In order to fully exploit the flexible adjustable potentiality of decentralized electric heating load，an optimization control method for peak regulation of decentralized electric heating considering user satisfaction is proposed．Firstly，the response characteristics of the single electric heating load are analyzed based on the operating characteristics of decentralized electric heating．Secondly，the decentralized electric heating loads are clustered using affinity propagation (AP) algorithm，and a rotating control strategy for the electric heating load clusters is developed based on the equal area rule．Then，a multi-objective optimization model is established that simultaneously considers the participation of electric heating loads in peaking and customer satisfaction．Finally，the proposed control method is validated based on the decentralized electric heating load parameters in a region of Xinjiang．The simulation results show that the proposed method can guarantee the user satisfaction with certain peak regulation capability.