare how the alternative policy proposals best support system reliability, are cost-effective, and reduce greenhouse gas (GHG) emissions.  The CAISO should define a set of metrics for reliability, cost-effectiveness, and emissions for comparing the proposed alternatives in the production simulations studies.

Cal Advocates has additional concerns regarding the impact of the various policy alternatives on GHG emissions, as discussed below in response to question 6.

[1] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, p. 3.

[2] Resource Adequacy Enhancements Final Proposal – Phase 1, March 23, 2021, p. 23.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Cal Advocates agrees that an accurate representation of each component of storage’s marginal costs associated with cycling is needed to achieve efficient storage dispatch.  This topic should be included in the initiative scope.

Cal Advocates understands that some resource owners have indicated that “spread” bidding in the real-time market does not provide them with sufficient control over resource charging and discharging and thus insufficient ability to control the SOC, which may result in sub-optimal dispatch.[1]  However, it is not clear that discrete bids would provide advantages over spread bidding.  Exclusively employing discrete bids could lead to missed opportunities for cost-effective energy storage charging and discharging and cost-effective provision of reliability services.   Furthermore, based upon the discussion in the stakeholder call, it does not seem that this is a general industry concern.  The optional biddable end-of-hour SOC parameter that will be implemented in the Fall of 2021 may be sufficient to address any remaining stakeholder concerns.  Cal Advocates recommends that the CAISO continue monitoring the performance of spread bidding and continue to study the conditions under which spread bidding may result in sub-optimal dispatch.

Cal Advocates agrees that revisions to bid cost recovery (BCR) for storage resources should be considered in the initiative.  Energy storage’s unique cycling costs and energy-limited operation may not be appropriately captured using the current BCR rules that net costs and revenues over a single twenty-four-hour period.  BCR should be based on the net costs of charging and discharging over a sufficiently long period to capture a full charge/discharge cycle.   Because energy storage may be dispatched to discharge energy that was charged during the previous day, it may be necessary to extend the BCR period beyond a single day   Furthermore, Cal Advocates does not agree that BCR should be based on a “typical 8-to-9-hour cycle” for four-hour duration storage.  Rather, the rules should be sufficiently general to accommodate longer duration storage.

[1] Energy storage bids into CAISO day-ahead and real-time energy markets with discrete bids at which it is willing charge and discharge.  In addition, a spread bid considers the difference between the offer price to charge and discharge relative to the difference between market prices between two given intervals.  If the difference, or spread, between the offered discharging and charging prices for a given interval is equal to or greater than the difference between market prices for the current and some future interval, the resource may be dispatched to charge or discharge in those intervals.   In this case, the resource may be dispatched even though its discrete offered charging (discharging) price is below (above) the market price.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Cal Advocates supports market-based approaches for ensuring SOC in preference to market constraints, such as the minimum SOC.  The CAISO should prioritize consideration of the biddable SOC product the energy shift product, as well as other market-based solutions.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

Cal Advocates agrees that variable-rate energy storage charging is an important issue that should be included within the initiative’s scope.  CAISO market systems do not currently model variable rates of charge, but rather assume a constant charging rate regardless of SOC.  Because lithium-ion batteries require more time to charge when at higher SOC than at lower SOC, there may be a greater risk to system reliability if storage charging begins too late in the day to meet peak system needs.  Resource owners should not be required to oversize storage projects to accommodate a deficiency in CAISO market systems, as is suggested in the Issue Paper.[1]  Oversizing storage will result in additional costs for ratepayers associated with the unused capacity.  Rather, the CAISO should explore revising its market systems to accurately model storage charging.

In addition, Cal Advocates agrees with the proposal to allow resource owners to submit multiple real-time bid curves that are dependent upon SOC.[2]  Doing so would more accurately model storage marginal costs and result in more efficient market outcomes.

[1] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, p. 16.

[2] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, p. 7.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

Exceptional dispatch may become more frequent as energy storage penetration accelerates and the market becomes more reliant on storage.  Resources that are exceptionally dispatched receive the higher of their bid price and the prevailing market price.[1]   Resource owners could submit very high bids in anticipation of the need for exceptional dispatches.   Frequent exceptional dispatch may thus lead to greater costs to ratepayers.  The CAISO should prioritize developing market rules and policies to mitigate the need for exceptional dispatch and ensure that exceptional dispatch does not lead to market distortions and adverse ratepayer impacts.

[1] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, p. 16.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

The CAISO’s Issue Paper states that given the unique characteristics of energy storage, new market rules and changes to the CAISO’s optimization models may be needed to fully capture energy storage’s capability to provide flexibility and maintain grid reliability, and to maximize energy storage’s effectiveness in achieving California’s clean energy goals.[1]  

The Issue Paper rightly places “achieving clean energy goals” among the primary objectives of the Energy Storage Enhancement Initiative.  Cal Advocates supports CAISO’s efforts to revise market rules and systems to minimize GHG and other criteria emissions.  However, the Issue Paper does not propose metrics to assess the effectiveness of the various proposals in achieving GHG reductions.

Energy storage is not an inherently clean energy resource.  Energy storage is only as clean as the resources that charge it and may be less clean due to storage efficiency losses.  Energy storage has the potential to reduce GHG emissions by charging during intervals when marginal GHG emissions are low and discharging when marginal GHG emissions are high.  Revisions to CAISO market rules and market systems should seek to ensure that market incentives are aligned with the state’s GHG reductions goals.

The example given in the Issue Paper in Section 3.2 “Ensuring State of Charge,” provides an illustration of a case in which maximum GHG reductions are not necessarily aligned with “efficient” market outcomes because GHG reductions are not considered in the market formulation.[2]  In the example for a critical peak day, 3.6 giga-watt hours (GWh) of storage is required to meet peak system load, although up to 8 GWh of stored energy could be made available from the 2 giga-watts (GW) of energy storage.  If the additional energy were stored during periods of low system marginal emissions, more storage energy could be dispatched at peak to displace higher marginal emissions resources.  None of the proposed revisions to CAISO market rules and systems directly address the effectiveness and efficiency of GHG reductions, nor do they propose market-based mechanisms for achieving additional GHG reductions.  The CAISO should include GHG reductions in the evaluation of its proposed market changes.

Cal Advocates generally supports the use of market-based mechanisms rather than constraints to achieve efficient market outcomes, including for the efficient and effective reduction of GHG emissions.  Of the proposed policies, the energy shift product comes closest to a market-based solution for reducing GHG emissions.  The Issue Paper indicates that the CAISO could procure additional energy shift if it is economic to do so.[3]  Among the considerations in procuring additional energy shift above that required to serve reliability needs could be the cost-effectiveness of achieving additional GHG reductions.  Such a market mechanism could serve as the basis for the development of a policy framework under which additional, cost-effective GHG reductions could be procured in CAISO markets.

The CAISO should conduct quantitative market-simulation based studies to compare how the alternative policy proposals best support system reliability, are cost-effective, and reduce GHG emissions.   The CAISO should define a set of metrics for reliability, cost-effectiveness, and emissions for comparing the proposed alternatives and conduct production-simulation studies of the alternative policy proposals.

The CAISO should also consider the impact on GHG emissions from the operation of energy storage in ancillary service (A/S) markets.  Energy storage is currently operating primarily in the CAISO A/S market rather than the energy market.[4]  Energy storage charging and discharging resulting from A/S awards is dispatched by the CAISO’s Automatic Dispatch System in response to second-by-second system reliability needs rather than in response to energy price signals.  As a result, energy storage’s participation in A/S markets does not necessarily follow the pattern of charging during periods of low marginal emissions and discharging during periods of high marginal emissions.  Energy storage will likely continue to provide A/S in addition to more energy as storage penetration increases.  Energy storage’s participation in A/S markets may result in increased rather than decreased GHG emissions, and these impacts must be considered in assessing the ability of energy storage to contribute to the state’s clean energy goals.

The various proposals listed in the Issue Paper that are intended to mitigate sub-optimal energy storage dispatch may differ in effectiveness in reducing GHG emissions. The CAISO should evaluate the net GHG emissions in both the energy and A/S markets under each of the proposed revisions of market rules and market systems. 

[1] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, p. 3.

[2] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, pp. 9-10.

[3] Energy Storage Enhancements Issue Paper, California Independent System Operator, April 28, 2021, p. 13.

[4] Energy Storage and Distributed Energy Resources Phase 4 Final Proposal, California Independent System Operator, August 21, 2020, p. 19.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Thank you for the opportunity to comment on the California ISO’s Energy Storage Enhancements Issue Paper. I write today on behalf of Camus Energy, a San Francisco based Grid Management as a Service provider aiming to help utilities and energy providers reduce cost, reduce carbon emissions, and improve resiliency in part by managing distributed resources. 

The California ISO has made tremendous progress in ensuring resource adequacy under unprecedented demand, policy, and climate conditions. Even with the climate crisis driving higher temperatures, more droughts, and greater danger from wildfires, the CAISO has been a leader in decarbonizing the power sector while building an increasingly equitable and increasingly reliable energy supply. Without a doubt, energy storage is an essential part of meeting California’s energy demand and climate policy goals. 

While the recently completed fourth phase of the energy storage and distributed energy resources (ESDER) initiative included development of market power mitigation for storage resources and tools to help scheduling coordinators manage state of charge, there is more work to be done in building a durable market that sufficiently values energy storage and ensures it i’s able to meet dispatch demands. 

To that end, Camus poses the following recommendations and questions? questions and suggestions in response to the CAISO’s recent Storage Enhancement issue paper:

• Agree with the suggestion that submitting bids 75 minutes is reasonable, but would like to see progress down to 15 minutes feasible or aligned with ERCOT.

• While perhaps outside of scope, advocate for more transparency in markets including accounting for which resources are dispatched and when, which will help public and private stakeholders.

• We are in support of a scarcity price cap and propose that this issue be revisited as more market data becomes available.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

NRG/Direct Energy appreciates CAISO's attention to the set of topics outlined in the Issue Paper.  Comments submitted on this initiative are limited to approaches to address Variable Charging Rates as outlined in Section 3.3. 

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

No comments at this time

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

No comments at this time

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

Section 3.3 of the Issue Paper describes the variable charging rate obstacle when scheduling storage resources, and proposes two potential solutions to this problem. We do not support CAISO’s comment suggesting this could be managed on the resource side by oversizing or modifying the Masterfile parameter to reflect a consistent charging rate regardless of the SOC. It would be economically undesirable to overbuild the asset in order to portray a uniform charge rate. NRG recommends the ISO move forward with the other proposed solution in the Issue Paper to enhance the NGR model to allow the market to keep track and optimize the variable charging rates automatically.   

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

No comments at this time

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

No comments at this time

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

EDF-R appreciates the CAISO’s making good on its commitment to launch this stakeholder initiative as soon as practical after the Market Enhancements for Summer 2021 initiative. EDF-R agrees with the CAISO and stakeholders that we are at an inflection point for storage integration -- CAISO sees need for an additional 10,000 MW by 2025 to replace retiring gas generation, and much of that is sure to be in the form of storage.  EDF-R encourages the CAISO to adopt a least-regrets approach for this initiative, not to eschew solutions purely on the basis or scope or magnitude, and to make all discussion and ranking of issues and solutions transparent. 

To the end EDF-R requests that the CAISO set the expectation now that the Energy Storage Enhancements initiative will be an ongoing initiative, like the CAISO’s Interconnection Process Enhancements[1] initiative, and structure the initiative appropriately. Certainly, this initiative cannot tackle every storage item, the list of needs should be cataloged and prioritized with stakeholder input. EDF-R requests that the CAISO hold a workshop after this round of comments and before the Straw Proposal to develop a list of topics and allow for one round of comment on that list before publishing the Straw Proposal. This second comment round would allow stakeholders to see each other’s proposals and provide input on ranking and prioritization. 

EDF-R’s suggestions for the list are:

• Request the CAISO hold a workshop before the straw proposal, and develop a list and rank for all stakeholders’ energy storage concerns
• Provide more information on technical limitations for expanding the Real-Time price horizon
• Assist stakeholders in developing a simplified approach to identifying retail load as it relates to energy storage facility station power 
• Consider changes to proposed MSOC and energy shift products
• Consider changes to storage exceptional dispatch 
• Adjust the Bid Cost Recovery window to charge/discharge window
• Improve the CAISO’s interconnection modification procedures to fast-track reviews for modifications required to provide non-degrading storage

EDF-R provides some additional context on these suggestions below but stops short of detailed explorations for brevity’s sake at this early stage.

