Domain Adversarial Transfer Learning for Fault Diagnosis of Rotating Parts in Cross Working Condition Injection Molding Machines

In response to the problem of decreased accuracy of equipment fault diagnosis models caused by differences in the distribution of degradation data of injection molding machine rotating parts under various operating conditions, transfer learning can effectively minimize cross-domain data distribution differences by removing the inherent relationship between the data distributions of the source and target domains, thereby improving the diagnostic accuracy of the model under different operating conditions. Considering this, this article combines domain adaptation and adversarial training to propose a cross-working condition fault diagnosis domain adversarial transfer learning method. By introducing adversarial training, the relationship between sample data under various working conditions is established, and the difference in sample data distribution is minimized to achieve fault diagnosis under different working conditions.