Energy storage configuration and day-ahead pricing strategy for electricity retailers considering demand response profit

Abstract Real-time price(RTP) is one of the common methods for electricity retailers to adjust load demands and gain revenue from demand response(DR) programs. However, effectiveness and accuracy of the RTP directly affect the DR quantity, which reflects as revenue returns in DR programs. A method is proposed to maximize profit of the electricity retailer by configurating energy storage system(ESS) and coordinate the operation of ESS with RTP to participate in DR programs, which enriches the profitability measures of retailers. Firstly, an improved CNN-LSTM algorithm is utilized to establish the DR dynamic characteristic model with expected consumption as input and RTP as output. Thus, the demand curve during each period can be generated, from which optimal RTP with maximum profit of the retailer can be estimated. Secondly, a bi-level optimization model is established to maximize the profit from the DR program in a whole day. The upper level real-time pricing model takes maximum profit of the retailer as the objective by optimizing the RTP with fixed ESS configuration. The lower level ESS configuration model takes maximum profit of ESS as the objective with RTP fixed by the upper level model, from which the rated power, capacity and daily operation strategy of ESS can be optimized and sent to the upper level model. Finally, the optimal profit gained by the retailer can obtained by iterative calculation between the upper and lower level models. Based on the historical data from PJM market, the case demonstrates that the algorithm proposed can describe the DR dynamic characteristic effectively and accurately. Moreover, by configurating ESS, the retailer’s profit extra increases by 7.19%.