Industrial circulating water system water supply pump fault diagnosis method based on GAN-CNN-BiLSTM

The invention provides a fault diagnosis method for a water supply pump of an industrial circulating water system, and the method comprises the following steps: firstly, collecting a driving end vibration signal which is strongly related to the state of the water supply pump through a three-axis acceleration sensor, carrying out the signal-image conversion of the vibration signal, and achieving the three-channel two-dimensional diagram of a one-dimensional vibration time sequence signal; performing data enhancement on an original sample by adopting a generative adversarial network, and expanding a water supply pump fault image data set; on the basis, a convolution-bidirectional long-short term memory neural network model is adopted to extract fault features; weighting the fault depth features by fusing an attention mechanism method to obtain key depth features; and finally, clustering the key depth features, determining the fault type, and establishing a water supply pump fault diagnosis model. Aiming at the