As a typical network model of artificial neural network, BP neural network has been widely used in fault diagnosis of distribution network. However, due to the randomness of initial weight and initial threshold, the diagnosis accuracy is not high in application. Aiming at this problem, a distribution network fault diagnosis method based on genetic algorithm optimization of BP neural network is proposed. The initial weight and threshold of BP neural network are optimized by genetic algorithm, and the fault diagnosis results are compared with those of traditional neural network in the calculation example. Finally, the simulation errors of the two are analyzed to verify the feasibility. The results show that the genetic algorithm provides relatively ideal initial weights and thresholds for the BP neural network, effectively reducing the error in the operational results and improving the accuracy of the diagnosis.