Fault diagnosis method of distribution transformer based on improved deep learning

In view of the low efficiency of the current distribution transformer fault diagnosis methods, a transformer state identification method based on improved deep belief network(DBN) is proposed in this paper. Firstly, the operation state data of distribution transformer is classified, analyzed and standardized. On this basis, the bidirectional random butterfly optimization algorithm is used to dynamically optimize the parameters of the DBN, so as to provide the efficient calculation and processing state of the whole cycle of diagnosis and analysis, and realize the accurate and effective fault identification and diagnosis of transformer. The simulation results show that the accuracy of the proposed fault identification method is 98.76% and the analysis time is 7.456s, which has good network performance.