مدیریت انتقال ریسک محدود در بازار عمده فروشی برق

Risk resulting from transmission loading relief calls made by transmission system operators can be managed with information estimated by a statistical model capable of predicting one day in advance the probability that a particular wholesale power transaction might be curtailed. The model predicts this probability with a reasonable degree of accuracy using information on variables that can be obtained publicly.

The North American Electric Reliability Corporation (NERC) has established procedures by which transmission system operators can manage the flow of power on lines that approach their loading limits – that is, become congested. These procedures, which are known as transmission loading relief (TLR) calls, are used by transmission operators to limit flows on congested lines so that transmission reliability standards, also set by NERC, can be maintained. TLR calls curtail transmission transactions in real time with very little notice. For power traders who engage in power transactions on a daily basis and rely on short-term non-firm transmission reservations to support them, TLR calls can prove costly when expensive substitutes must be found for the curtailed power. Consequently, in view of the uncertainty about the firmness of transmission service, power traders need to consider transmission curtailment risk in valuing the forward transmission reservations they have made to support their power trades. This article lays out a transmission curtailment risk management process that uses the logistic or “logit” statistical model to make day-ahead hourly predictions of TLR curtailment probabilities. This article draws upon earlier research (Morey et al., 2007) sponsored by the Electric Power Research Institute (EPRI). The article is organized as follows. Section II defines transmission deliverability risk and the typical circumstances that give rise to curtailments on transmission interconnections within the Eastern Interconnection. Section III reviews our TLR probability modeling method, which uses the logit model and depends upon publicly available data. Section IV addresses the predictive accuracy of the TLR forecasting model. Section V describes the process by which an estimated TLR forecasting model can be used on a day-ahead basis to assist a power trader in risk adjusting a transmission reservation. Section VI provides an example that illustrates the theoretical discussion of previous sections. Section VII offers conclusions.

This article explains how power traders can evaluate transmission curtailment risk through the use of a logit model to better forecast transmission loading relief calls that can curtail wholesale power transactions. The TLR forecasting model enables a trader to predict the probability that a curtailment-level TLR call will be made in each hour of the following day on a transmission interconnection over which the power trader holds a firm or non-firm reservation. The predicted probabilities can be used to risk-adjust the expected value of the transmission reservation in either of two ways. The risk-adjusted value, in combination with other market information that a power trader may possess, can be used to make decisions about holding, buying, and selling short-term non-firm reservations on particular transmission paths.