Firstly, a probability model is established for the randomness of distributed power sources and loads, and then a voltage stability index of the distribution network is established. Secondly, The probability model, and the voltage stability index is applied to the voltage optimization model of the distribution network, and the probability flow of the distribution network is calculated based on the unscented transformation method. The probabilistic multi-objective voltage optimization model of the distribution network with the objective function, and the improved particle swarm algorithm is used to solve the model. Finally, t The calculation example shows that the proposed method can reduce the network loss of distribution network by configuring shunt capacitors, static var compensators, voltage and reactive power control equipment of on-load tap changer improve the voltage stability, over-limit and fluctuation problems of the distribution network including DG, and improve the ability of the distribution network to cope with source and load uncertainty.