Correction Method of Baseline Load Estimation Result Based on the Identification and Evaluation of the Rebound Effect

As the basis for incentive-based demand response (DR) settlement, accurate customer baseline load (CBL) estimates are critical. However, the existence of the rebound effect will cause users to rebound their load after participating in incentive demand response events. This phenomenon will change the user’s electricity consumption pattern, and if the additional factors introduced by the rebound effect are not considered, estimating CBL using traditional methods will have significant bias. Therefore, this paper proposes a method for the identification and comprehensive evaluation of the rebound effect. After that, a neural network is built to get the mapping relationship between rebound effect evaluation index and the CBL deviation. Finally, a way of getting the deviation of the CBL result caused by RE is proposed. In order to correct the CBL, the paper first uses wavelet decomposition to identify the rebound effect, and then, this paper established the comprehensive evaluation index system to comprehensively evaluates the rebound effect. Finally, the BP neural network is selected to correct the deviation of the baseline load estimation result caused by the rebound effect. Simulation experiments show that this method can correct the deviation of baseline load estimation results caused by rebound effect.