[1] See CAISO’s 2015 IPE as an example. Providing unique topic numbers for each item (that do not change throughout the process) is hugely helpful for tracking items as they are split off onto different tracks or are dropped http://www.caiso.com/Documents/IssuePaper-StrawProposal_InterconnectionProcessEnhancements2015.pdf  

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

EDF-R strongly supports further discussion on all these topics. 

EDF-R suggests that Bid Cost Recovery window to charge/discharge window should be dynamic based on nameplate power/capacity ratio., and if that is not feasible, at least should be adjusted to 8 hours instead of 24. 

EDF-R continues to be interested in exploring the possibility of extending the real-time market horizon. The CAISO’s response that the implementation is not technically feasible is unsatisfying at best. If it is indeed the case that expanding the horizon is technically infeasible, EDF-R requests the CAISO put the issue to bed finally by providing sufficiently detailed justification for this position. EDF-R requests the CAISO provide a summary of what the current market system can support, what technology improvements would be required to increase the horizon (to 6 hours? 10?) and cost information so that the stakeholder community can evaluate the value proposition of the required technical upgrade.)

EDF-R suggests that the CAISO make advisory prices available which would be helpful in real-time market responsiveness for energy storage resources and would enhance market transparency. 

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

EDF-R does not support any policy on state of charge that is designed as a mandate or a limit. EDF-R does not believe the CAISO has demonstrated a need for a blunt or restrictive market action and does not believe actual battery performance this summer will provide that justification either. 

The best way to ensure generation resources perform is to send proper price signals. EDF-R supports efforts to enhance market price signals to incent generation to be available when needed. 

In the spirit of providing CAISO some market tool to ensure state-of-charge, EDF-R would be interested in exploring an MSOC-type procurement which is triggered on days where load is expected to outpace generation, and the MSOC procurement is assigned a minimum compensation value with potential for increased compensation. EDF-R’s concern is that without a minimum assigned value, something like the energy shift product that is able to clear at $0 will clear at $0 as was the case with historical flexible ramp product. This could be driven by poor implementation which puts all the burden on energy storage resources to provide a service for no compensation.  If CAISO continues to pursue the energy shift product, EDF-R proposes a biddable product but with a minimum price greater than $0.  Similarly, EDF-R is concerned about the proposed biddable stored energy product. Implementation is crucial for this to work and be priced fairly, and EDF-R is concerned about the complexity on deriving an accurate price.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

EDF-R supports including exceptional dispatch as a scope item in this initiative. The CAISO tariff acknowledges storage as NGR with the ability to shift energy, rather than generate it, and as such exceptional dispatch of storage should come paired with the order to charge the storage for that dispatch, and storage should receive appropriate compensation for both. 

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

Station power improvements

EDF-R requests the CAISO work with stakeholders to develop a simplified approach to identifying retail load as it relates to station power specifically for energy storage projects. The CAISO’s current position on station power effectively boils down to “station power is retail load thus not the concern of the CAISO.” This position is tenable for traditional generators as they do not shift energy and thus any negative read on their metering systems can be assumed to be retail load. Indeed, this is exactly how some station power agreements with LSEs are constructed. EDF-R acknowledges station power is tricky subject but raises this issue as a concern because as it stands energy storage implementations are being grossly overcharged for retail load or installing duplicative and absurd dual metering systems to separate the load: imagine if California built two state highway systems because Ford drivers didn’t want to share the road with Toyota drivers. EDF-R requests the CAISO provide sufficient load data to the ISO settlement system so that participants can use this data to calculate actual station power load. 

Modification Process improvements:

EDF-R requests the CAISO add interconnection request modification process enhancements to the scope of this initiative. The CAISO has no current plans to holistically address interconnection process enhancements currently. EDF-R request the CAISO modification process changes that reduce the uncertainty and lead time present in the current modification review process required for maintaining a lithium-ion energy storage facility in a non-degraded state.

There are three general scenarios for augmenting a lithium-ion energy storage facility to maintain a non-degraded state:

1. Add more batteries 
2. Add more batteries with additional DC/AC inverters
3. Add more batteries with additional DC/DC converters

This maintenance needs to be performed at least on an annual basis, but possibly more frequently. Adding more batteries (Maintenance Scenario 1) is not a viable consideration because, while is technically possible, it is in opposition to the battery manufacturer’s recommendations. In conversations held in late April the CAISO indicated that a full post-COD modification review ($10,000 deposit, technical data and a timeframe of at least 45 days) is required to approve any inverter or converter changes (Maintenance Scenarios 2 and 3). 

EDF-R respectfully requests that the CAISO consider the two proposals:

Proposal 1 

Energy storage customers may request pre-approval for their maintenance plans. EDF-R envisions a process where, using the existing modification review framework (including required deposit and technical data), an interconnection customer could request and receive approval for scheduled post-maintenance configurations well in advance of the date the maintenance would be performed.

Proposal 2

The CAISO adds Battery Energy Storage System Maintenance activities up to a certain level as a change that is allowable without a modification review in Section 6.2.1 of the BPM for Generator Management, “Modifications That Are Approved Without Material Modification Assessment.” 

Lithium-ion energy storage facilities degrade at a rate 10 times that of a solar PV install and this maintenance frequency is worthy of a specific and streamlined process. 

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Independent Energy Producers (IEP) broadly agrees that the issue paper has identified the issues that need to be resolved in the near term to facilitate the participation of storage assets in the wholesale market and to ensure that energy storage systems are managing their state of charge to be ready to discharge when needed to maintain reliability. On the latter point, IEP suggests that the state of charge issues, which are discussed in Section 3.2 of the issue paper, should be deprioritized and addressed in a Phase 2 of the initiative in order to expedite resolution of the other operational issues. CAISO recently adopted an interim state of charge policy that energy storage systems must reach a full state of charge by the evening ramp during tight supply conditions. The adoption of the interim policy mitigates the exigency of addressing state of charge. Alternatively, CAISO could transfer the issues discussed in the Section 3.2 to the Day-Ahead Market Enhancements initiative.  

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

IEP agrees that improving the representation of marginal costs of storage in CAISO’s market operations is an urgent need for storage market participants and should be included in the scope of this initiative. The deficiencies in the current representation of storage marginal costs and operational capabilities in CAISO’s market models described in Section 3.1 have already resulted in economic harm to storage resources.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

As stated in our response to Question 1, CAISO should either assign the state of charge issues to a second phase of this initiative or consider moving this topic to the Day-Ahead Market Enhancements initiative. Because CAISO recently adopted an interim policy in the Resource Adequacy Enhancements initiative to require energy storage facilities to maintain a minimum state of charge for critical hours when the Residual Unit Commitment process indicates that supply conditions will be tight, the need for longer-term solutions is less urgent. Although IEP prefers in-market solutions to the rule-based interim policy, CAISO should prioritize the other issues with the goal of resolving them by the end of the year before turning its attention back to state of charge.

As the issue paper explains, expanding the real-time market to accommodate managing the state of charge is not technologically feasible and would require the optimization to look ahead “14 or more hours” (Issue Paper, p. 12), at which point the “real-time” market is bordering on a day-ahead timeframe. Because market-based solutions to managing state of charge inherently entail longer timeframes, IEP suggests that CAISO consider moving the state of charge discussion to the DAME forum.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

Improving market operations to incorporate information on variable charging rates is imperative. IEP is aware that not all CAISO operators understand why storage operators in some instances have not followed charging instructions from CAISO. However, they did not follow the instructions because it was physically impossible for them to do so. This misunderstanding of physical constraints of the storage resource results in dispatch instructions that proved to be infeasible and consternation from both CAISO and resource operators. To avoid continued economic harm to storage operators, CAISO must resolve this problem quickly.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

IEP agrees that CAISO should include exceptional dispatch in scope. As the issue paper describes in Section 3.1, CAISO dispatch instructions and compensation rules need to account for the opportunity costs that storage resources face due to their limited discharge durations.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

The Joint CCAs appreciate the CAISO’s efforts to make enhancements to its energy storage model. With the increasing number of renewable resources on the grid, the Joint CCAs believe that energy storage will continue to play a vital role in providing reliability. In October of 2020, a group of CCAs released a request for offers (RFO) for long duration storage resulting in a wide variety of technologies. Through this RFO, the Joint CCAs took a technology agnostic approach and is possibly considering non-lithium-ion long duration storage solutions. The CAISO’s Non-Generator Resource model does not account for emerging storage technologies that may require different parameters to participate in the CAISO market. Below are a few specific examples:

1. Although the CAISO has pumped storage hydro and multi-stage generating models, the Joint CCASs are unsure if certain technologies can participate under these models. Several non-li-ion technologies do not have the ability to provide seamless transition from charge to discharge.
2. To account for low round trip efficiency rates, certain technologies have paired with li-ion to provide better efficiencies and seamless transitions from charge to discharge. Can the CAISO accommodate for two storage resources under a single resource ID with different ramp rates and charge/discharge capacity?

The Joint CCAs request that the CAISO assess its current energy storage models and identify if it is compatible for non-lithium-ion technologies. The Joint CCAs look forward to working with the CAISO on these issues and requests that a workshop be held to discuss various new technologies and determine the feasibility within CAISO models.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

LS Power thanks CAISO for launching the Energy Storage Initiative to address these important operational issues. Hundreds of megawatts of battery storage are coming online in 2021 and thousands more megawatts are needed to meet California’s clean energy and reliability goals. Storage resource owners/operators such as LS Power are completely aligned with CAISO with respect to meeting these goals. Storage resource owners also have the same operational goal as CAISO during tight system conditions: to have the battery full and ready to discharge when needed (such as during evening peak hours).

It is important that CAISO wholesale market prices both accurately reflect reliability need and guide dispatch the vast majority of the time, so that energy storage resources can contract with load serving entities to mitigate their market exposure. Resource adequacy contracts alone do not cover the revenue requirements for new-build energy storage projects, so a smoothly functioning energy market that allows storage to effectively participate and earn revenues is critical to the task of evolving from today’s fleet to the clean and reliable fleet called for in analysis such as in the CPUC’s Integrated Resource Planning process or CEC’s SB 100 report. In addition to market participation, energy storage projects can enter into energy contracts, and for these storage project needs to be able to know with high confidence that it will be charged and discharged by CAISO according to the prices in its bid curve in order to meet its obligations. These energy contracts can simultaneously help protect consumers from price spikes and provide energy to the grid when it is needed most, in addition to helping attract the investment needed to finance project development. LS Power emphasizes that CAISO should seek to implement in-market solutions and use transparent prices to achieve desired goals wherever possible. This is good market practice, is squarely in line with CAISO’s stated principles, and will minimize risk for any changes that need FERC approval.

LS Power agrees with that the list of issues identified in the Issue Paper are the right areas of focus, and suggests a phased approach to working through the topics.

• Phase 1 – LS Power suggests CAISO’s initial focus should be on the following:
  • 3.1 Representing Marginal Costs (specifically MIO, spread bidding, and BCR)
  • 3.3 Variable Charging Rates
  • 3.4 Exceptional Dispatch

These are areas of immediate concern, affecting storage resources and CAISO grid operators today, and CAISO should aim to finish this phase by December 2021. Implementing these Phase 1 solutions could reduce the need to address some of the issues noted below in Phase 2, which is another reason we propose this phased approach.

• Phase 2 – LS Power suggests the second phase focus on area 3.2 Ensuring State of Charge (SOC). Storage operators already have a strong financial inventive to ensure that their SOC is sufficient to meet Day Ahead schedules, so resolving the existing market issues in 3.1 effectively should largely solve this issue. CAISO is also launching the end-of-hour (EOH) SOC tool in Fall 2021. This tool could have a significant impact on ensuring SOC and stakeholders should work together to make sure it is implemented effectively. Finally, CAISO just moved to establish the minimum SOC requirements that are in place until the end of 2022, so there are over 18 months remaining to thoughtfully develop alternative solutions. Gathering actual data on resource performance given all of these recent changes would be advisable before coming up with yet another solution to a problem that should not exist if resource owners could control SOC effectively through bid curves.

