With the large-scale development of wind power, the security and economy problem of wind power integration becomes particularly serious. The traditional chance constrained model does not fully consider the possible losses when the system runs beyond the confidence interval, which may have great risks to the scheduling decision. Based on the theory of Conditional Value-at-Risk, this paper proposes a quantitative method to measure the risk of power deviation of wind power grid connected, and introduces the concept of source network coordinated optimal operation in the scheduling process, and constructs an optimal scheduling model based on the theory of Conditional Value-at-Risk and source network coordinated operation. Multi-objective Compound Backtracking Search (MOCBS) algorithm is used to solve the optimal scheduling model efficiently and accurately. Finally, IEEE30 nodes are taken as an example to verify the model. The results show that the proposed model can effectively measure the potential risks of system operation, and verify the superior performance of MOCBS algorithm.