Rolling bearing fault diagnosis method based on TCN-BiGRU-Attention

In response to the complex working environment of rolling bearings and the problem of low fault identification due to noise interference in vibration signals, a rolling bearing fault diagnosis method based on TCN-BiGRU-Attention is proposed. Firstly, VMD decomposes the vibration signal of the rolling bearing to obtain k IMF components; secondly, by evaluating the index, IMF components with rich fault information are selected, and nine features such as mean, variance, peak value, kurtosis, effective value, peak factor, impulse factor, waveform factor, and margin factor are calculated; finally, the features are input into the hybrid neural network model, where TCN is used to extract the spatial features of time series data, BiGRU is used to capture the temporal features, and the attention mechanism is used to enhance the model's ability to identify key features. The method is verified using the rolling bearing vibration signal of Case Western Reserve University in the United States, and the experimental results show that the method can effectively extract fault features and accurately identify the fault state of rolling bearings, showing a high diagnostic accuracy rate and good application potential.