Abstract
Unsupervised domain adaptation (UDA) has garnered widespread attention in the field of rolling bearings fault diagnosis, thanks to its remarkable performance in transfer-based diagnostic tasks. However, in a noisy environment, the existing domain adaptive networks are difficult to accurately mine domain invariant features and effectively align edge distribution and conditional distribution, which lead to the problem of low diagnostic accuracy in the existing models. In response to the above challenges, a cross-condition fault diagnosis method for bearings based on feature enhancement and semantic information constraints (CC-FESIC) is proposed. Firstly, a networked blind deconvolution (NBD) module is designed based on the time-domain autoencoder and frequency-domain filter, which enhances the expression of domaininvariant features in the fault signal domain of rolling bearings. Secondly, the fusion of domain-invariant features is accomplished through a multi-scale feature extractor. Again, a joint distribution framework with semantic information constraints (JDF-SIC) is constructed, the framework realizes the joint distribution alignment of features respectively through multi-core maximum mean discrepancy (MK-MMD) and the improved MK-MMD (SMK-MMD). In addition, the joint distribution alignment is constrained through a semantic information constraints (SIC) module. Finally, extensive crosscondition diagnostic experiments are carried out on two public bearing datasets to verify the effectiveness of the proposed method.
Funding Information
  • Priority Academic Program Development of Jiangsu Higher Education Institutions
  • Fundamental Research Funds for the Central Universities (2021YCPY0203)
  • Scientific Research Innovation Capability Support Project for Young Faculty (ZYGXQNJSKYCXNLZCXM-E2P)

This publication has 48 references indexed in Scilit: