In order to ensure the safe and efficient operation of the transformer in the power system, this paper proposes a method to fault the transformer by combining the RBF neural network with the fuzzy control algorithm.The neural network learning system with 6 layers is designed, and the fuzzy membership function is introduced into the second layer, which accelerates the learning speed of the neural network. Data Based on Transformer Faults This paper analyzes the types of faults by analyzing the internal gas content. The fuzzy RBF neural network designed by this paper is used to diagnose fault diagnosis. The diagnostic results show that the diagnostic model has better effect on the fault diagnosis of the transformer.