Fault Diagnosis of Electro-Hydraulic Switch Machine Based on IGWO-LSTM

In this paper, a fault diagnosis method of electro-hydraulic switch machine based on improved grey wolf optimization algorithm (IGWO) and long short-term memory network (LSTM) is proposed for the oil pressure curves corresponding to the eight common fault modes and normal modes of ZYJ7 electro-hydraulic switch machine. Firstly, the improved fully integrated empirical mode decomposition and adaptive noise (ICEEMDAN) are used to decompose the time series of nine oil pressure curves on nine oil pressure curves to obtain the intrinsic mode function (IMF), and the fuzzy entropy information is extracted from the IMF as the data set of the diagnostic algorithm. Secondly, the IGWO algorithm is used to optimize the parameters of the LSTM deep learning model. Finally, the data set is input into the trained IGWO-LSTM model for fault diagnosis, and the diagnosis results are obtained. The simulation results show that compared with multi-layer perceptron (MLP) and convolutional neural network (CNN), IGWO-LSTM achieves 94.44% accuracy on the test set and improves the recognition rate of fault types.