Customer Clustering and Typical Daily Load Profile Identification Considering the Demand Response and the Inherent Uncertainties

Demand response (DR) is a reserve resource that affects the system reliability by guiding consumers to change their electricity consumption patterns actively, thus achieving the reduction of peak load. However, the price elasticity curve is difficult to predict, which provides an uncertain dimension for reliability assessment of generation and transmission systems, so it is urgent to analyze the load profile after considering the DR and the inherent uncertainties. Therefore, this paper first proposes DR uncertainty index HI to describe the uncertainty of consumer’s behavior when the electricity price changes. Then, a new Sigmoid cloud model is proposed to depict the uncertain mapping relationship between electricity price and HI under the influence of complex factors. Based on this, this paper proposes the modified demand elasticity matrix, obtains the functional relationship between the uncertainty variation of DR and load, then, an uncertainty model of price-based DR is established. Finally, in order to reduce the number of loads, k-means algorithm is used to cluster the load considering the DR uncertainty. The correctness and effectiveness of the proposed method is investigated through the case study on the RBTS and RTS reliability test system.