Deep learning based fault diagnosis method for S700K switch machines

Aiming at the problem that it is difficult to extract the fault features of the power curve of the 5700 K switch machine and it is difficult to jointly optimize the signal processing and classification algorithms, a fault diagnosis method for the switch machine combining convolutional neural networks (CNN) and Transformer is proposed. Firstly, a convolutional neural network is used to carry out preliminary feature extraction on the pre-processed power curves extracted from the signal centralized monitoring system; then, the extracted features are subjected to absolute position encoding and relative position encoding, and the multi-attention mechanism is used to capture the long-distance dependency; after further feature processing by feed-forward neural network, the classification results are outputted through the fully-connected layer.