CAISO could also include remaining open issues, such as Bids Submission Timeline and End of Horizon Opportunity Costs, in Phase 2. These are “nice-to-have” items that would likely improve the market outcomes but are not high priority. CAISO should aim to finish Phase 2 by mid-2022.

LS Power provides further information on specific issues and proposed solutions in the sections below. Various stakeholders have reached out to the LS Power team with questions trying to better understand these issues and how they affect storage resources in the market today, and we expand on issues briefly below for the benefit of the CAISO stakeholder community participating in this process.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

As stated above, LS Power thinks issues related to representing marginal costs is a high priority item for this initiative. In particular modifying MIO and spread bidding (these are really two parts of the same issue), and bid cost recovery (BCR) rules should be addressed in Phase 1.

The other issues of end of horizon opportunity costs and updating bids submission timeline are of interest, but LS Power does not think they are as high priority and can therefore be deferred until a Phase 2. LS Power discusses Phase 1 issues in more depth below and defers comments on Phase 2 issues until a later date.

Influence of Advisory Intervals - MIO

How MIO works today: Briefly, MIO is a design aspect of the Real Time market that estimates market conditions several intervals in the future, known as the advisory intervals, in order to optimize the dispatch signals given to resources in the immediately following interval, known as the binding interval. This way CAISO could start ramping up slow moving but low-cost traditional generators if CAISO anticipated a need occurring later in the hour, keeping costs low. However, most battery storage resources differ from conventional resources in various key aspects, and the MIO algorithm has unintended consequences as a result. Specifically, the MIO algorithm is extremely sensitive to changes in CAISO’s estimate of future system conditions, and frequently those estimates do not materialize in practice. These inaccurate forecasts are a major reason why just extending the time horizon for MIO would not solve these problems that would only be a “solution” if the dispatch algorithm had perfect knowledge of the future.

Having no inertia, storage resource dispatch signals (DOTs) are thus moved up and down in the binding interval in a way that is beyond the control of resource owners, any time CAISO estimates a very low or high price materializing in the advisory intervals. The algorithm does not consider Day Ahead schedules in any way when determining Real Time market dispatches. Because MIO in real-time does not consider awards in prior market runs, it is not uncommon for the MIO algorithm to lock in financial losses in storage operation during various intervals of the day.

The result of this takes two forms:

1. When CAISO expects a future high price, it starts charging the resource at prices that may be outside of the resource’s bid curve. The expected future high price may or may not actually occur, leading to potential financial losses and unnecessary battery cycling.
2. When a future low price is anticipated, CAISO starts discharging the resource at prices that may be well below the prices that the resource has stated a desire to charge at in its bid curve. This discharge could occur regardless of potential Day Ahead award schedules and could make it difficult for the resource to stay charged ahead of its Day Ahead discharge hours.

Both outcomes make it very difficult for a resource owner to control their resource’s SOC of charge through the only mechanism available to them: the bid curve. Outcome B also creates reliability risks: it discharges a resource, potentially for low prices, leaving it empty relative to where it should be. To be clear, there are currently no market guardrails against this undesirable outcome happening prior to the evening peak, except on days when CAISO has declared a minimum SOC requirement.

Additionally, MIO does not “remember” previous market awards and past predictions, it is always forward looking with no memory. As a result, we commonly see that a storage resource may be discharged in one interval, in anticipation of a lower price in the advisory intervals, and then charged at a higher price a few intervals later.

The chart below shows a typical example of how MIO violates bid curves and creates risks for both resource owners and grid reliability. The chart shows the battery’s dispatch operating target in the shaded black boxes and the real-time locational marginal price in blue. Throughout the two hours shown below, the battery’s bid curve expressed willingness to charge at prices below $50/MWh and discharge at prices above $100/MWh as bids to keep the battery full in preparation for evening Day Ahead energy awards.

Over the period shown below, CAISO dispatched the battery to discharge below its bid curve at a price of $36/MWh, charged it back up at $37/MWh, and then discharged the battery further at $23-24/MWh. This creates multiple risks. All of the discharges in violation of the battery’s bid curve left it emptier than desired prior to Day Ahead awards later in the evening. The resource lost money on the cycling and is not eligible for bid cost recovery under current CAISO rules. Finally, the unnecessary charging and discharging created additional wear and tear on the asset, reducing its useful life.

MIO Recommendation: MIO should be constrained to a much smaller number of advisory intervals for Non-Generator Resources (NGRs) (maybe 2-3 intervals, i.e. 10-15 minutes, instead of the current 13 intervals, i.e. 65 minutes). Also, an adjustment could be made to specifically avoid Outcome B noted above, where the unit’s SOC is lowered in anticipation of future low prices. The reliability risk to the grid and financial risk to storage operators from that scenario is so great that LS Power and others would surely prefer to forgo any theoretical opportunity to “earn a spread” in those intervals when a low price does subsequently appear after the unit has been discharged for prices less than was bid into the market.

Spread Bidding

The issue of spread bidding is intertwined with MIO challenges and cannot really be separated. The issue paper was a bit odd in its treatment of “Spread Bidding”, which is really just applying a name to the way storage is already dispatched in the Real Time Market. Today, LS Power’s understanding is that CAISO treats a storage resource’s Real Time bid curve as a Spread Bid.

Note that the only tool a storage resource owner has to control the operation of their plant in the CAISO market is their bid curve. Storage resources are fully automated and respond as closely as possible to the dispatch CAISO sends to them, and as such we have no ability to manage SOC with any tool other than our bid curve at this time. The bid curve must convey physical costs like variable operations and maintenance, and opportunity costs, such as the importance of being full to back an energy contract or to meet a Day Ahead schedule and avoid exposure to massive imbalance losses.

There is currently no way for a resource to opt out of the Spread Bid treatment, but ideally there should be. Consider two resources with similar technology offering a $40/MWh spread between willingness to charge and discharge at a given time.

• One resource is offering to charge at $0/MWh and discharge at $40/MWh—both LMPs are commonly seen on springs day at CAISO nodes in solar resource areas. This resource may not have impending Day Ahead energy awards to deliver and thus be trying to charge only when nearby solar PV is being curtailed, something it guesses at using the price signal. This may be very desirable behavior for the grid and for the environment, ensuring that the battery uses limited space available to soak up energy only when it is cheapest and cleanest.
• Another resource is offering to charge at any price below $500/MWh and discharge at any price above $540/MWh. This resource really wants/needs to be full (for example, in preparation to deliver an evening Day Ahead energy award, or in anticipation of impending real-time volatility) and is trying to convey its high opportunity cost to the market. This is very different behavior, but may similarly have desirable outcomes for CAISO grid reliability, as the resource is ensuring that it has adequate charge to meet its commitments and be available to the Real Time market as the fastest and most flexible peaking resource a grid operator could ask for.

Based on our understanding today CAISO dispatches resources in both examples the same way due to the MIO algorithm treating bid curves as “spreads” and not as a statement of the actual prices a resource is seeking to be dispatched above or below.

Lastly, CAISO’s explanation of spread bidding and MIO in previous stakeholder processes only makes sense in the context of theoretical examples which use one price to discharge and one price to charge. As discussed in previous stakeholder processes such as ESDER4, real bid curves have up to 10 monotonically increasing segments, none of them necessarily at 0 MW, and it is unclear what it means to have a “Spread Bid” with these real bid curves.

Spread Bidding Recommendation: We recommend that CAISO address the MIO issues first. If the larger MIO issues could be resolved, spread bidding could be implemented as an option for resources that desire to have CAISO fully manage their functioning. Not all resources will need spread bidding so if implemented it should already be done as an option. It is important for resource owners/operators to be able to rely on using resource bids and market prices, rather than the spread, in order to meet energy contract obligations or other business use cases. If MIO and spread bidding in a period result in the resource being depleted and unavailable to meet a price spike, that resource could potentially be exposed to huge imbalances. For storage operators, being able to rely on market prices to meet contractual obligations is a must to foster a financially healthy sector and attract the levels of investment needed to build the thousands of megawatts of storage California requires.

Bid Cost Recovery (BCR)

BCR is also an urgent item that needs to be addressed in this stakeholder process under Phase 1.

CAISO must review the current BCR calculation for the Real Time market as it is applied to NGRs. Due to the specifics of that calculation, NGRs are almost never made whole for losses incurred by resources following CAISO dispatch (i.e. the settlement charge code is almost always zero at the end of the day). Intra-day financial losses can and do occur frequently due to the undesirable outcomes mentioned above from both Real Time/MIO dispatch issues and exceptional dispatch (ED). In our discussions of the calculation with one CAISO Settlements group, we learned that storage resources have never received BCR, which points to investigating why BCR is not working for storage.

To be clear, our understanding is that BCR works well for traditional generators that only have positive levels of output and whose operational costs have a different relationship to their bid curve (i.e. operation requires a specific number of dollars for each MWh generated) than energy storage does (which uses a curve formatted the same way to convey a necessary spread between charge and discharge prices, as well as to price in the opportunity cost of being empty at the wrong time). But both resources have BCR calculated with the same math.

Just one example of why BCR does not work well for storage is how the math treats a resource that is not charging because it is full. Most batteries spend several hours completely full every day, waiting to be discharged during the time of highest need as expressed by the market in the form of high prices. If a resource is always willing to charge say at any price below $50, and in a given interval the LMP is $49 but the resource is not able to charge because it is full (100% SOC), BCR treats the resource as though it made a profit of $49/MWh by not charging in that interval, because the bid curve expressed a willingness to charge at that price. There is no defensible logic to this, it is just an algorithm that worked well for generation resources being applied inappropriately to storage resources. It was easy to overlook this as recently as early last year when there were <200 MW of storage on the 50GW CAISO grid, but it is time for it to be fixed.

BCR Recommendation: BCR rules should be completely rewritten for NGRs in a way that takes into account the specific economics of energy storage. One plausible option would be for CAISO settlements to compute BCR using a backtest of what a resource would have earned if it was a price taker and its bid curve had been honored exactly in each hour given the resource’s physical characteristics and actual prevailing LMPs in each of the Day Ahead, Fifteen Minute, and Five Minute Market runs for the day (after any adjustment to the bids is made for market power mitigation of course). The BCR payment would make up the difference between actual market revenue and what the resource would have earned had it been dispatched according to its bids and prices only (i.e. it would undo any damage from MIO or ED dispatch that took it out of the market). This type of a script is relatively simple to develop, and indeed many market participants already run something like this in their own offices each day.

This is only one plausible solution and we recognize it is a complicated subject. We look forward to collaborating on it.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

LS Power suggests that this issue be deferred to a Phase 2 with a goal of a final proposal by mid-2022. While it is an important issue, CAISO has already taken steps to address it and it is not urgent to make further market changes, given that the minimum state of charge requirement is in place through 2022. Additionally, solutions identified in Phase 1 may change potential solutions to ensuring SOC. Furthermore, CAISO is launching the EOH SOC tool in Fall 2021. This tool could have a significant impact on ensuring SOC and stakeholders should work together to make sure it is implemented effectively.

Several stakeholders have suggested extending the Real Time market look out horizon as a solution to ensuring SOC. We agree with CAISO that this option should not be pursued as a potential solution. The challenges with MIO illustrate that the predictions of market prices and actions in the future are imperfect. LS Power thinks that extending the Real Time horizon could exacerbate this issue and create more uncertainty.

LS Power defers comment on other issues related to ensuring SOC to a later date.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

LS Power thinks that variable charging rates is a high priority issue that should be included in Phase 1. The problem is pervasive, affecting every battery storage project that is operating at a near full or near empty SOC for reasons of simple technical limitations. CAISO operators have called our storage operations personnel about this issue multiple times and it should be addressed for operators’ benefit above all. We believe there is a simple fix from a technology perspective to address this issue

In particular, as identified in the Issue Paper, LS Power notes the primary issue is CAISO sending high power charge commands when a battery is almost full, which is an infeasible dispatch (ID) the resource cannot follow. CAISO regularly sends these IDs to NGRs due to limitations of the NGR model, specifically an inability to express the physical limitations that all batteries have with respect to charge rates that change as a function of SOC. A straightforward fix is for CAISO to utilize one of the existing telemetry points. LS Power urges CAISO to consider this as a solution or implement any other potential solutions. This issue should be addressed quickly, as ID by definition cannot be followed by the resource receiving them, and the result is reliability risk for CAISO and financial risk for resource owners, and lot of unnecessary confusion between CAISO and Storage operators.

Lithium-Ion, like all electrochemical battery types that we are aware of, can be rapidly charged across much of its usable SOC range, but its charge acceptance drops rapidly as it approaches 100% SOC. In other words, the maximum charge rate of a battery varies as a function of SOC, decreasing considerably when the battery is almost full, and this is true of all batteries. This is shown in the figure below, where the plant can charge from 0% up to nearly 88% of its maximum SOC at the plant’s maximum MW rating, however the charge rate falls off exponentially as it approaches 100% SOC, and thus the last few MWh of charge take place at a much lower rate and thus take much longer.

At this time, the NGR model does not have any way to act on this information, and as a result any time the battery is not 100% full, CAISO’s algorithms assume the battery can be charged at its maximum charge rate. This happens to LS Power battery storage projects multiple times per day on average, particularly because the resource sits 90-100% full most hours of a typical day. LS Power frequently receives ID when the plant is 99% full and capable of being charged at only 1-5 MW yet receives charge commands of much higher levels. Note that this is also an issue for discharging when the plant is near empty (close to 0% SOC), but occurs much less often and to a lesser degree.

One question in the Issue Paper is whether a resource should oversize or modify the Masterfile parameter to reflect a consistent charging rate regardless of SOC. LS Power does not think this is a practical solution so should not be considered. Given that the effect kicks in around 85-90% SOC for most batteries, oversizing the batteries would mean every project buy 10-15% more batteries as a workaround for this issue. With thousands of MW of battery capacity coming to the CAISO grid soon, this would put unnecessary costs onto resources and their off-takers that could quickly mount into the hundreds of millions of dollars additional cost to CAISO ratepayers.

Variable Charging Rate Recommendation: There are no fixes for this that can be implemented on the project side alone, there must be coordination between CAISO and the plant’s telemetry. LS Power suggests two possible fixes.

1. Allow NGRs to upload a MW vs. SOC piecewise curve as part of the master file. This curve could be used for a simple check against the SOC each time a DOT is being calculated. The curve would look much like figure above, only with SOC as the X axis instead of time. For an example project of 100 MW, if SOC was 0-88% of the Maximum Energy, the plant’s PMin/Max Charge Rate is 100 MW, but if SOC is 99% of Maximum Energy the plant’s PMin/Max Charge Rate drops to ~1 MW or so.

• Use of AGC telemetry in RT dispatch: LS Power’s understanding is that there are two telemetry points, an Upper Operating Limit and a Lower Operating Limit, which are used by the AGC system for dispatching units that are providing Regulation. CAISO could additionally make use of those points for projects that are not providing regulation. The plant could keep the telemetry points updated at all times with its current Max Charge Rate for the Lower Operating Limit, or the current Max Discharge Rate for the Upper Operating Limit based on the current SOC level and CAISO can use these to make DOT decisions. These points have the same units (MW) already, so it should be a fairly straightforward implementation on the CAISO side to constrain awarded DOTs in the fifteen and five minute periods to be less than or equal to the Upper limit and greater than or equal to the Lower limit.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

LS Power thinks ED is a high priority issue and should be included in Phase 1. In 2020, CAISO operators frequently relied on ED to control storage resources in tight conditions. Specifically, this usually took the form of the CAISO operations desk making phone calls to the resource’s scheduling coordinator and operators, telling them that the resource is being removed from the market and needs to set its output at a specific level for a certain time period.

ED effectively takes storage out of the Real Time market, with grid operators losing the resource’s ability to quickly and flexibly respond to reliability needs, and in some cases creating significant lost opportunity cost for the resources. Excessive use of this tool creates market inefficiencies, which in turn leads to increases in cost to serve load. Upon review of CAISO settlement statements, there is no evidence of the resource being made whole for losses resulting from ED either. The normal settlement logic flows through for all charge codes, and in theory BCR is supposed to make a resource whole, but BCR does not work in this or any other scenario for energy storage as currently implemented, as discussed above.

ED Recommendation: CAISO operations desk personnel need to have more information about a storage resource’s market schedules and their physical characteristics so they can make appropriate decisions on ED. Situations where an ED instruction by CAISO leads to an opposite dispatch action as compared to resource’s market awards (such as a resource getting emptied by ED before its Day Ahead discharge award) should be avoided, unless it is for reliability purposes. Settlement rules for use of ED should be reviewed under this stakeholder initiative. A storage resource on ED should be made whole for any lost opportunity costs or imbalance charges due to ED. For instance, an ED instruction that empties the resource ahead of its Day Ahead discharge awards during evening peak could lead to huge financial exposure for the resource, and it should be made whole. LS Power is always available to collaborate on improving telemetry, documentation, and any training CAISO Operations deems necessary so all parties can better understand operating characteristics of this resource class.

LS Power also thinks allowing CAISO to target a SOC rather than specifying a target MW amount may help resolve undesirable outcomes when EDs are issued to storage. CAISO should specifically review how BCR would need to work for resources subjected to an ED and removed from normal Real Time functionality as well.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

LS Power appreciates CAISO’s timely attention to the issues identified in this initiative. LS Power suggests the initiative may need to stay open beyond Phase 2 to incorporate issues identified in the next year as more storage projects come online and resource owners and CAISO get more operational experience.

One issue to add to Phase 2 is on NGR Ancillary Service Awards. Currently, NGR resources providing Ancillary Services get awards based on their 1 hour dispatch capability. This is primarily due to the current 1 hour optimization construct for the Day Ahead market and is not in line with CAISO’s tariff and BPM definitions of Ancillary service, which state this to be a 30 min product. The issue for NGRs would have been resolved if CAISO proceeded to implement a 15 min market in Day Ahead. However, since CAISO is no longer proceeding with a Day Ahead 15 min market, the NGR Ancillary Service issue should be resolved. NGRs with a state of charge should be required to be capable of a 30 minute discharge to be awarded regulation-up, spinning reserves and non-spinning and 30 minute charge to be awarded regulation down.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

LSA and SEIA ask that the CAISO include the important issue of controlling grid charging of storage capacity in Mixed-Fuel Resources (MFRs) in Co-located Resource (CLR) configurations, to accommodate compliance with the Investment Tax Credit (ITC).

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

LSA and SEIA strongly support including in the scope of this initiative the topic of limiting Mixed-Fuel Resource (MFR) charging of storage capacity with energy from the CAISO grid (as opposed to charging that capacity with renewable energy generated on-site). 

CAISO management stated several times during the Hybrid Resources Initiative (HRI) that it would consider this issue further in a future “follow-up” storage initiative.  This initiative has arrived, as the current Energy Storage Enhancements (ESE) Initiative.  Given these past statements, LSA and SEIA are puzzled at the omission of this issue from the Issue Paper.

Background

LSA and SEIA have been active participants in the CAISO’s Hybrid Resources Initiative (HRI), including multiple rounds of comment submittals in both Phase 1 and Phase 2.  LSA and SEIA generally have supported most CAISO HRI proposals but expressed serious concerns in that initiative about a foundational CAISO Management principle stated at the start of that initiative:  that MFR operating limits due to storage Investment Tax Credit (ITC) recovery must be managed by resource owners, not CAISO market rules.

The ITC, which provides up to 30% cost recovery over 5 years, will be an important factor keeping energy storage costs to ratepayers reasonable.  However, the credit is only recoverable to the extent that charging energy comes from on-site renewable resources.   Charging from the grid does not count as “renewable,” even though a large share of that energy might, in fact, be provided from renewable sources. 

Thus, it is critically important for ITC recovery that resource owners be able to restrict grid charging.  CAISO Management stated at the start of the HRI, and numerous times throughout, its belief that ITC issues are related to tax policy, and that it was not the CAISO’s responsibility to modify its rules to ensure ITC recovery for resource owners. 

Nevertheless, the CAISO stated at p.11 of the HRI Revised Straw Proposal (http://www.caiso.com/InitiativeDocuments/RevisedStrawProposal-HybridResources.pdf):

The CAISO notes the ITC as a business driver for hybrid and co-located resources because it reduces the costs associated with developing new resources that contain solar and storage combinations. The CAISO believes that this ITC eligibility issue should not drive market design proposals because the ITC is an out-of-market incentive.

When LSA and SEIA raised this issue during the stakeholder process, and in a letter and Public Comment to the CAISO Board in November 2020, CAISO Management noted that, while the issue was not included in the HRI scope, it could be addressed in a follow-up initiative planned later – the current ESE Initiative.

There are several significant, substantive reasons why this issue should be added to the scope of the ESE Initiative.  These reasons are listed below and then explained further in the rest of this response.

• The Investment Tax Credit (ITC), the source of the grid-charging limitations in question, is a federal government policy.  As such, the CAISO should craft its policies and procedures to support compliance and not frustrate it.

• Lack of tools to readily control grid charging will lead to sub-optimal market results, as Scheduling Coordinators (SCs) for MFRs must reflect extremely conservative assumptions about renewable-energy production in their market bids.

• Since CAISO has refused to accommodate these requirements, Market Participants have filled that void via potentially dysfunctional practices that could undermine the usefulness of energy storage to the CAISO grid, e.g., physically and/or contractually disabling storage capacity so that it cannot charge from the grid even where economic, or if CAISO needs that additional State of Charge (SOC) for reliability purposes.

Inclusion in the initiative will likely yield many suggestions by stakeholders.  Potential solutions include, for example, allowing the storage Pmin in the Master File to reflect the negative energy (charging) range but allowing the Aggregate Capability Constraint (ACC) Pmin to be zero.

Facilitation of compliance with federal requirements

The CAISO has often changed its market rules to accommodate energy policies adopted by state and/or federal agencies or governments.  Such changes include market rules to enable market participation by demand-side, aggregate, distributed, and storage market resources, expediting “shovel-ready” projects during the Great Recession, and implementing Resource Adequacy requirements, among many other actions.

The ITC – including requirements limiting grid charging of MFR storage - reflects federal law and policy, and CAISO should likewise examine its market policies and procedures to help with compliance.   The lack of effective tools to comply with ITC requirements will reduce the benefits of these resources to the CAISO grid, and LSA and SEIA strongly believes more must be done. 

At a policy level, significant risk of ITC under-recovery will make it difficult for mixed-fuel projects to obtain the financing needed to successfully bring the considerable capacity in the interconnection queue to market.

Sub-optimal market results

Specifically, lack of further action will likely reduce market offers from both solar and storage capacity.  This problem will occur for both proposed mixed-fuel resource configurations – Hybrid Resources (HRs) (single Resource ID for solar and storage resources) and CLRs (separate solar and storage Resource IDs).  For example:

• MFRs in the CLR configuration must carefully match their bids for the two resource IDs to ensure that, on a 5-minute basis, storage CLR charging does not exceed energy produced from the solar CLR.  This careful bidding may result in schedules where storage CLR charging does not exceed solar CLR output. 

However, if real-time VER CLR production is less than scheduled, the storage CLR must still follow its charging schedule, i.e., the additional charging energy will inadvertently come from the grid, jeopardizing ITC recovery.  This will force output forecasting (and market offers) for the VER CLR, and thus charging schedules (and subsequent market offers) for the storage CLR, to be extremely conservative.

•  MFRs in the HR configuration will have to base their market bids for their single Resource ID on an extremely conservative VER output forecast, yielding an extremely conservative storage State of Charge and reducing the level of market bids, and services offered, to a lower level than could be offered without the need for this conservatism.

Sub-optimal MFR design

ITC/grid-charging matter has since become more urgent since the start of the HRI.  The following have been occurring as Market Participants struggle to address it without CAISO help:

• The vast majoring of MFR Power Purchase Agreements (PPAs) negotiated since the ITC was adopted contain provisions simply prohibiting grid charging.  CAISO failure to provide realistic options to implement these provisions will thus undermine these agreements, which are needed for financing and construction of these resources.

• Developers are designing their facilities to be physically incapable of grid charging.  This means that these facilities will intentionally be disabled so that they cannot charge from the grid, depriving the CAISO and the market generally of significant operating flexibility that they might otherwise provide.

• PTOs are implementing their own grid-charging limitations.  For example, consider recent draft GIA amendment language to implement an energy storage addition via the Material Modification Assessment (MMA) request process.  Because the IC had said it intended to charge the storage from on-site VER capacity, the proposed PTO language would require the project to: (1) install a “control limiting device” or control system to prevent any grid charging; and (2) designate any grid charging as a breach of contract risking loss of the GIA.

Conclusion

The CAISO should include an examination of MFR Co-located Resource grid-charging limitations to comply with ITC requirements in the scope of the ESE Initiative.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

NCPA recognizes the critical role batteries play in achieving the state’s renewable goals. NCPA also understands that integrating non-generator resources into transmission systems and energy markets developed around natural gas generation will present challenges. However, NCPA is concerned that CAISO and stakeholders may be considering solutions that may actually create significant disruptions or inequities in the markets at great cost to ratepayers.

NCPA believes it would be very complex to modify MIO, spread bidding, or the cost recovery rules for storage. If necessary, NCPA suggests addressing cycling costs and opportunity costs in separate stakeholder initiatives. NCPA encourages CAISO and market participants to continue working within the confines of the existing market to optimize energy costs.

NCPA believes CAISO markets and RAAIM/UCAP are sufficient tools to incentivize availability in critical net peak hours and is concerned that new products will result in unnecessary costs to ratepayers. NCPA does not support including exceptional dispatch in this initiative for similar reasons.

CAISO and stakeholders admit that there is virtually no storage operating in the market. As of May 13, 2021, batteries only accounted for 1% of generation serving load. CAISO and the stakeholders will be in a position to better explain or even understand issues that batteries are experiencing in the market after they have been operating and testing various strategies over the summer. Until then, CAISO’s temporary MSOC tool will ensure the units are available during availability assessment hours.    

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Marginal costs should be out of the scope of this initiative, because CAISO should address them comparably to the way it addresses such costs for other resources. CAISO identified three primary cost drivers in ESDER Phase 4: 1) energy cost, 2) opportunity costs, and 3) cycling costs. Battery schedulers should be responsible for bidding their resources in a manner that will result in net energy revenues. If the CAISO still dispatches the units uneconomically, resulting in net energy costs, then load will make the resource whole through Bid Cost Recovery. CAISO should have a separate stakeholder initiative to determine cycling costs or include them in Variable O&M review in order to holistically address this potential deficiency.

CAISO should also proceed in a technology-neutral manner. CAISO currently addresses opportunity costs with restrictive Use Limited designations that do not appropriately account for critical use limited resources. Batteries are only one of those categories. CAISO should initiate a new stakeholder process to update its Use Limited program to address three deficiencies: 1 use limits and opportunity costs associated with batteries, 2) use limited hydro that does not fit in the current model, and 3) gas fired units that are nearing end of life and have a limited number of runs left (only environmental limitations are considered, rather than the remaining useful life of the resource).  These units are particularly critical to bridge the gap between current and future renewable goals, and CAISO should be concerned with keeping them available for reliability needs. CAISO should be addressing all of these limitations in a technology-neutral and non-discriminatory manner

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Existing CAISO markets and RAAIM penalties should be sufficient tools to ensure state of charge during net peak hours. For example, if the storage unit economically bids in the DAM and receives favorable results in terms of timing, pricing, and cycling, it can lock those in with RTM self-schedules. If the DA results are not favorable, then it should refine its bids in the real-time market. NCPA’s understanding is that there is no PMin on batteries, so operators should be able to bid high or low in hours when charging or discharging is inefficient. If the current paradigm is not accommodating batteries, then battery schedulers must better explain why and offer suggestions to improve existing market instruments rather than ask for special products. CAISO should aim to keep the markets as technologically agnostic as possible, and should pursue neutral reforms that benefit storage and other kinds of use-limited resources before attempting to design special products for storage resources. 

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

NCPA supports CAISO’s request for feedback from battery operators to determine the pervasiveness of the issue and whether or not there is sufficient documentation to support including charging rates as a new Masterfile parameter.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

NCPA does not support including exceptional dispatch as a scope item in this initiative. Forcing load to pay generators to hold their charge should not be necessary for batteries to successfully participate in the CAISO market. Ultimately, CAISO operators should accurately determine when to begin charging resources to ensure availability when they anticipate the need will materialize and not use special storage ED rules as an expensive aid.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

NCPA encourages CAISO and stakeholders to continue to collaborate and gain experience with bidding storage into the market this summer as more batteries come online. Further, NCPA suggests that CAISO allow all battery schedulers to test various storage bidding strategies in CAISO’s market simulation environment while they execute conservative strategies in production during this volatile period. We acknowledge that CAISO designed its market around gas-fired units, but CAISO has already integrated a diverse range of non-gas fired resources and the limited number of large-scale batteries on the system indicates that it is too soon to determine whether new products are required. NCPA suggests pursuing the targeted, technology-neutral changes discussed above, while tabling this stakeholder initiative until fall and then hosting a number of technical lessons-learned workshops in which battery operators and other stakeholders can thoroughly describe their experiences and the issues encountered in CAISO’s market. Then CAISO may publish an updated, more informed, issue paper and stakeholders can then more thoughtfully determine the most effective, least cost solutions available for resolving them.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

PG&E welcomes the CAISO’s launch of the Energy Storage Enhancements initiative and appreciates the opportunity to provide comments on the Issue Paper (released on April 28, 2021).  Below is a summary of our general comments:?  

• Limitations of the CAISO’s Outage Management System (OMS) present significant challenges to scheduling coordinators and therefore should be addressed as a part of this initiative. 

• PG&E proposes the idea of an Energy Storage Reserve product in lieu of an Energy Shift product. 

• PG&E suggests making the hybrid model available to NGR resources not paired with other fuel types in order to address market participants’ concern that batteries need to have more flexibility in adjusting their bids and schedules in real time. 

• PG&E supports the idea to hold a workshop to hear from battery operators about their specific experiences and associated pain points in the CAISO markets. 

• PG&E requests the CAISO to restart the Storage as a Transmission Asset (SATA) initiative as a potential element of this initiative. 

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Although batteries are unique (along with pumped storage) in pairing consumption and generation of energy in the CAISO wholesale markets, they are not unique in needing to incorporate opportunity cost considerations into their bidding due to cross-temporal effects and values outside of market horizons.  At this early stage in the introduction of battery storage into the markets, PG&E believes the CAISO should take a conservative approach toward proposing changes in the definition of marginal costs or bid cost recovery.  Marginal costs should continue to be calculated based on incremental energy within a single interval, and consideration of opportunity costs in default energy bids (already discussed at length in the ESDER4 process) should be treated comparably to opportunity costs on other resources: I.e., the CAISO clearly does not question the existence of opportunity costs above default energy bid levels, nor does PG&E, but unless these can be clearly documented, they should not be considered for inclusion in the DEBs. 

Similarly, bid cost recovery has produced results that have disappointed resources bidding into the CAISO markets for more than a decade, but the fundamental principle continues to be fair and equitably applied: when a resource submits binding bids for a particular market (I.e., day-ahead or RTM), it shows a willingness to accept any level of revenue consistent with these bids, down to and including “zero” with respect to bid costs, totaled over the market’s defined scope.   

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Some merchant storage and merchant storage advocates have objected strongly to the use of the Minimum State of Charge (MSOC) constraint to protect peaking energy schedules from batteries during net load peak hours.  Before considering the solution directions suggested by the CAISO to address these objections, PG&E notes that at the end of 2021, as a result of ESDER4 implementation, scheduling coordinators for batteries will have the capability to set storage targets on an hourly basis in real-time, which will potentially provide them with the ability to guarantee the state of charge (SOC) required to support peaking schedules without direct CAISO intervention. 

There remains a risk that batteries receiving peak energy awards may find it economically justified to risk not being able to satisfy their discharge awards, even on days when taking this economic risk results in system reliability risks from the CAISO’s vantage.  The CAISO therefore has proposed the concept of an energy storage product to assure both that the CAISO meets its reliability needs and that batteries can be paid appropriately for holding storage in reserve. 

The CAISO discusses two possible approaches to paying for SOC.  The first approach is to define an “energy shift product” that gives the CAISO rights to both charge and discharge energy in a single transaction.  PG&E interprets this as essentially an opportunity cost payment for imposing a MSOC constraint on a resource that is discharged, or available for discharge, during some subset of hours, possibly defined by day-ahead market results.  One implementation, for example, would offer an opportunity cost-based payment for willingness to adhere to an MSOC constraint: this payment would be determined in the day-ahead market along with the day-ahead charge and discharge schedule consistent with the payment.  The CAISO could evaluate the combination of the charging cost and MSOC opportunity cost versus discharge value in the day-ahead optimization. 

A second implementation, and one PG&E considers much more promising, would be for the CAISO to procure storage capability to cover discharge needs from batteries in the aggregate, rather than tying that capability to particular battery charge-discharge schedules.  For example, the CAISO might procure 2500 MWh of discharge energy at the end of hour ending 18 as a kind of ancillary service, which may be provided by any battery willing to guarantee a SOC consistent with its award.  This product could be referred to as Energy Storage Reserve, since it represents the quantity of energy that the MSOC ensures is available to meet the CAISO’s needs during constrained conditions.  However, the storage obligation would not be tied directly to day-ahead awards, and so could be met by batteries even if they had no charge and/or discharge awards from the day-ahead market: for example, batteries with residual state of charge from previous days, batteries priced out of the money but capable of charge, and discharge for reliability needs, and batteries whose primary use cases are not based on CAISO market revenues. 

PG&E sees several benefits from the Energy Storage Reserve concept: it allows the CAISO flexibility in determining its aggregate SOC needs; could potentially be extended to real time incremental procurement of energy; allows the entire storage market to allocate the costs and revenues from energy obligations among themselves based on economic value; would enable an end of day storage obligation to be paid for and thus enable battery storage to be procured for the morning peak as well as the evening peak; and would make unnecessary the kinds of changes to optimization horizon that are not feasible for the CAISO to implement given current technology and algorithmic limitations. 

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

In the most general case, variable charging rates as a function of state of charge would require a complex space of battery operations, which would express both the operating mode of the battery (charging, discharging or idle) and a discretization of the battery’s state of charge to allow differential charging rates to be defined within a given range.  The complexity of such a model would likely be comparable, if not more complex, than the existing pumped storage and MSG multi-state models.  PG&E’s opinion is that it is not yet clear such a significant modeling effort would be justified, as the real purpose of what has been suggested so far is avoidance of operations in ranges that should be avoided, possibly associated with unusual stress on the battery. 

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

PG&E supports the inclusion of exceptional dispatch in the scope of this initiative.  CAISO operators will require both specialized tools and clear guidance as to how to access charge and discharge capabilities of batteries at the same time that they potentially provide their instructions as some combination of dispatch and state of charge targeting.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

PG&E has previously expressed its belief that the limitations of the OMS system in not adequately enabling restrictions to be reported on both discharge and charge capabilities and state of charge capabilities simultaneously. The limitations of the OMS system further complicate changes already being made within the RA Enhancements initiative. Two examples of this include outage substitution requirements and the treatment of planned vs. forced outages. Considering how overlapping events may affect batteries’ capacity to fulfill RA requirements as well as their exposure to undue UCAP adjustments, improvements to the OMS system should be considered in the scope of storage enhancements at this time. 

To explain this issue further, the OMS software currently does not offer the same functionality for non-generator resources (NGRs) as it does for other generation resources. NGRs are unable to submit overlapping outage cards which makes scheduling coordinators unable to accurately and appropriately report outages and de-rates on such resources, for example when a transmission induced outage occurs while an existing outage (e.g. with a different nature of work) is underway. Scheduling coordinators do not have the appropriate tools to effectively communicate both outages to the CAISO as per the requirements outlined in the CAISO Tariff and Outage Management BPM. 

