Demand Response Potential Prediction Method for Residential User Clusters with High Penetration Behind-the-Meter Distributed Photovoltaics

With the continuous development of power systems, the demand for flexibility of power systems has surged. Demand response has become a key measure to address this shortage, with the evaluation of demand response potential being fundamental. This paper proposes a method for predicting the demand response potential of residential user clusters in scenarios with high penetration behind-the-meter distributed photovoltaics. Firstly, the demand response behavior of these users is modeled, with an optimization model established to minimize the weighted average of electricity cost and comfort, generating load data under an incentive-based demand response plan. Secondly, a prediction method based on network decoupling is proposed to estimate the demand response potential of user clusters, predicting the potential for future demand response days based on the type of day. Finally, case studies demonstrate that the proposed method has a certain superiority in prediction accuracy.