Refined peak shaving potential assessment and differentiated decision-making method for user load in virtual power plants

Abstract

There is a consensus regarding the need to realize the transformation of renewable energy by enhancing demand-side regulating ability. This paper proposes a peak shaving potential assessment model based on the price elasticity mechanism and consumer psychology, focusing on the adjustable user load in virtual power plants. The values of deterministic parameters and the distribution of the uncertain parameter of the model are obtained through the long short-term memory network (LSTM) and mixture density network (MDN). Then, the refined distribution of peak shaving potential considering external conditions, incentive inputs, and spatial and temporal scales is obtained. Based on the evaluation results, a peak shaving decision-making model for virtual power plants is constructed using a scenario scheme. Differentiated schemes for traditional, risk-averse, and risk-seeking virtual power plant decision-makers are considered. Case studies using the data of a virtual power plant pilot area show that the proposed model can better characterize the features of virtual power plant users, and a refined control strategy with better economic benefits can be obtained.