Weibull-Gompertz Regression Model used in Accelerated Life Testing for Competing Failure Modes Consideration

摘要：

In this study the failure behavior of Accelerated Life Testing (ALT) data is examined through Weibull distribution. Instead of fixed shape parameter, which is usually taken in common approach, it assumes that the shape parameter follows Gompertz typed relation with the stress, and the scale parameter an Arrhenius relation for material degradation due to the effect of chemical reaction. Since the characteristics of damage about specimen change as the stress level raising, i.e. different shape parameter should be taken into consideration. Such adjustment allows an additional degree of freedom to fit with the physical situation. Another factor also yields the problem much more complicated due to multi-failure modes occurred in the testing. Thus the maximum likelihood method is used in the analysis. Two examples are studied to demonstrate the proposed approach. The results lead to more mathematical sense as comparing with the traditional approach does.

关键词： accelerated life testing gompertz growth curve competing failure modes shape parameter

作者: YANG Yiheng WANG Kuoshong

作者单位: Department of Mechanical Engineering, National Central University,Taoyuan 32001, Taiwan

会议类型: 国际会议

会议名称: 2010 International Conference on Risk and Reliability Management(2010年风险与可靠性管理国际会议RRM 2010)

会议地点: 北京

会议语种:英文

页码: 241-249

在线出版日期: 2010-10-23（万方平台首次上网日期，不代表论文的发表时间）