Fault diagnosis of asynchronous motors based on 2DCNN-GRU network optimization

The current paper proposes a method of asynchronous motor fault diagnosis based on the combination of Two-Dimensional Convolutional Neural Network (2DCNN) and Gated Recurrent Unit (GRU) with optimised network parameters to address the issues of low accuracy of the existing single model based on neural network in motor fault diagnosis. Firstly, the 2DCNN and GRU networks are used to extract the spatial and temporal characteristics of the asynchronous motor fault signals, respectively. Next, the main parameters of the 2DCNN-GRU network model are optimised using the Weighted Mean Of Vectors Algorithm (INFO), and then the SoftMax function is used to fault classification. The proposed method is experimentally evaluated in this study using a three-phase asynchronous motor fault diagnosis, and the findings demonstrate the high accuracy of the suggested model.