Motor rotor fault diagnosis method based on CNN-BiLSTM-residual module-attention mechanism

The invention discloses a motor rotor fault diagnosis method based on a CNN-BiLSTM-residual module-attention mechanism, and belongs to the technical field of fault diagnosis, and the method comprises the following steps: S1, employing an improved Chernobelian disaster algorithm ICDO to optimize VMD parameters; s2, after the initial fault data of the motor rotor is decomposed through VMD, IMF components with fault feature signal components are screened out according to a correlation coefficient criterion and kurtosis, and then signal reconstruction is carried out; s3, solving a time domain feature and a frequency domain feature of the reconstructed signal, and screening data features by using a comprehensive analysis method; and S4, inputting the screened data features into the CNN-BiLSTM-residual module-attention mechanism network to carry out fault diagnosis. According to the motor rotor fault diagnosis method based on the CNN-BiLSTM-residual error module-attention mechanism, motor rotor fault diagnosis can