ABSTRACTS
For the timely identification of the potential faults of a rolling bearing and to observe its health condition intuitively and accurately, a novel fault diagnosis and health assessment model for a rolling bearing based on the ensemble empirical mode decomposition (EEMD) method and the adjustment Mahalanobis–Taguchi system (AMTS) method is proposed. The specific steps are as follows: First, the vibration signal of a rolling bearing is decomposed by EEMD, and the extracted features are used as the input vectors of AMTS. Then, the AMTS method, which is designed to overcome the shortcomings of the traditional Mahalanobis–Taguchi system and to extract the key features, is proposed for fault diagnosis. Finally, a type of HI concept is proposed according to the results of the fault diagnosis to accomplish the health assessment of a bearing in its life cycle. To validate the superiority of the developed method proposed approach, it is compared with other recent method and proposed methodology is successfully validated on a vibration data-set acquired from seeded defects and from an accelerated life test. The results show that this method represents the actual situation well and is able to accurately and effectively identify the fault type.
Acknowledgments
The authors would like to thank the associate editor and the anonymous reviewers for their detailed comments and suggestions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Additional information
Funding
Notes on contributors
![](/cms/asset/a15d8d5c-84f4-4f33-87e4-02ab3334d748/tsys_a_1397804_uf0001_oc.jpg)
Junxun Chen
![](/cms/asset/aa8e0d62-7f63-4289-a645-29daf3fcee7b/tsys_a_1397804_uf0002_oc.jpg)
Longsheng Cheng
Longsheng Cheng received his BS degree from Anhui Normal University, China, in 1985, MS degree from East China Institute of Technology, China, in 1988, and his PhD degree in System Engineering from Nanjing University of Science and Technology, China, in 1998. From 1998 to 1999, he joined City University of Hong Kong as a Research Assistant. Since 2005, he has been a Professor of Management Sciences and Applied Statistics at School of Economics and Management, Nanjing University of Science and Technology. His current research interests include prognostic and health monitoring, machine learning, quality engineering.
![](/cms/asset/475fee77-12dd-4f33-9cb5-6854d8ad0e8a/tsys_a_1397804_uf0003_oc.jpg)
Hui Yu
![](/cms/asset/eba3b250-676d-4eac-b11c-544cfa102a8d/tsys_a_1397804_uf0004_oc.jpg)
Shaolin Hu
Shaolin Hu received his B.S degree in mathematics from Anhui Normal University, and his M.S degree in statistics and her Ph D degree in system engineering from the Xi'an Jiaotong University, China. In 2002, he joined the School of Information, University of Science and Technology of China, where he was a post-doctoral researcher until 2004. In 2006, he joined the Royal Swedish Institute of Technology, where he was a visiting professor until 2007. He is currently a distinguished professor with the School of Automation, Foshan University, and supervisor of doctorate candidate at the School of Automation and Information Engineering, Xi'an University of Technology, China. His current research interests include system safety, data fusion, fault diagnosis and big data analysis.