Dual adversarial and contrastive network for single-source domain generalization in fault diagnosis
This addresses fault diagnosis challenges in process industries where data is scarce and from a single mode, though it appears incremental by combining adversarial and contrastive learning for domain generalization.
The paper tackles the problem of fault diagnosis in industrial systems with limited single-mode fault data by proposing a dual adversarial and contrastive network (DACN) to extract domain-invariant features for unseen modes, achieving high classification accuracy on benchmarks like the Tennessee Eastman process and continuous stirred-tank reactor while maintaining a small model size.
Domain generalization achieves fault diagnosis on unseen modes. In process industrial systems, fault samples are limited, and it is quite common that the available fault data are from a single mode. Extracting domain-invariant features from single-mode data for unseen mode fault diagnosis poses challenges. Existing methods utilize a generator module to simulate samples of unseen modes. However, multi-mode samples contain complex spatiotemporal information, which brings significant difficulties to accurate sample generation. To solve this problem, this paper proposed a dual adversarial and contrastive network (DACN) for single-source domain generalization in fault diagnosis. The main idea of DACN is to generate diverse sample features and extract domain-invariant feature representations. An adversarial pseudo-sample feature generation strategy is developed to create fake unseen mode sample features with sufficient semantic information and diversity, leveraging adversarial learning between the feature transformer and domain-invariant feature extractor. An enhanced domain-invariant feature extraction strategy is designed to capture common feature representations across multi-modes, utilizing contrastive learning and adversarial learning between the domain-invariant feature extractor and the discriminator. Experiments on the Tennessee Eastman process and continuous stirred-tank reactor demonstrate that DACN achieves high classification accuracy on unseen modes while maintaining a small model size.