Typical scenarios generating is one of methods for dealing with uncertainty of wind and PV outputs in power system planning and operation.However, the differences among output distribution functions at different times are not considered by the existing generation methods for typical scenarios.Given this background, in view of the uncertainty and correlation of outputs for distributed renewable energy generation, the differences among outputs′ distribution functions at different times are considered innovatively, the Copula function is used to establish the joint output model for multi-wind farm time-series.A large number of initial scenario sets are generated by probability sampling and splicing from the Copula model, and K-means clustering algorithm is used for reducing scenarios and generating typical joint output scenarios.Case studies show that the typical joint output scenarios of wind power are consistent with the correlation among wind power output and the differences among outputs′ distribution functions at different times.The output scenarios of multiple wind farms in the same area can be generated by the proposed algorithm with higher accuracy, which can guide the optimal operation of power systems more effectively.