Health Assessment Method for Electronic Components Subject to Condition Monitoring and Hard Failure

• Published in: IEEE transactions on instrumentation and measurement Vol. 68; no. 1; pp. 138 - 150
• Main Authors: Zhao, Shuai, Makis, Viliam, Chen, Shaowei, Li, Yong
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accelerated tests; Condition monitoring; Condition monitoring (CM); Degradation; Electronic components; Failure; Field effect transistors; Hazards; Health; health assessment; Markov chains; MOSFET; MOSFETs; Parameter estimation; power metal–oxide–semiconductor field-effect transistors (MOSFETs); proportional hazards (PHs) model; Reliability; Reliability aspects; residual life (RL) prediction; Semiconductor device modeling; Semiconductor devices; Statistical models
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2018.2839938

In this paper, a health assessment method for electronic components subject to condition monitoring (CM) is presented, which can be used to estimate the conditional reliability characteristics given the current operational age and the corresponding degradation state. The degradation process is characterized by a continuous-time Markov chain, which is incorporated into a Cox's proportional hazard model to describe the hazard rate of the time to failure. The two main challenges encountered in the health assessment of electronic components subject to CM and hard failure, i.e., the large number of degradation states and the general deteriorating transition mechanism, can be properly addressed by the proposed method. To illustrate the effectiveness of the proposed method, a case study of power metal-oxide-semiconductor field-effect transistors is performed as a representative example of a general electronic application. The method for the model parameter estimation is developed and applied using the degradation and failure data obtained from an accelerated testing experiment, based on which the conditional reliability function, the mean residual life, and other health characteristics can be explicitly calculated using the proposed discretization technique. The excellent accuracy of the health assessment demonstrates the effectiveness and advantages of the proposed method applied to a real electronic component subject to CM and hard failure.