AriSEIA Submits Comments on APS VPP Valuation Methodology

READ THE FILING

Arizona Corporation Commission
1200 W. Washington Street
Phoenix, AZ 85007
 
RE: APS Virtual Power Plant (BYOD) Pilot Program, Docket No. E-01345A-22-0144
 
Chairman, Commissioners, and Staff,
 
Arizona Public Service (APS) is predicting unprecedented load growth over the next decade. To meet this rising need, the utility must aggressively add capacity which will put dramatic upward pressure on rates. One way to mitigate that upward rate pressure is to avoid direct utility investments where possible and to leverage customer owned assets to provide services that would otherwise require utility investment and risk ratepayer funds. To this end, the Commission ordered Arizona Public Service (APS) to implement the money-saving Bring Your Own Device (BYOD) program (also known as a virtual power plant (VPP)) which uses customer owned batteries to meet peak demand. The evidence in the rate case found that such a program could give APS access to batteries at a cost well below the cost of utility owned options.

AriSEIA/SEIA have raised several concerns about APS’s approach and its valuation methodology during the ongoing BYOD stakeholder process. APS has expressed consistent opposition to this program that could displace some utility investment opportunities from the start so it is unsurprising to AriSEIA/SEIA that APS’s methods and assumptions are designed to skew the outcome of this process in its favor. While we have some questions about the program implementation costs, the primary issue is with the value of avoided capacity and energy. The Company’s approach for valuing these categories does not reflect actual conditions on the ground and, as a result, produces avoided cost values that are substantially lower than justified.  Further, under the Company’s value, nearly all potential value from participating perfectly will be eaten up by the opportunity cost of forgoing peak time of use (TOU) reductions. Stated more plainly, APS’s proposal undervalues the important services BYOD can provide to such an extent that ratepayers will be unlikely to participate at all and would be better off simply reacting to the TOU rate. Rather than provide the appropriate price signal to customers to manage their battery for the broader good of the grid and all ratepayers, APS’s proposal will result in retrenchment to optimizing individual bill savings.

Capacity Value Issues

APS had indicated that it will use the Public Utility Regulatory Policies Act (PURPA) avoided capacity cost for the BYOD. Based on its Federal Energy Regulatory Commission (FERC) filing, this appears to be roughly $80/kW-year.  There are two key issues with using the PURPA avoided capacity value as a proxy for avoided capacity costs. 

First, the Company’s latest PURPA filing assumes that customer-sited microgrids are the capacity resource of choice. This is a non-conventional and inappropriate choice. APS was recently prevented by the Commission from pursuing additional customer-sited microgrid projects due to the impingement of its utility monopoly into a competitive market. This alone disqualifies this technology as the future capacity resource of choice, and any PURPA capacity costs based on this technology. Further, the Company has projected more than 4,000 MW of new capacity need over the following decade, and its integrated resource plan (IRP) shows that it will be procuring copious amounts of centralized generation and battery storage. Using a more conventional avoided resource such as a battery or gas combustion turbine (CT) or Company unit is more appropriate. AriSEIA/SEIA’s filing in this case used the annual revenue requirement of a utility-scale battery as a proxy, finding its avoided capacity value was north of $200/kW-year.

Second, PURPA is not driving investment in Arizona. The Company indicated that it has not signed any new PURPA contracts for years and the IRP filing does not rely on this financing mechanism for future procurements. Rather, it is procuring resources through competitive all-source contracts and through market-based purchases. Tying the value of this program to a moribund policy that is not producing new capacity is simply the wrong framework and should not be approved.

Energy Value Issues

The Company is using a simulated dispatch model to project avoided energy costs. This approach is necessarily tied to the input assumptions used and will necessarily not reflect actual real-world conditions, particularly during the scores of high-load/high-cost hours that the BYOD program will target. APS indicates that the average avoided energy cost is between $33/MWh and $52/MWh, a value that is simply inconsistent with historic market purchases during high-cost hours.

APS provided the quantity and price of its wholesale market purchases from 2018-2022 in the rate case.[1] An analysis of this data shows that the Company routinely paid in excess of $200/MWh for market purchases, with occasional purchases in excess of $1,000/MWh.  AriSEIA/SEIA’s analysis showed that the average weekday market purchase cost between 2019 and 2022 was over $100/MWh between 5 PM and 9 PM, the exact hours the BYOD program would target.[2] 

But if one looks at the actual highest-cost purchases, the avoided energy potential is much higher. AriSEIA/SEIA determined the 500 highest cost hourly purchases throughout the year and then analyzed the purchases that fell in the core summer months of June to September from 2018 through 2022.[3]  The results show that year after year, the avoided energy cost values during the highest cost hours are at times an order of magnitude larger than what APS proposes.

