Bing Long ab, Houjun Wang a, Qiang Miao b, and Michael Pecht c
aSchool of Automation Engineering, University of Electronic Science and Technology of China (UESTC),
Chengdu 611731, China
bSchool of Mechanical, Electronic and Industrial Engineering, UESTC, Chengdu 611731, China
cCenter for Advanced Life Cycle Engineering (CALCE), University of Maryland, College Park, MD 20742, USA
Abstract:
It is important to find the potential failure of subsystems or components for complex electronic systems such as radar, aviation systems, and then repair or replace them before the whole system fails. To repair or replace potential failure subsystems or circuits before a whole system fails, a new online prognostics and health management (PHM) system has been developed based on Mahalanobis distance (MD), the multisignal model, and the least squares support vector machine (LS-SVM). MD is used as a health index for monitoring each test parameter in the system. If health indices indicate that the test parameters will fail soon, the multisignal model is triggered to locate the potential failure subsystems at the system level. Then, LS-SVM is triggered to identify the failure modes or components in the potential subsystems. This paper presents an experiment with a radar transmitter system, which shows that the new PHM strategy can find potential failure circuits or components before the whole system fails. This method is useful for condition-based maintenance (CBM) of complex electronic systems.
Complete Article is available from the publisher and to CALCE consortium members.
© IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.