4.6 Article

Deep Joint Distribution Alignment: A Novel Enhanced-Domain Adaptation Mechanism for Fault Transfer Diagnosis

期刊

IEEE TRANSACTIONS ON CYBERNETICS
卷 53, 期 5, 页码 3128-3138

出版社

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TCYB.2022.3162957

关键词

Maximum likelihood estimation; Task analysis; Fault diagnosis; Circuit faults; Vibrations; Measurement; Deep learning; Conditional distribution discrepancy; domain adaptation (DA); fault transfer diagnosis; joint distribution alignment (JDA); unlabeled target-domain samples

向作者/读者索取更多资源

This study proposes a new domain adaptation mechanism called deep joint distribution alignment (DJDA) to simultaneously reduce the discrepancy in marginal and conditional distributions between source and target domains. By aligning the means and covariances and using a Gaussian mixture model and statistical metric to reduce distribution discrepancy, DJDA can achieve the highest degree of domain confusion. Experimental results demonstrate that DJDA outperforms other typical domain adaptation models in fault transfer diagnosis.
Various domain adaptation (DA) methods have been proposed to address distribution discrepancy and knowledge transfer between the source and target domains. However, many DA models focus on matching the marginal distributions of two domains and cannot satisfy fault-diagnosed-task requirements. To enhance the ability of DA, a new DA mechanism, called deep joint distribution alignment (DJDA), is proposed to simultaneously reduce the discrepancy in marginal and conditional distributions between two domains. A new statistical metric that can align the means and covariances of two domains is designed to match the marginal distributions of the source and target domains. To align the class conditional distributions, a Gaussian mixture model is used to obtain the distribution of each category in the target domain. Then, the conditional distributions of the source domain are computed via maximum-likelihood estimation, and information entropy and Wasserstein distance are employed to reduce class conditional distribution discrepancy between the two domains. With joint distribution alignment, DJDA can achieve domain confusion to the highest degree. DJDA is applied to the fault transfer diagnosis of a wind turbine gearbox and cross-bearing with unlabeled target-domain samples. Experimental results verify that DJDA outperforms other typical DA models.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.6
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据