Another issue that arose tangentially in the Issue Paper and discussions with market participants was that of the need for batteries to have more flexibility in adjusting their bids and schedules in real time.  PG&E suggests that one direction of development that would address this concern would be to make the hybrid model available to NGR resources not paired with other fuel types.  The hybrid model is a variant of the NGR model which consists of the following elements: 1) a single resource ID designating the output of one or more resources, 2) no enforcement of SOC constraints such as initial SOC, MSOC or end-of-hour SOC parameter (because there is no direct control of SOC in the market processes) and 3) the Dynamic Limit tool, which sets bid limits much closer to time of delivery in the RTM.   

 The disadvantage of the hybrid model, as PG&E understands it, is that use of the Dynamic Limit tool is almost certain to reduce the resource’s UCAP in the long-term, so that it will be unable to preserve its full dispatch range as RA capacity.  It’s worth pointing out that there is some vagueness about whether the hybrid model, at least without access to the Dynamic Limit tool, is already accessible to batteries that wish to choose it; although it has not been mentioned in any recent discussions, at one point that model was considered to be available to any battery just as the NGR-REM and NGR-LESR models were available. 

Also, although PG&E is sympathetic to the CAISO’s desire not  to lead endless open-ended stakeholder discussions of possible future innovations, PG&E does believe there would be value in holding a workshop specifically to hear from battery operators about their experiences in the CAISO markets so far, and their identification of pain points in the use of the current market models.  PG&E’s own battery experience has been limited to batteries two orders of magnitude smaller than some of those that have now been in the CAISO markets for a year, and PG&E would welcome insights that may assist it in the near future with its own impending large battery.  PG&E would prefer to see such a workshop distinguished from any discussions of specific future market design implementations. 

Lastly, PG&E requests the CAISO to restart the Storage as a Transmission Asset (SATA) initiative as a potential element of this initaitive. SATA was temporarily suspended in January 2019.  In its January 2019 presentation to stakeholders, the CAISO stated it would “reinitiate SATA once ESDER4 has resolved the identified gaps”¹(SATA Stakeholder Meeting Presentation January 14, 2019). PG&E applauds the CAISO for completing the ESDER4 initiative in September 2020 and understands that ESDER4 was able to address some of the gaps identified in the SATA initiative.  Also, the CAISO’s Hybrid Resource initiative successfully completed two phases and developed market enhancements for existing generation resources paired with storage.  In addition, PG&E looks forward to actively engaging in this recently launched  initiative addressing the remaining identified gaps and resolving concerns related to optimization, dispatch, and settlement of storage assets.  PG&E recommends the CAISO integrate the SATA initiative as a potential element of the Energy Storage Enhancements initiative.  

In the 2020-21 Transmission Plan, the CAISO identified two previously approved transmission projects that can be partially replaced by appropriately sited battery storage.  For the North of Mesa Upgrades Project, the CAISO is recommending procurement of a 50MW 4-hour BESS at Mesa 115kV substation to address a maintenance window need² (CAISO 2020-21 Draft Transmission Plan page 118). The Draft Transmission Plan points out that this storage recommendation will provide a sufficient maintenance window within the winter months for facilities in the  area as required by the CAISO planning standards.  It appears the CAISO-recommended storage resource at North of Mesa Project will operate as a transmission asset for a portion of the year.  In essence, through the TPP, the CAISO is recommending a SATA without the market rules and policy framework in place to enable storage resources to provide cost-based transmission service and to also participate in the ISO markets.   PG&E requests the CAISO commit to integrating the SATA in the Energy Storage Enhancements efforts as doing so will provide rate payer benefits and greater flexibility to the grid.  

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

PacifiCorp appreciates the opportunity to provide comments on this Initiative and appreciates the CAISO’s openness to shaping the scope of this important initiative in a collaborative way. As mentioned in detail in our specific comments below, PacifiCorp strongly encourages CAISO to examine practical option for extending the look ahead horizon of the market. We also believe that it is prudent for CAISO to ensure that the different needs and operational practices of EIM entities compared to CAISO are of sufficient focus in this initiative specifically as it related to ensuring state of charge.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

CAISO’s best practice indicates that entities submit updated base schedules and bids for a rolling 5 hours. While some organizations follow this guidance, it is our understanding that significant deviation in individual approaches by market participants may exist. We recommend CAISO consider turning this best practice into an enforced rule and updating its BPM to reflect this change. This would increase the quality and, perhaps more importantly, the completeness of the data, available for the optimization run thus improving the quality of the forecast beyond T-40. We recognize that such a change would likely trigger changes to participant’s OATT’s but maintain that fixing the underlying data challenges would represent a more holistic solution than a shortening or elimination of the advisory interval.

PacifiCorp considers expansion of the market’s time horizon as the optimal solution and strongly encourages CAISO to investigate what options in fact exist in the mid-to long term. Alternatives to extending the current market framework might be the adaptation of the DA mechanisms and/or use of hourly blocks to extend the market duration. In the meantime, PacifiCorp considers the addition of End of Horizon Opportunity Costs a possible solution for the short to mid-term. PacifiCorp requests that CAISO please explain in detail and with examples how End of Horizon Opportunity Costs would work.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Given the expected ubiquity of storage across the WECC in the future, PacifiCorp considers it prudent for CAISO to ensure that the rules for ensuring state of charge are designed in a way that considers EIM entities who experience different net load peaks than the CAISO and thus the rules and processes need to be sufficiently flexible to account for the differences in net load peaks for CAISO and EIM entities.

PacifiCorp strongly encourages CAISO to re-examine its position on extending the look ahead horizon of the market and consider possible solutions such as using hourly blocks and/or an adaptation of the DA process to inform the market runs over a longer period of time than status quo.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

PacifiCorp is supportive of including variable charging rates in the scope of this initiative.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

While PacifiCorp understands the interest in this topic as it pertains to storage optimization, we are concerned that providing operators with more ability to use out of market adjustments is not aligned with the philosophy of an organized market and is in opposition to the CAISO’s position during the Day Ahead Market Enhancement Initiative where CAISO stated that reduction of out of market adjustments was beneficial to the market function. In light of that, PacifiCorp strongly encourages CAISO to re-consider its position on this point.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Please see item 5 below as PNM would like to see some other items potentially included and prioritized in this initiative.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

PNM does not have any comments at this time on this item.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

PNM would be interested in seeing a tool to help manage state of charge included.  As an EIM Entity, PNM’s desire for SOC management may be slightly different than SOC management of energy storage directly in the CAISO. Some of the new projects PNM has coming online in the next year require that the energy storage be managed to a specific state of charge on average across the day (i.e. average SOC across the day should be less than 50%.  This implies that we can’t charge the energy storage to 100% and then leave it there for long periods of time or across midnight period for discharge the following morning ramp/peak).

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

PNM does not have any direct experience with this issue, but we wonder if multiple ramp rates could be modeled in the Master File (MF) based off of SOC instead of MW output like how the ramprate tab in the MF works today. 

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

PNM does not have any comments at this time as it does not appear to be a direct impact to EIM Entities. 

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

??????PNM would like to see the issue paper include the following items from an EIM participation perspective

• Include capability for participation of energy storage to charge based off bids while restricted from only charging from the co-located renewable resource.  As an EIM entity outside the state of CA, procurement of energy storage is being paired with renewable via Purchase Power Agreements.  The PPAs include language to take advantage of the investment tax credit (ITC) and restrict the energy storage from only charging from the co-located/paired solar.  PNM would like to see energy storage be able to be bid into the EIM for both charging and discharging to help with passing flex ramp sufficiency tests but doesn’t see that ability in current EIM rules to also respects the requirement of charging from the co-located solar.
• Provide ability to limit total discharge of an energy storage system over the course of day (i.e. 1 full charge/discharge cycle per day or 365 per year)

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

SDG&E appreciates that the CAISO is considering energy storage enhancements, and in general is supportive of exploring methods to improve energy storage resources participation in the CAISO market.  While SDG&E believes most of the possible enhancements are worth exploring, it may be worth prioritizing which enhancements should be addressed first in order to get the most benefit from this process.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

SDG&E agrees that issues related to marginal cost should be included in the scope for this initiative and should be a priority for the CAISO.  It will become at matter of increasing importance to allow energy storage resources to reflect their true marginal costs as accurately as possible as these resources continue to increase proportion on energy supply within the CAISO.  SDG&E is supportive of examining MIO and Spread bidding to determine  whether they will truly allow for a more accurate reflection of costs as examine potential unintended consequences of the modifications.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

The CAISO suggests a tool to combine the potential for reliability service with its market. Use of the tool is presumably restricted to summer or months forecasted to be stressed. The auction for this service will resemble an ancillary service process possibly with sub auctions required for LCR areas or Zones. SDG&E might support the MSOC as long as its use is restricted to a few critical days per year and/or its effects on market participation  is more carefully examined.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

The ability to have variable charging rates would be beneficial to optimizing battery energy storage resources participation in the market.  The CAISO says there is currently no way to do this, so it is worth exploring whether there would be a way to implement this capability.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

Exceptional dispatch should be a scope item for this initiative.  It is appropriate for Battery Energy Storage resources to be compensated for Exceptional Dispatch when the dispatch results in an opportunity cost loss.  Additionally if allowing a new type of exceptional dispatch to procure a specific state of charge would improve reliability then it should be considered, but this change should be examined closely to determine if it provides an actual benefit.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

The CAISO states that it is technologically infeasible to expand the Real-Time Market (RTM) and SDG&E agrees that currently expanding the RTM is technologically infeasible.  However, the CAISO should have a plan in place to expand the RTM or at least identify what would have to happen for the RTM expansion to become technologically feasible.  Per the CPUC Reference System Plan (D. 20-03-028) over 12,000 MW of battery storage capacity is expected to be online by 2030.  Whatever measures the CAISO adopts in this initiative will only be a stopgap.  To truly solve the problems that may arise from increasing battery storage resources, the CAISO will eventually have to expand the RTM.  SDG&E implores the CAISO to begin working on that RTM expansion plan now.

Additionally, SDG&E supports CAISO’s belief of holding off on any enhancements to BCR at this time.  Allowing these other enhancements to come to fruition first would be the best course of action in our belief. 

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

The Cities of Anaheim, Azusa, Banning, Colton, Pasadena, and Riverside, California (the “Six Cities”) are generally supportive of the broad scope the CAISO has outlined for this initiative, at least at its initial stage.  Although the Six Cities appreciate the CAISO’s efforts to use CAISO and stakeholder resources most efficiently, the Six Cities agree with suggestions made during the May 5^th stakeholder meeting that a technical workshop might allow for a constructive dialog regarding the range of topics that should be included in this initiative and their relative priorities. 

It also may be worthwhile to consider phasing this initiative into topics that consist of more limited tariff or market rule changes that could be stakeholdered and adopted on a comparatively short timeline, and topics for which either more study is needed or where stakeholder consensus may be less clear.  For example, one of the ways the CAISO identifies to address state of charge-related issues is to expand the look-ahead window for the Real Time Market.  The Six Cities do not question the CAISO’s conclusion that this would be a significant change, but there could be value in assessing the relative costs and benefits of doing so and considering how to weigh these costs and benefits given the changing nature of the CAISO’s resource fleet, the increase in expected storage resource deployment, and the tightening capacity market.  A longer time horizon to evaluate all of these factors in this initiative could be helpful.  In contrast to this issue, other topics within this initiative might lend themselves to resolution and implementation on a somewhat shorter timeframe. 

While the Six Cities look forward to participation in this initiative, at this time the Cities have limited comments regarding the specific topics that the CAISO has identified in the Issue Paper.  However, the Six Cities remain concerned that the CAISO needs to acknowledge and work with stakeholders to fully address grid-charging restrictions associated with storage resources that are funded through Investment Tax Credits (“ITC”).  In the recently-concluded Phase 1 Resource Adequacy Enhancements initiative, the Six Cities identified concerns with the CAISO’s seemingly unequal treatment of Co-Located Resources (relative to Hybrid Resources) in terms of their obligations to participate in the Minimum State of Charge (“MSOC”) program if these resources are used to provide Resource Adequacy capacity.  The Six Cities continue to urge the CAISO to ensure that storage resources are able to participate in CAISO markets to the maximum extent possible while not undermining their ability to comply with ITC-based charging restrictions.  The Six Cities will continue to evaluate the CAISO’s proposals with this principle in mind and urge the CAISO to consider the reality that a substantial number of new storage resources may have been financed and are economically viable only to the extent that ITC-related charging restrictions can be followed. 

