期刊
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
卷 186, 期 -, 页码 -出版社
ACADEMIC PRESS LTD- ELSEVIER SCIENCE LTD
DOI: 10.1016/j.ymssp.2022.109884
关键词
Discriminative feature learning; Joint domain adaptation; Distribution alignment; Classification loss; Fault transfer diagnosis
An improved joint distribution adaptation (IJDA) mechanism is proposed to enhance the distribution alignment and match the marginal distributions as well as conditional distributions of two domains. It combines maximum mean discrepancy and CORrelation Alignment (CORAL) to enhance domain confusion and constructs an improved conditional distribution alignment mechanism. In addition, a new I-Softmax loss is introduced to contribute to feature learning and learn more separable features. Experimental results on six cross-machine diagnostic tasks demonstrate that the proposed DDTLN achieves higher performance in transfer fault diagnosis compared to other typical domain adaptation methods.
Many domain adaptation methods have been presented to deal with the distribution alignment and knowledge transfer between the target domain and the source domain. However, most of them only pay attention to marginal distribution alignment while neglecting the discriminative feature learning in two domains. Thus, they still cannot satisfy the diagnosis requirement in some cases. To enhance the distribution alignment and match the marginal distributions as well as conditional distributions of two domains, an improved joint distribution adaptation (IJDA) mechanism is proposed. In IJDA, to enhance domain confusion, maximum mean discrepancy and CORrelation Alignment (CORAL) are combined as a new distribution discrepancy metric. Furthermore, an improved conditional distribution alignment mechanism is constructed. To contribute to feature learning and learn more separable features, a new I-Softmax loss that can be optimized like the original Softmax loss and possesses a stronger classification ability is proposed. With the IJDA mechanism and I-Softmax loss, the deep discriminative transfer learning network (DDTLN) is built to implement fault transfer diagnosis. Under the unlabeled target-domain samples, the experimental results on six cross-machine diagnostic tasks verify that the pro-posed DDTLN has a higher performance of transfer fault diagnosis than other typical domain adaptation methods.
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。
推荐
暂无数据