Abstract: In order to reduce the impact of prediction error on unit output and fully mobilize the enthusiasm of users to participate in demand response, a multi-time scale optimization model of integrated energy system considering the contribution of user demand response is proposed. Firstly, an improved complete ensemble empirical mode decomposition with adaptive noise-circle-grey wolf optimizer-back model based on the improved/fully adaptive acoustic ensemble empirical mode decomposition-circle algorithm is established propagation-autoregressive integrated moving average model(ICEEMDAN-CGWO-BP-ARIMA) to generate multi-time scale scenarios to address the uncertainty of load and wind output. Secondly, the concept of demand response contribution is introduced, and a multi-time scale response strategy is formulated, in which the operating cost of the system and the minimum penalty for curtailment of wind and solar power are taken as the scheduling goals in the pre-day stage. In the intra-day phase, the day-ahead scheduling plan is revised with a time scale of 15 minutes. In the real-time phase, the intraday scheduling plan is revised with a time scale of 5 minutes. Finally, Yamlip is used to build a model for the proposed example, and the Gurobi optimizer is used to solve the model, which analyzes and verifies the positive effect of the proposed model on reducing the cost and improving the operation stability of the system.