BP neural network is often used in fault diagnosis of spring operating mechanism of high voltage circuit breaker because of its strong linear mapping ability and self-adaptive ability. However, it is easy to fall into local minimum, which limits the convergence speed and classification accuracy of the network. This paper presents a fault diagnosis method of high voltage circuit breaker operating mechanism based on L-M algorithm to optimize BP neural network, analyzes the mathematical model and mapping relationship of neural network, optimizes the traditional BP network by using L-M algorithm, solves the problems of local minimization and flat area in traditional BP neural network gradient descent method, and effectively improves the training speed of the algorithm, as well as the accuracy of classification. The diagnosis results show that the BP neural network optimized by L-M algorithm can effectively realize the fault diagnosis of high voltage circuit breaker operating mechanism. The research content of this paper provides ideas and methods for fault diagnosis of high voltage circuit breaker operating mechanism, which is of great significance to improve the safety and reliability of high voltage circuit breaker.