A Decision Framework for Optimal Pairing of Wind and Demand Response Resources

Day-ahead electricity markets do not readily accommodate power from intermittent resources such as wind because of the scheduling difficulties presented by the uncertainty and variability in these resources. Numerous entities have developed methods to improve wind forecasting and thereby reduce the uncertainty in a day-ahead schedule for wind power generation. This paper introduces a decision framework for addressing the inevitable remaining variability resulting from imperfect forecasts. The framework uses a paired resource, such as demand response, gas turbine, or storage, to mitigate the generation scheduling errors due to wind forecast error. The methodology determines the cost-effective percentage, or adjustment factor, of the forecast error to mitigate at each successive market stage, e.g., 1 h and 10 min ahead of dispatch. This framework is applicable to any wind farm in a region with available pairing resources, although the magnitude of adjustment factors will be specific to each region as the factors are related to the statistics of the wind resource and the forecast accuracy at each time period. Historical wind data from New England are used to illustrate and analyze this approach. Results indicate that such resource pairing via the proposed decision framework will significantly reduce the need for an independent system operator to procure additional balancing resources when wind power participates in the markets.