The community integrated energy system（CIES）can significantly improve energy utilization efficiency and facilitate the consumption of renewable energy through multi-energy complementary and collaborative optimization scheduling.It has become a new energy utilization approach to balance multi-energy supply and demand on user side.Taking a community in Xiong’an New District as the research object，a two-stage optimization approach for the CIES considering off-design conditions is proposed.At the first stage，non-dominated sorting genetic algorithm（NSGA-II）with elite preservation strategy is taken to optimize the equipment type and capacity of the community energy station.It is a multi-objective planning optimization problem，aiming at coordinately scheduling the economic costs and environmental costs.At the second stage，operation optimization is made.Considering the uncertainty of wind and PV outputs and taking the system economy，environmental cost and users’ comfort as optimization objects，the multiple Pareto fronts obtained at the first stage are solved by Benders decomposition optimization method and the optimal outputs of different generators in various plans are obtained.The comprehensive costs of different plans are the important indicator for determining the best planning scheme.Case studies show that the designed planning scheme takes the operation uncertainty into consideration at the second stage，and realizes the refined operation simulation by fully considering the characteristics of equipment under variable working conditions.The approach can effectively reduce the system operating cost and guarantee the reliability of power supply，which is practical for instructing the CIES planning.
