Motor bearing fault diagnosis method based on reconstructed generative adversarial network

The invention discloses a motor bearing fault diagnosis method and system based on a reconstructed generative adversarial network, and the method comprises the steps: collecting majority class health samples and minority class fault samples under an unbalanced data set, and marking the majority class health samples and minority class fault samples as real samples; a generator network and a discriminator network are constructed, the generator network comprises an input layer, four convolution layers and an output layer, and the discriminator network comprises an input layer, two convolution layers and a full connection layer; a reconstruction layer network is constructed, and the reconstruction layer network is composed of two sub-network encoders and decoders; training a generator network based on the random noise signal, and marking the output signal as a generation sample; jointly training a discriminator network based on the real sample and the generated sample; training a reconstructor based on the genera