With the continuous deterioration of the ecological environment, the energy crisis is becoming more and more obvious. In the past decades, the renewable energy has been developed rapidly. However, intermittent power supply, such as wind and photovoltaic power supply, is random and volatile, which will bring some difficulties to the optimal scheduling of power system after it is connected to the grid. In order to solve this problem, an operation multi-objective optimization model of integrated energy system (IES) was proposed to optimize system operation cost, carbon emission and primary energy consumption to the maximum extent. Then, this paper proposes a new multi-objective optimization algorithm, called multiple target cross entropy algorithm based on decomposition (MOCE/D), trans-forming multi-objective operation problems into a series of single objective optimization subproblems, and by considering the sup-ply and demand balance of power, natural gas, cold and thermal energy and the operation limitation of each module to solve the integrated energy system model of non-convexity, nonlinear, and multi-local optimal problems. Finally, the overall performance of the proposed MOCE/D algorithm is comprehensively studied. The results of statistical simulation show that, compared with the traditional CCHP system, the proposed operation multi-objective optimization model of integrated energy system (IES) can effec-tively reduce the operation cost, carbon emission and primary energy consumption. In addition, MOCE/D has better optimiza-tion effect and stronger competitiveness compared with other algorithms.