Spatial graph convolutional neural network via structured subdomain adaptation and domain adversarial learning for bearing fault diagnosis

Unsupervised domain adaptation (UDA) has shown remarkable results in fault diagnosis under changing working conditions in recent years. However, most UDA methods do not consider the geometric structure of the data. Furthermore, the global domain adaptation technique is commonly applied, which ignores the relation between subdomains. This paper addresses mentioned challenges by presenting the novel deep subdomain adaptation graph convolution neural network (DSAGCN), which has two key character-istics: First, a graph convolution neural network (GCNN) is employed to model the structure of data. Second, adversarial domain adaptation and local maximum mean discrepancy (LMMD) methods are applied concurrently to align the subdomain's distribution and reduce structure discrepancy between rel-evant subdomains and global domains. CWRU, PU, and JNU bearing datasets are used to validate the DSAGCN method's superiority between comparison models. The experimental results demonstrate the significance of aligning structured subdomains along with domain adaptation methods to obtain an accu-rate data-driven model.(c) 2022 Elsevier B.V. All rights reserved.

Automated summary

Unsupervised domain adaptation has been successful in fault diagnosis under changing working conditions, but most methods neglect the geometric structure of the data and the relationship between subdomains. This paper proposes a novel deep subdomain adaptation graph convolution neural network that models the data structure and aligns subdomain distributions using adversarial domain adaptation and local maximum mean discrepancy methods to achieve accurate data-driven models.