Real-Time Prediction Method for Performance Degradation Trend Based on Reliability Experimental Data

Abstract:

According to the real-time prediction for performance degradation trend, the commonly used method is just based on field data. But this methods prediction result will not be so much ideal when the fitting of degradation trend of field data is not good. To solve the problem, the paper introduces a new method which is not only based on field method but also based on reliability experimental data coming from the history experiment. We use the relationship between the field data and reliability experimental data to get the result of the two kinds of data respectively and then get the weights according to the two prediction results. Finally, the final real-time prediction result for performance degradation tendency can obtain by allocating the weights to the two prediction results.

• Periodical: Applied Mechanics and Materials (Volumes 321-324)
• Pages: 757-761
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.321-324.757
• Online since: June 2013
• Authors: Chen Liang Song, Zhen Liu, Bin Long, Cheng Lin Yang
• Keywords: AR Model, Performance Degradation Trend, Reliability Experimental Data, Weights Allocation

References

[1] Deyan Chen, Prediction of Chinese Urban and Rural Income Gap Based on ARMA Model: Journal of Liaoning Shihua University. Liaoning 2011.

[2] Liming Zhang, Xinwen Zhao, Qi Cai and Guangjiang Wu, Prediction Degradation Trend of Nuclear Equipments Based on GM (1, 1)-Markov Chain: Atomic Energy Science and Technology. Beijing 2010.

[3] Zhen Liu, Houjun Wang, Bing Long and Zhiguo Zhang, Research on Condition Trend Prediction Based on Weighed Hidden Markov and Autoregressive Model: ACTA Electronics SINICA. Beijing 2009.

DOI: 10.1109/icemi.2009.5274896

[4] Haiyou Jiang, Research on the Reliability Estimation Theory Based on Performance Degradation Data: Southeast University Master Degree Thesis. Jiangsu 2009.

[5] Xianda Zhang: Modern Signal Processing (Second Edition) (Tsinghua University Press, pp.69-79, Beijing 2010).

[6] William Q. Meeker, Luis A. Escobar: Statistical Methods for Reliability Data (Wiley-Interscience Publication, p.642, New York 1998).