The application of transfer learning in fault diagnosis has been developed in recent years. It can use existing data to solve the problem of fault recognition under different working conditions. Due to the complexity of the equipment and the openness of the working environment in industrial production, the status of the equipment is changeable, and the collected signals can have new fault classes. Therefore, the open set recognition ability of the transfer learning method is an urgent research direction. The existing transfer learning model can have a severe negative transfer problem when solving the open set problem, resulting in the aliasing of samples in the feature space and the inability to separate the unknown classes. To solve this problem, we propose a Weighted Domain Adaptation with Double Classifiers (WDADC) method. Specifically, WDADC designs the weighting module based on Jensen–Shannon divergence, which can evaluate the similarity between each sample in the target domain and each class in the source domain. Based on this similarity, a weighted loss is constructed to promote the positive transfer between shared classes in the two domains to realize the recognition of shared classes and the separation of unknown classes. In addition, the structure of double classifiers in WDADC can mitigate the overfitting of the model by maximizing the discrepancy, which helps extract the domain-invariant and class-separable features of the samples when the discrepancy between the two domains is large. The model’s performance is verified in several fault datasets of rotating machinery. The results show that the method is effective in open set fault diagnosis and superior to the common domain adaptation methods.

Due to the lack of fault signals and the variability of working conditions in engineering practice, there is still a gap between the conventional deep learning fault diagnosis models and the practical application. Aiming at the problem of few-shot fault diagnosis in variable conditions, we propose a novel few-shot transfer learning method based on meta-learning and graph convolutional network for machinery fault diagnosis. The 2D convolution module is used to extract latent features. Then the extracted features and their labels are combined as the nodes, and the similarity between the nodes is used as the connection relationship between the nodes, so as to realize the construction of the graph sample. Subsequently, graph samples are input into the graph convolutional network to evaluate the similarity and complete the classification of faults. Crucially, the idea of metric-based meta-learning is integrated into the graph convolutional network to set tasks and extraction methods. Finally, the analysis and comparison of the diagnostic accuracy under different sample capacity and transfer conditions were demonstrated. The results show that the method can achieve 97.25% diagnostic accuracy with only a few samples in the scene of variable working conditions.

No abstract