A Fault Diagnosis Method for Rolling Bearings Based on Parameter Transfer Learning under Imbalance Data Sets

Fault diagnosis under the condition of data sets or samples with only a few fault labels has become hot spot in field machinery diagnosis. To solve this problem, method based on deep transfer learning is proposed. Firstly, discriminator generative adversarial network (GAN) improved by enhancing its sparsity, and then adopts mechanism to continuously optimize recognition ability discriminator; finally, parameter (PTL) applied trained target domain problem small number label samples. Experimental results show that good performance.