A Comparison of Two Accelerated Degradation Models with Temperature and Humidity as Accelerating Stresses

Abstract:

In this paper, a comparison of two kinds of acceleration models under temperature and humidity joint stress is given. The traditional generalized Arrhenius model and Eyring Model give two ways to describe the accelerated life of products. First, through some mathematical transformation and synthesizing of these two models we have concluded two acceleration models under temperature and humidity joint stress which are described in detail in this paper. Estimations of model parameters are also given. Secondly, by comparing accelerated life coefficient and coefficient of variations, we can take a glimpse of the models in their fitness of actual conditions. Then, since the models presented in this paper follow the features of nested models, a likelihood ratio test is conducted which takes a further step in the work of model selection. At last, these two models are compared through an application example – Smart Electricity Meter.

• Periodical: Applied Mechanics and Materials (Volumes 300-301)
• Pages: 1162-1170
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.300-301.1162
• Online since: February 2013
• Authors: Wen Yu Wang, Xiao Bing Ma, Shi Hua Chang, Rui Kang
• Keywords: Accelerated Degradation Model, Accelerated Life Coefficient, Coefficient of Variations, Model Selection, Nested Models, Smart Electricity Meter

References

[1] D.Q. Zheng, Z.P. Zhang, H.B. Li, and Y.P. Hu, Statistical analysis of constant-stress accelerated life testing under Weibull distribution based on grouped data, Structure & Environment Engineering, vol. 35, no. 6, pp.40-44, (2008).

[2] Z. Hui, C. Shihua and G. Lin, L. Ying. Reliability and Life Assessment of Smart Meter based on Pseudo Life of Multiple Parameters [C]. 2012 International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering: IEEE, 2012: 169 – 172.

DOI: 10.1109/icqr2mse.2012.6246212

[3] K.F. Widaman and J.S. Thompson. On specifying the null model for incremental fit indices in structural equation modeling [J]. Psychological Methods，2003，8(1): 16 – 37.

DOI: 10.1037/1082-989x.8.1.16

[4] X.Y. Li and T.M. Jiang, Review of multiple-stress models in accelerated life testing, Systems Engineering and Electronics, vol. 29, no. 5, pp.828-831, (2007).

[5] M. Adamantions. Modeling & Analysis for Multiple Stress-Type Accelerated Life Data [A]. Proceedings Annual Reliability and Maintainability Symposium [C]: IEEE, 2000: 138 – 143.

[6] W.Q. Meeker and M. Hamada. Statistical tools for rapid development & evaluation of high-reliability products [J]. IEE Transactions on Reliability, 1995, 44(2): 187-198.

DOI: 10.1109/24.387370

[7] V. Crk. Reliability Assessment from Degradation Data [J]. Proceedings Annual Reliability and Maintainability Symposium, (2000).

DOI: 10.1109/rams.2000.816300

[8] C. Zhi, J.L. Juin and Y. Yun. A new extapolation method for long-term degradation prediction of deep - submicron MOSFETs [J]. IEEE Transactions on Electron Devices, 2003, 50(5): 1398-1401.

DOI: 10.1109/ted.2003.813473

[9] L. Gallais, J.Y. Naotoli and C. Amra. Statistical Study of Single and Multiple Pulse Laser-induced Damage in Glasses [J]. Optics Express, 2002, 10(25): 204-219.

DOI: 10.1364/oe.10.001465

[10] H. Liu, Y. Xu and L. Wang. The Principal and Methods of Model Selection in Application of SEM [J]. PSYCHOLOGICAL EXPLORATION, 2007(1), 27: 75-78.