Fault diagnosis of charging facilities based on improved RNN

Abstract

Abstract To ensure the safe use of electric vehicles in the charging process is an important issue for the electric vehicle industry to overcome. Therefore, this paper proposes an improved RNN neural network fault diagnosis model for charging facilities. Firstly, typical fault types are extracted based on fault analysis of charging facilities. Then, the Whale optimization algorithm (WOA) is used to initialize the RNN network parameters and form WOA-RNN network model. Finally, the improved model is used to analyse the fault, and the comparison of the fault diagnosis accuracy before and after the improvement shows that the proposed algorithm has a high accuracy, and the feasibility of this method is proved.