Improving the self-healing ability of distribution network under fault conditions is of great significance to ensure the stable operation of power grid, the reliability of power supply to users, and the intelligence of power grid. For distribution network with distributed new energy, this paper studies the self-healing control problem of distribution network considering the randomness of distributed new energy generation. This paper discusses the operation structure and model of distribution network with distributed photovoltaic new energy, constructs a mixed-integer stochastic programming model for distribution network self-healing control considering the uncertainty of distributed photovoltaic power generation, and proposes the scenario tree method and linearization method to transform the original self-healing control problem into a mixed-integer linear programming for solution. The effectiveness of the self-healing control method considering the randomness of distributed power generation and load described in this paper is verified by experimental analysis.