A bearing fault diagnosis method based on a convolutional spiking neural network with spatial–temporal feature-extraction capability

Abstract

Abstract Convolutional neural networks (CNNs) are widely used in the field of fault diagnosis due to their strong feature-extraction capability. However, in each timestep, CNNs only consider the current input and ignore any cyclicity in time, therefore producing difficulties in mining temporal features from the data. In this work, the third-generation neural network—the spiking neural network (SNN)—is utilized in bearing fault diagnosis. SNNs incorporate temporal concepts and utilize discrete spike sequences in communication, making them more biologically explanatory. Inspired by the classic CNN LeNet-5 framework, a bearing fault diagnosis method based on a convolutional SNN is proposed. In this method, the spiking convolutional network and the spiking classifier network are constructed by using the integrate-and-fire (IF) and leaky-integrate-and-fire (LIF) model, respectively, and end-to-end training is conducted on the overall model using a surrogate gradient method. The signals are adaptively encoded into spikes in the spiking neuron layer. In addition, the network utilizes max-pooling, which is consistent with the spatial-temporal characteristics of SNNs. Combined with the spiking convolutional layers, the network fully extracts the spatial-temporal features from the bearing vibration signals. Experimental validations and comparisons are conducted on bearings. The results show that the proposed method achieves high accuracy and takes fewer time steps.