What follows is my submission to the IESO Renewable Integration stakeholdering initiative [SE-91] wherein I oppose the IESO’s proposal to transfer the cost of wind forecasting from wind power generators directly to consumers. I believe that approximately $500K/yr is at stake in this discussion.
Thank-you for the opportunity to make a late submission regarding the “Renewable Integration Design Principles” (SE-91) as presented in the conference call the IESO hosted on February 1, 2010.
The main purpose of this submission is to oppose the IESO’s revised proposal which now would bill consumers for the cost of centralized wind production forecasting.
At the September 23, 2010 meeting, the IESO proposed a generator-pays model for centralized forecasting. My recollection of that meeting was that this proposal was uncontroversial and widely agreed. The minutes of the meeting, issued October 28, do not indicate any backlash from the generators on this point.
I note that the Wind Power Standing Committee web page does not include a link to the SE-91 page, wherein the IESO reversed its approach to forecasting cost responsibility. Observers of the Wind Power Standing Committee who had not received specific notice of SE-91 might have had difficulty following the IESO’s change in thinking which appears to have surfaced publicly for the first time in a SE-91 discussion paper dated December 9, 2010.
In the discussion paper, the reversal of the generator-pay position of the IESO is not acknowledged. The IESO’s justification for the new consumer-pay approach is that the forecast is analogous to the purchase of ancillary services. The IESO argues that better wind forecasting will improve the market’s efficiency and security. While it is true that better forecasting will enhance the market, the analogy to ancillary services is lacks a logical foundation.
The reason ancillary services are paid for through uplift is simply because there is no clear method to allocate the costs to a specific market participant who either caused the cost or who directly and uniquely benefited from the service. In the case of wind forecasting, the specific market participants causing the cost are the wind generators. There is no reason to apply the cost to other market participants.
The underlying logic of Ontario’s Market Rules includes requiring generators to report their availability and to offer their generation into the market. To file their availability reports and make their offers, generators naturally have to forecast their fuel availability. All of this work is performed at the generator’s own cost. The wind resource forecast is the equivalent of the fuel procurement process of a fossil or nuclear generator, the costs of which are borne by the generator. The fact that wind generators have difficulty forecasting their near term production and therefore complying with the Market Rules, is a practical problem associated with their technology. Centralized forecasting is designed to mitigate the failure of the wind generators to comply with the Market Rules. The challenge that wind operators have in meeting the requirements of the Market Rules is something wind developers have known in making their investments. While it is true that reliable, predictable generator behaviour and compliance with the Market Rules benefits the market, this fact does not justify recovering from consumers a cost that generators in the normal instance bear themselves.
During the all the public consultations sessions hosted by the OPA discussing the development of the FIT program — all of which I attended — I do not recollect any instance when the financial models the OPA was developing considered transferring the burden of Market Rules compliance with respect to availability notification from generators to rate payers.
The IESO’s proposal to burden rate payers with the costs of centralized wind forecast appears to be a retroactive and unfair change in cost responsibility and should not be pursued.
In addition to expressing my concern with respect to the allocation of forecasting costs, I also want briefly comment on the the IESO’s proposal to relax the dispatch instruction compliance deadband rules for wind generations, the calculation of deemed generation, and the overall approach for considering the eligibility of wind generators to receive CMSC payments.
The IESO is proposing not just to transfer forecasting costs to consumers but also to provide relaxed dispatch compliance deadband rules for wind generators. It is foreseeable that these relaxed deadband rules will increase consumer costs as the size of the wind fleet increases. Operating reserve and ancillary services requirements associated with wind generation are under review through the NERC Integration of Variable Generation Task Force. I urge the IESO to closely monitor the record of wind generators with respect to dispatch compliance and the resulting cost implications. In the event that cost to consumers become significant, less relaxed compliance rules should be considered.
During the conference call, I requested, albeit without clear articulation, that the IESO ought to carefully calculate the deemed generation upon which generators will receive payment. I urge the IESO to make every reasonable effort to ensure that ,in the determination of deemed generation, rate payers not be charged for generation that a wind operator could not in fact have delivered, had they not been constrained off. Some rigorous method of verification will be required to confirm availability claims of wind generators during constrained off events.
Regarding the IESO’s proposal to make Congestion Management Settlement Credit (CMSC) payments available to wind generators, the discussion paper of December 9th and subsequent discussion slides provide very little background information. No estimates of costs are provided or review of alternatives. For the purposes of these comments, I am assuming that any CMSC payments to wind generators will simply be part of the make whole payment to keep them in line with their FIT contract payment. If this assumption is incorrect, I would ask that the IESO provide a more complete explanation. One substantive issue to consider is whether CMSC payments are most likely to arise during periods of low HOEP. If this is the case, then I would suggest that it is important to ensure that these costs to be not be recovered in a way that associates the recovery with demand charges. To optimize the efficiency of the system and to minimize overall costs to consumers, demand charges should recover demand-related costs only. The IESO should consider the proposed CMSC structure from the perspective of efficiency taking into account the overall design of the market.
All of which is respectfully submitted,
February 2, 2010