To fully explore the regulation potential of various reserve resources, the optimal scheduling problem of an integrated energy system when it participates in both the electrical energy and reserve markets is studied. Considering the relevant characteristics of energy and load in the same area and the risk faced by the system during operation, a stochastic optimization model for the day-ahead scheduling of the integrated energy system is established. First, the Latin hypercube sampling technique based on Cholesky decomposition is used to generate a wind-solar-load sample matrix that conforms to the source-load correlation in the integrated energy system. Secondly, the affinity propagation (AP) clustering algorithm is used to generate typical scenarios. Then, with the objective of minimizing the total cost of scheduling composed of day-ahead scheduling cost, expected cost of adjustment and conditional value of risk, a stochastic optimization scheduling model of the integrated energy system is constructed. Finally, a simulation model of the integrated energy system is established to verify that the proposed method can characterize the source-load correlation in the integrated system and achieve a balance between the economy and safety of system operation.