In order to control air pollution and protect the environment, more and more power energy alternatives covering electric heating and electric vehicles, emerge in distribution networks. Power supply capability of distribution network receives a significant effect and the traditional distribution network operation optimization method needs to be improved. According to the randomness characteristics of electric heating and electric vehicles, the methods of Latin super cube sampling, Cholesky decomposition and synchronous back generation reduction are adopted to realize the rapid generation of probability multiple scenes. Based on the probability multiple scenes, combining robust optimization and stochastic programming method, the scenario analysis method is accepted. With satisfying robust constraints, taking the optimal expectation values of running cost and load balancing degree as objectives, robust network reconfiguration model in distribution network with power energy alternatives integration is built. Ant colony algorithm is used to solve the model. Finally, a case study is given to verify that the proposed method can effectively improve the power supply capacity of the distribution network with large scale power energy alternatives integration.