Even in 2019, which was an outlier in terms of the low quantity of high-cost market energy purchases, the average purchase during the high-cost hours was nearly $400/MWh. In 2021 and 2022 (and likely 2023), the price and quantity of high-cost purchases surged, with the average high-cost hour moving north of $800/MWh. 
Against this irrefutable historic purchase data, which has cost the Company’s customers tens of millions of dollars per year, the Company’s offer of as little as $33/MWh in avoided energy is simply unacceptable. AriSEIA instead recommends a minimum value of $500/MWh for avoided energy purchases for the BYOD program, a value which is roughly 1/3 of the highest cost purchase in recent years.

APS’s Proposal Is Eroded by TOU Opportunity Costs

By lowballing the avoided cost value on both the energy and capacity side, APS is setting the BYOD program to undercompensate customers for the real services they provide which will cause the program to fail. The residential value available to customers is only $40/kW-year based on the average reduction over the course of a program year.  The Company is authorized to call up to 60 events of up to 4 hours each between the hours of 4 PM and 10 PM.  Given that only half of these hours fall within the current peak TOU period (and fewer than half if events are called on weekends), customers participating in the BYOD event will have to consider the opportunity cost of peak TOU reductions. In other words, customers electing to participate in BYOD and give APS the ability to access the energy and capacity when they need it, may be inclined to simply participate in the TOU rate design thereby depriving APS and all its customers of the savings possible with BYOD.  

APS’s current peak and off-peak rates are roughly $0.34/kWh and $0.12/kWh, with a rate spread of $0.22/kWh.  Each kWh that a customer discharges their battery during an off-peak hour event call thus has an opportunity cost of $0.22. From this, it is possible to calculate the total opportunity cost that a BYOD customer faces by participating in the program.

In a best-case scenario, every event would be as short as possible (perhaps 2 hours) and fall during the peak TOU period.  In this case, discharges from the battery would not incur an opportunity cost and the customer could capture the full $40/kW-year benefit the Company proposes. In a worst-case scenario, every event would be 4 hours long and fall on weekends. This means that 100% of the event hours would incur the opportunity cost. A middle of the road scenario might assume event calls from 6 PM to 9 PM weekdays, with 2/3 of the hours occurring off-peak.
Suppose one analyzes an 11.5 kWh battery. The results of the three different scenarios are tabulated below. Even in the middle scenario, so much of the value of the program is eaten up by the opportunity cost that it is hardly worth the effort for a customer to sign up for the program. And in a worst-case scenario, it actually costs the BYOD customer money to participate in the program.

APS’s Stakeholder Process Violates Order No. 79293

Order 79293 (the “Rate Case Decision”) orders APS to “meet with AriSEIA/SEIA and any other interested parties to discuss collaboratively and attempt to reach agreement on the language of the BYOD Pilot POA.”[4] APS has held several stakeholder meetings in which it has told the stakeholders what it plans to do. AriSEIA requested an additional meeting to walk through the concerns identified above. APS sent out the proposed valuation that same afternoon. At the following stakeholder meeting, several stakeholders pointed out that the APS methodology was flawed and the valuation was too low and would make the program unsuccessful. APS requested stakeholders provide feedback in writing. Stakeholders asked for the Company’s workpapers. Stakeholders received two spreadsheets with two business days to review before the date in which APS asked for written feedback.

To date, it does not appear that APS plans to consider any stakeholder feedback in the plan of administration (POA) it intends to file within a month. APS is going through the motions of a stakeholder process, but there is nothing to indicate they intend to discuss collaboratively or attempt to reach agreement with stakeholders.

Respectfully,
 
Autumn T. Johnson
Executive Director
AriSEIA 
(520) 240-4757
[email protected]

[1] AriSEIA 4.03_ExcelAPS22RC03362_Hourly Market Purchases 2018-2022

[2] Lucas Direct at 59.

[3] This is twice as many as are allowed in the BYOD program, which authorizes 60 event days with events up to 4 hours.

[4] APS Rate Case Decision, Order No. 79293, 452:17-18, available here https://docket.images.azcc.gov/0000210704.pdf?i=1722230808744.