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

The Six Cities do not have comments regarding this topic at this time. 

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

The Six Cities generally support inclusion of this topic within this initiative, but have not yet identified a preferred solution to this issue. 

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

 The Six Cities do not have comments regarding this topic at this time. 

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

The Six Cities do not have comments regarding this topic at this time, although the Six Cities would like to better understand the basis for the perceived need for Exceptional Dispatch of storage resources for purposes of managing their state of charge and reiterate the concerns explained above regarding the need to ensure adherence to ITC-based charging restrictions.  To the extent that the CAISO would exceptionally dispatch a resource to procure state of charge, then, again, the Six Cities request that the CAISO work with stakeholders to ensure that grid-charging restrictions are not violated. 

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

 The Six Cities do not have additional comments at this time. 

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Please see attachment.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

Please see attachment.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

Please see attachment.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

Please see attachment.

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

Please see attachment.

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

Please see attachment.

1. Please provide a summary of your organization’s general comments on the issue paper, including reprioritization (if any) of the proposed issues in scope):

Please email Kallie Wells at kwells@gridwell.com if interested in reading full comments with example pictures. The comment template cannot support pictures and despite many tries by us and the CAISO, our comments cannot be uploaded as an attachment.

Western Power Trading Forum (WPFT) thanks the CAISO for commencing this important stakeholder initiative focused on energy storage resources.  WPTF appreciates the CAISO working together with stakeholders to lay the foundation as the first step of the policy effort.  CAISO is starting by identifying and prioritizing issues related to the integration, modeling, and participation of energy storage in the ISO’s real-time market.  WPTF appreciates this approach, and that the CAISO is not prematurely defining the scope and/or schedule of the initiative.  This is inclusive of both ideas and stakeholders and will inevitably lead to more robust discussion and holistic policy development and result in more durable market design.

WPTF is intrigued by the thought of considering fundamental market changes and encourages the CAISO to work for robust long-term policy solutions.  The resource mix is evolving, and it is imperative to design market policy with this in mind.  We know we will continue to see increasing levels of energy storage and renewables combined with energy storage and we must develop the market for this resource mix as we look ahead.  Band aid solutions and design may provide a short-term fix but does not address the larger structural changes.  To be clear, WPTF is not opposed to implementing immediate changes to address known issues, but does not want those changes to derail or render more holistic changes needless. It is essential we design and implement robust policy from the beginning.  WPTF thanks CAISO for an open policy development timeline to allow for deep and enduring policy development.  

WPTF would like to take this opportunity to reiterate the importance of really considering all the elements needed for a well-functioning market that is comprised primarily of energy storage and renewable resources. Entities are making business decisions as we draft these comments, crafting and signing contract language that will influence how storage resources participate in the market. It is absolutely essential that the CAISO gets this right and is very transparent in how storage resources will be modeled, able to participate, and awarded energy and ancillary service products through the market. More importantly though, that the CAISO is motivated to consider market design changes that ensure successful integration of the resource mix. The CAISO and stakeholders really have to embrace the idea that the resource mix is changing. With that comes the recognition that what used to be efficient for a predominantly thermal resource fleet may impede the ability for storage resources to effectively participate in the market in a way that results in the optimal dispatch and use of the overall resource mix while maintaining reliability, which will be key in the CAISO continuing to reliably operate a renewable rich resource fleet. Over the past couple years, we have seen the CAISO continue to implement constraint after constraint in the market as a way to integrate storage resources. We suspect this is a direct result of thinking the current market design can be “tweaked” to manage storage resources rather than being open to fully integrating storage resources. While market constraints may work from a physical operational standpoint, and one may argue reliability perspective, it comes at the cost of enabling the competitive market forces to generate the optimal and most efficient use of the emerging resource mix. WPTF strongly encourages the CAISO to take a step back and really consider a market design that relies less on constraints and more on market based incentives and modeling functionality. We are not naïve in thinking this will be an easy and quick policy process, but are motivated and ready to take on the challenge as it will shape how the market functions over the next several decades.     

To effectuate such a paramount effort, WPTF requests a series of workshops to work through the range of concerns and potential for innovative ideas that arise from comments submitted on the Issue Paper before moving to the Straw Proposal. The Issue Paper is a good starting point for discussion, and the stakeholder call provided an opportunity for stakeholders to start considering and debating the variety of issues and topics to consider in this policy effort.  WPTF believes a range of reasonable concerns and potentially innovative ideas may come out of comments submitted on the Issue Paper, and these ideas warrant discussion before going to a Straw Proposal.  WPTF requests a series of workshop as an important next step in this effort prior to putting together a Straw Proposal. WPTF found the stakeholder engagement important on the initial call but there is more work to be done on identifying and elaborating on issues for energy storage resources and wholesale market participation which should be done prior to the CAISO crafting and publishing a Straw Proposal. Ideally the first workshop would consider if and how fundamental market changes are needed to fully integrate energy storage resources in a renewable heavy resource mix.

WPTF highlights some of the current issues with energy storage wholesale market participation here and elaborates on issues below.

2. Provide your organization’s comments on whether to include issues related to representing marginal costs as a scope item for this initiative (including whether to include modifying the Multi-Interval Optimization (MIO) and Spread bidding, and modifying the cost recovery rules for storage):

One of the most significant challenges facing energy storage resources today in the CAISO market is the impact of the advisory prices and shorter outlook horizons used in the real-time market. WPTF strongly encourages the CAISO to focus its efforts and prioritize addressing this issue, which may be accomplished through a more holistic review/fundamental changes to the existing market structure. There are already two potential solutions that have been put on the table for consideration – extending the real-time outlook horizon and implementing the end-of-hour state of charge biddable parameter.

Extending the real-time look out horizon has been a solution suggested by stakeholders for at least the past 5 years. Every time it is suggested, the CAISO’s response is its not technologically feasible to extend the outlook horizon. The response this time was unchanged. WPTF strongly believes there is the ability for the CAISO to achieve an extended outlook horizon, and while it may not be feasible given the computing power they currently have, that is not to say that the CAISO can’t buy more computing power such that extending the real-time lookout horizon becomes feasible. We understand computing power may not be cheap and there has been zero transparency regarding how much additional computing power would need to be purchased by the CAISO. Thus before the CAISO dismisses this solution (again), the stakeholders deserve to have an idea of what cost would be associated with increasing the real-time lookout horizon. It could be the case that the cost does outweigh the potential benefits, but the CAISO has yet to present any data/information that would enable the stakeholder community to opine on if its worth the additional cost or not.

WPTF also strongly encourages the CAISO to prioritize implementation of the end-of-hour state of charge biddable parameter. That market solution was initially proposed by WPTF in a stakeholder meeting 2 years ago. The proposal went uncontested by stakeholders throughout that policy process and approved by the BOG last year. We are disappointed to see that two years later it still has yet to be filed at FERC.

Additionally, even with the two aforementioned improvements to the current market design, WPTF believes that the market design resulting from this policy effort should focus on further mitigating the impact of binding and advisory prices on energy storage resources. The markets decision on when to discharge an energy storage resource is also linked to a prior decision to charge the resource. That is to say that the overall charging and discharging schedule of an energy storage resource is directly impacted by two distinct market prices in two different hours/intervals. Given that the day-ahead market optimizes over the entire trade day simultaneously and all hourly awards and prices are binding, the day-ahead market is better situated to optimally charge and discharge storage resources. However, the real-time market not only makes binding charging/discharging decision based on advisory prices it sees in a future time period relative to the current interval’s binding price, it also does not look out over the entire trade day but rather “sees” future periods in differing/overlapping chunks of time. The shorter, overlapping outlook horizons in combination with making binding dispatch instructions based on prices that may or may not actually materialize is one of the most significant challenges for the ability of energy storage resources to be optimally dispatched by the CAISO market, resulting in inefficient market outcomes and inability for storage resources to be able to fully capture their value in the market.

The example below is intended to illustrate this challenge that energy storage resources face on a daily, interval by interval basis. The blue cells are the binding dispatch instructions and energy prices resulting from that market run. In this example, the first market run determines its optimal to charge the resource in interval 1 (binding interval) based on the interval 1 binding price as well as the advisory prices for intervals 2-4. Under this charging/discharging schedule the resource has the potential to earn $2,625 and end with a 0% SOC. However, as the subsequent market runs continue and the corresponding prices from the first market run change, the charging/discharging schedule starts to diverge from what was initially considered optimal. By the end, the resource has only earned $125 in market revenues and is left with 25% SOC.

While we strongly support the CAISO considering fundamental market changes, if determined necessary, to address this issue, there are also interim steps that can be taken such that market participants can better understand the dynamics and impact of the advisory prices on energy storage resources. As a scheduling coordinator, its challenging to structure bids in a way that mitigates this risk, especially given that there is zero transparency on what the advisory prices are. Thus, one potential interim step would be for the CAISO to release the advisory prices such that scheduling coordinators have more information when structuring market offers.

3. Provide your organization’s comments on whether to include a tool to ensure state of charge topic as a scope item for this initiative and early preference for a solution to this issue:

As apparent throughout these comments, state-of-charge management for energy storage resources is fundamental; nearly every element of energy storage participation, modeling, and operational management is impacted by the SOC. While there are several other issues already mentioned that relate to SOC management, WPTF wanted to take this opportunity to raise other concerns/issues with how the CAISO market currently treats SOC.

In the market the CAISO currently applies what is referred to as an efficiency parameter to the energy storage to account for the operational reality that for every 1 MWh of energy stored, the energy storage is only able to discharge a fraction of that 1 MWh – typically in the range of 85% - 90%. WPTF believes there are two ways in which this parameter can be applied to the energy storage resource – limiting the amount it can discharge or reducing the amount of stored energy. While these appear at first to be essentially the same concept, WPTF is concerned that there are some operational realities that need to be considered.

Based on the formulation in Section 6.6.2.3 of the CAISO’s BPM for Market Operations, the efficiency parameter is applied to a battery storage resources charging schedule.  So, for every 1 MWh charged it can only dispatch .85MWh. This is reflected in a lower SOC value. As a result, the only way a storage resource will be able to reach 100% SOC is by actually charging 1.15 MWhs for every 1 MWh of energy storage capability. Take for example a 100MW, 400MWh battery. Based on our understanding of how the efficiency parameter is currently reflected in the market, for the resource to reach 100% SOC it will have to charge 470MWhs (assuming an 85% efficiency parameter). WPTF is concerned that from an asset management perspective this becomes a risk associated with over-charging the element. First, WPTF would like to confirm that our understanding is accurate in how the parameter is implemented. And secondly, WPTF would like to hear from developers if this is a risk and if so, is a potential solution to apply the efficiency parameter on the discharging schedules? For example, the same resource would charge 400MWhs to achieve 100% SOC but then still only be able to discharge 85% of that by limiting the discharge schedule to 85% of the total stored energy (e.g., 400MWhs in this example). Thus, WPTF offers that it may make more sense to have the efficiency parameter on the discharge side of the resource.

Additionally, WPTF believes this effort should strongly veer away from considering any state-of-charge constraint in real-time to ensure energy storage resources are able to meet their day-ahead schedule, similar to the recently proposed MSOC. The day-ahead market is, for all intents and purposes, a financial market (with the exception of long-start resources). The real-time market is the market that creates physically binding awards and schedules to meet real-time load conditions. The concept that energy storage resources should be held to their day-ahead schedule in a market that inherently reflects different conditions, resource schedules, and market prices, is flawed.  WPTF strongly agrees with the perspective of Vistra that was eloquently articulated in recent FERC comments. Specifically, that “reliability and economic efficiency would be best served by giving the resource the ability to choose to respond to real-time information – which may diverge from the day ahead schedule.” Thus, WPTF believes that any market design change intended to help manage the energy storage resources in the real-time market should be market-based solutions and incentives rather than physical constraints that lead to inefficient outcomes and potentially result in reliability issues.

