Adaptive graph-guided joint soft clustering and distribution alignment for cross-load and cross-device rotating machinery fault transfer diagnosis

Abstract

Domain adaptation (DA) is an effective solution for addressing the domain shift problem. However, existing DA techniques usually directly match the distributions of the data in the original feature space, where some of the features may be distorted by a large domain shift. Besides, geometric and clustering structures of the data, which play a significant role in revealing hidden failure patterns, are not considered in traditional DA methods. To tackle the above issues, a new joint soft clustering and distribution alignment with graph embedding (JSCDA-GE) method is proposed. Specifically, weighted subspace alignment (WSA) is proposed to align bases of source and target subspaces by combining instance reweighting and subspace alignment strategies. Then, JSCDA-GE formulates an objective function by incorporating dynamic distribution alignment (DDA), soft large margin clustering (SLMC), and graph embedding (GE) in a unified structural risk minimization (SRM) framework. Ultimately, JSCDA-GE aims to learn a generalization classifier for fault diagnosis. Its effectiveness and superiority have been confirmed through thirty-six tasks on two bearing databases.

A dual-weighted adversarial network for partial domain fault diagnosis of machinery

“…Additionally, DWAN exhibits the smallest mean standard deviation, confirming its robustness. (2) The overall performance of DANN is stronger than that of IWAN with partial DA tricks. Its overall diagnostic accuracy is significantly higher than that of IWAN by about 8.53%, which can be explained by the fact that the direct application of partial DA tricks that do not match the dataset may lead to class-level mismatches across domains.…”

“…issues. Guaranteeing the security and optimal performance of rotating machinery heavily relies on the implementation of an accurate fault diagnosis [2]. Among various diagnostic methods, data-driven fault diagnosis methods have grown in popularity because of their particular benefits, including high accuracy and reduced reliance on expert experience.…”