The power industry is an important part of my country's strategic goal of carbon neutrality. To promote this goal, it is necessary to mobilize power users to participate in power side demand response. On the one hand, from the user’s perspective, different users face electricity price fluctuations to control expected cost of electricity and electricity cost fluctuations have different preferences. On the other hand, from the energy supply side, as users are affected by the characteristics of production, e-commerce retailers have different difficulties in adjusting electricity consumption behaviors of different users. Therefore, this paper uses the demand response model to characterize user behavior, and simulates the electricity price risk link based on the MCMC sampling method. On the basis of considering the benefits of carbon emission reduction, with users as the main body, this paper comprehensively constructs a multi-objective optimization model that considers total electricity bill control and risk control, and uses adaptive genetic algorithm to solve the case. The results show that the model proposed in this paper can identify the risk preferences and production characteristics of different users, help users maximize the utility of their preferences, and improve their enthusiasm for participating in demand response.