4. Provide your organization’s comments on whether to include variable charging rates as a scope item for this initiative:

5. Provide your organization’s comments on whether to include exceptional dispatch as a scope item for this initiative:

6. Provide any additional comments on the issue paper, or any additional scope items your organization feels should be included for this initiative. You may upload examples and data using the “attachments” field below:

Energy Storage Resources Providing Regulation

The Energy Storage Enhancements effort should ensure the market is able to access regulation capabilities efficiently and effectively from energy storage resources. Currently there are four known issues, discussed in more detail below, with how the market awards and dispatches energy storage resources providing regulation. These issues are resulting in an inefficient and in some cases infeasible outcomes. First, the day-ahead market does not take into consideration expected deployment of ancillary services when awarding in the day-ahead market and this can result in infeasible day-ahead schedules. Second, WPTF has concerns on the conflicting minimum state-of-charge (MSOC) recently filed at FERC and the regulation state-of-charge (SOC) requirements put forth in Proposed Revision Request (PRR) 1334. Third, there is minimal transparency into if/how the CAISO will model and certify the battery component of a co-located resource for regulation that is unable to charge from the grid. And lastly, the way the CAISO currently determines the regulation range for hybrid resources limits the market’s access to the full regulation capability of the resource.

Infeasible Day-ahead regulation awards

The CAISO market awards energy storage resources regulation capacity in a way that results in infeasible schedules and thus inefficient market outcomes. The CAISO market procures 100% of its regulation up and regulation down capacity in the day-ahead market. Because the market co-optimizes regulation and energy in the day-ahead market, infeasible regulation awards then result in inefficient market outcomes as the market comes out of the day-ahead process with schedules and awards that will, by structure, have to be modified in real-time.

The day-ahead market does not assume any deployment of regulation when awarding energy storage resources regulation up and/or down. The actual deployment of regulation in real-time directly impacts the resources state-of-charge in that moment and going forward. Thus, without the day-ahead market adjusting the state-of-charge used in the day-ahead market for some assumed level of regulation deployment, it sets up the energy storage resources in such a way that come real-time they cannot provide the awarded day-ahead schedule even absent any deviations from the day-ahead schedule prior to the awarded regulation hours.

For thermal resources, this is not necessary as they have the ability to provide the regulation hour after hour independent of (1) its energy schedules prior to and during the awarded regulation hours and (2) how the market is deploying the resource for the regulation capacity. Storage resources, on the other hand, cannot continually provide the same level of regulation hour after hour because the state-of-charge is directly impacted by (1) its energy schedule prior to and during the awarded regulation hours and (2) the actual deployment of the regulation capacity.

Take for example a 100 MW/400 MWh battery resource that has been certified to provide 200 MWs of regulation up and down and receives the following day-ahead schedule. Note that the following schedule is considered a feasible schedule from the day-ahead market (i.e., respects the energy storage modeling constraints in the market). Also assume it has a 0% SOC at the beginning of HE 9.

In real-time, assume the resource is actually deployed for approximately 40% of its awarded regulation capacity. The following chart compares three different state-of-charges: (1) Day-ahead SOC based on day-ahead market awards (2) Real-time SOC based on day-ahead energy awards and actual deployment of regulation and (2) Minimum SOC required for resource to provide its day-ahead energy and regulation awards. Its important to note that the real-time SOC assumes the resource is following its day-ahead schedule (which is feasible in real time up to a point identified below) and is being deployed for 40% of its awarded regulation up.

Based on the chart above, one can make the following observations:

• HE 11 the day-ahead and real-time market SOC start to deviate because the resource is being used for the regulation up capacity provided
• The resource has sufficient SOC to support both its day-ahead energy and regulation awards up until midway through HE13
• HE 13 int 7, the real-time SOC is not high enough for the resource to continue providing its regulation (recall no energy schedule until HE 15 again)
• HE 15 the resource does not have sufficient SOC to support its energy schedule
• HE 15 int 6 the resource is depleted and cannot meet its energy schedule

WPTF believes rather than having to implement additional SOC constraints (similar to the one discussed in the next issue), there is a more market based approach that will not only better align the SOC between the day-ahead and real-time markets but result in a more efficient outcome. For example, the CAISO could consider adjusting the SOC in the day-ahead market using some assumed level of regulation deployment in real-time. WPTF recognizes that there will be the challenge of deciding what level of deployment to assume, but any non-zero value will be better than currently assuming none of the regulation capacity awarded will actually be needed and called upon.

Conflicting SOC constraints

WPTF appreciates that the CAISO recognized the issue with real-time SOC when resources are being deployed for the regulation capacity. In part to address the concern, the CAISO issued draft BPM language in PRR 1334. PRR 1334 proposes to ensure Non-Generator Resources (NGR) providing regulation have enough SOC in the real-time market to continuously deliver the awarded regulation for at least 30-minutes. WPTF submitted comments (summarized below) on the PRR, but brings the issue up here as well.

In the submitted comments in the PRR process, WPTF noted two main concerns. First, the proposed regulation SOC constraint conflicts with the CAISO’s recently FERC filed Minimum State of Charge constraint (MSOC). Specifically, when a resource is providing regulation down, the proposed SOC constraint will require the resource to be partially discharged and impose a maximum SOC such that it has sufficient remaining charging capability; meanwhile the MSOC constraint will impose a minimum SOC. WPTF is concerned with how the market will simultaneously have the resource meeting both maximum and minimum SOC constraints.

For example, take a 100 MW/400 MWh battery resource with an energy schedule of 100 MWs from HE 17- HE 20. Under the MSOC proposal, this resource will need to be fully charged by the end of HE 16. Assume in HE 16 it has a 0 MW energy schedule but providing 100 MWs of regulation down. Based on this PRR, it is WPTF’s understanding that the market will ensure the resource has more than 50 MWh of charging capability remaining throughout the entire HE 16. In other words, this PRR will not allow the resource to be more than 87.5% charged in HE 16. However, the MSOC proposal will require the resource to be 100% charged by the end of HE 16. WPTF is unclear how these two constraints will be satisfied by the market optimization under this scenario. In the CAISO’s response, they noted that the market will prioritize energy (i.e., MSOC constraint) over the regulation SOC constraint as that is how the market today prioritizes energy over regulation. However, WPTF questions if this prioritization is appropriate and aligns with how the market currently prioritizes awarded day-ahead regulation schedules.

The second issue WPTF raised is that the proposed regulation SOC will require a resource to have a SOC sufficient enough to support 30 mins of continuous deployment of the regulation award. However, it is unclear to WPTF why the CAISO would require a resource to be able to provide 30 mins of the regulation up/down if less than 30 mins of the trade hour during which the resource is providing the regulation capacity remains. Take for example the same 100 MW/400 MWh resource but assume it is providing 100 MWs of regulation up in HE 14 and nothing in HE 15. Based on this PRR, the resource will be required to have at least 50 MWh of stored energy through the entire HE 14 to support its regulation up award. If the market is now in the last 15-minute interval of HE 14, the maximum regulation up dispatch signal the resource could possibly be called to provide in that trade hour would require 25 MWhs of stored energy (100 MWh dispatch for regulation up but for only one 15-minute interval). WPTF questions why it is necessary to require 50 MWh of stored energy in this case. The same is true for the RTD market; once the market is at the 7^th 5-minute interval in HE14, the required amount of stored energy per this PRR will exceed the maximum amount of energy that could possibly be required based on the regulation up dispatch.

Here again, WPTF believes implementing more of a market based approach such as the one described previously would result in a more efficient outcome.

Regulation Capabilities reflecting ITC

The majority of storage resources coming online in the CAISO market will be unable to charge from the grid due to the ITC; the ITC only allows energy storage resources to charge from renewable generation. There has been ample discussion in the industry, and quite frankly heartburn, over the unknown with regards to how the CAISO market will take into consideration this constraint. While there are several issues throughout these comments related to the ITC, this one is specific to how the CAISO will certify energy storage resources to provide regulation that are subject to the ITC and how the market will also see that constraint.

Specifically, WPTF would like CAISO to provide comment on how they will consider regulation for co-located storage resources that have a pmin of 0MW because they cannot charge from the grid.  Providing regulation is a known and key benefit of energy storage, and the market should ensure it continues to be able to access that capability.  Regarding the certification process, does the CAISO envision only certifying the storage component of the co-located resources and still enable the battery to be fully certified for its regulation range (from negative pmin to positive pmax) or is it only tested for half of its capability (0 pmin to pmax)? Furthermore, regardless of how the certification is conducted, will the CAISO expect to then see the regulation range reflected in the Masterfile to be from 0MW to the certified upper regulation point (e.g, pmax of the energy storage resource) such that the resource cannot be awarded regulation down that would then result in a potential negative AGC signal?

Regulation Range for Hybrid Resources

WPTF encourages the CAISO to modify market rules allowing hybrid resources to provide regulation, even when operating at the upper end of its operating range.  Today, the CAISO limits the ability for hybrid resources to only provide regulation at the lower end of its operating range. For example, if there is a 60 MW solar paired with a 10 MW battery, today the CAISO will certify the hybrid facility to have a regulation range from -10 MW to 10 MWs (assuming no ITC). Thus anytime the solar facility is generating more than 10 MWs, the resource is considered outside of its regulation range. When a resource has an energy schedule outside its regulation range, the resource cannot be awarded regulation up or down. Assume the resource had a solar forecast of 30 MWs but the battery had 10 MWhs of stored energy. The battery can still provide 10MWs of regulation up while the solar is generating 30 MWs of energy; however, the current market design does not allow that to occur. Rather the resource is restricted to only providing 30MWs of energy. 

This issue and a potential solution are depicted in the charts below. The potential solution would allow the regulation range to essentially float around the energy schedule of the renewable component. This would then enable the CAISO market to access the regulation capability of the battery storage resource regardless of the renewable component’s forecast.

Other Concerns

WPTF has three additional concerns/issues with how energy storage resources currently participate in the market that we feel warrant raising in this forum.

First, under the co-located model it can be the case that the underlying renewable and energy storage resources are operated with two different scheduling coordinators. As such, they are unable to coordinate bidding strategies between the two resources and the market does not have any link between the two resources other than the ACC, which only limits the total output to the interconnection amount. The inability for the market to coordinate results is an inefficient market solution. For example, assume the solar resource has a forecast of 50MWs and the energy battery resource is scheduled to charge 20MWs. The solar resource would have an energy schedule of 50 MWs and the energy storage a charging schedule of 20MWs, thus the net output to the grid is 30 MWs and the energy storage resource is charging from the solar output meeting the ITC requirements. It is our understanding that currently if the renewable resource is curtailed in the real-time market based on economic offers submitted, the energy storage resource would still be scheduled to charge. However, under this scenario, the energy storage resource would then have to charge from the grid (which its unable to do as a result of the ITC) or deviate from its scheduled (which is potentially a violation of the CAISO tariff). WPTF believes this in an inefficient market outcome that, given the prevalent energy storage resources under ITC restrictions, should be addressed.

Second, WPTF would appreciate the CAISO commenting on how energy storage resources under ITC restrictions are expected to be modeled in the CAISO market. Specifically, if they have software such that the resource will physically disconnect from the grid if energy is flowing from the grid to the energy storage resource (i.e., charging from the grid) would the CAISO expect the Masterfile to reflect a negative minimum load value?

Lastly, WPTF would also like to explore energy storage, a default energy bid, and the ability to bid above $1,000/ MWh. During a recent CAISO meeting, it was noted that energy storage resources, regardless of if imports or other internal CAISO resources are able to bid above $1,000/MWh under the recently FERC approved Import Bidding and Market Parameters filing, cannot bid above $1,000/MWh. The justification provided is that the only way internal resources can offer above $1,000/MWh is if they request, and the CAISO accepts, a reference level adjustment request. However, given that energy storage resources do not have a fuel cost component, WPTF would appreciate additional clarification and confirmation from the CAISO that energy storage resources can request reference level adjustments if/when the opportunity cost based on forecasted market prices results in costs greater than $1,000/MWh.

Thank you for consideration of these comments.