In the context of the rapid development of energy Internet, integrated energy system has gradually become an important way to realize multi-energy low-carbon and efficient operation. In order to realize the rational planning of the integrated energy system configuration, a multi-objective integrated energy system planning model considering the uncertainty of landscape was proposed. The typical output curve of landscape was obtained by the method of typical curve mining combined with the probability of sequence distribution. Aiming at the lowest system operating cost and power supply rate, an improved multi-objective particle swarm optimization algorithm was adopted to solve the problem. The generated Pareto solution set was screened by hierarchical composition class mixed strategy game, so as to meet the different preferences of decision makers. The sensitivity analysis of the key parameters affecting the system planning scheme is carried out. Based on the actual data of a comprehensive energy demonstration project in China, the planning scheme is presented, and the simulation results verify the effectiveness of the proposed method.