融合加速退化和现场实测退化数据的剩余寿命预测方法
收稿日期: 2013-12-09
修回日期: 2014-03-03
网络出版日期: 2014-03-10
基金资助
国家自然科学基金(61273058)
Residual Life Prediction Method Fusing Accelerated Degradation and Field Degradation Data
Received date: 2013-12-09
Revised date: 2014-03-03
Online published: 2014-03-10
Supported by
National Natural Science Foundation of China (61273058)
针对先验信息为加速退化数据的情况,提出了利用非共轭先验分布进行Bayesian统计推断的剩余寿命预测方法.不预先假定Wiener过程参数值的分布类型,利用加速系数将加速应力下的参数值折算到工作应力水平下,进而使用Anderson-Darling方法确定参数值的最优拟合分布类型.在对参数值进行折算时,根据周源泉提出的理论对Wiener过程参数与加速应力之间的关系进行了推导.参数估计时,通过极大似然法得到超参数的估计值,利用WinBUGS软件实现Markov Chain Monte Carlo仿真得到参数的后验均值.通过某型军用电连接器寿命预测实例验证了所提方法的实用价值和研究意义,结果表明本方法可有效解决先验信息为加速退化数据时进行剩余寿命预测的难题.
王浩伟 , 徐廷学 , 赵建忠 . 融合加速退化和现场实测退化数据的剩余寿命预测方法[J]. 航空学报, 2014 , 35(12) : 3350 -3357 . DOI: 10.7527/S1000-6893.2014.0010
To address the situation that prior information is accelerated degradation data, a residual life prediction method based on Bayesian inference with non-conjugate prior distribution is proposed. Without presupposing the parameters' distribution types of a Wiener process, the parameters at accelerated stress levels are transformed into normal use stress level through acceleration factors and then the optimally fitted distribution types are determined using Anderson-Darling method. When transforming the parameter values, the relationships between parameters of the Wiener process and accelerated stress are deduced according to the theory proposed by Zhou Yuanquan. The estimates of hyper-parameters can be obtained by maximum likelihood methods, and posterior means of parameters can be inferred by Markov Chain Monte Carlo using WinBUGS software. The practical value and research significance of this study are demonstrated through a certain missile electrical connector lifetime prediction example. The results show that the proposed method can effectively solve the problem of residual life prediction when prior information is accelerated degradation data.
[1] Guida M, Penta F. A Bayesian analysis of fatigue data[J]. Structural Satety, 2010, 32(1): 64-76.
[2] Jiang X M, Yuan Y, Liu X. Bayesian inference method for stochastic damage accumulation modeling[J]. Reliability Engineering and System Safety, 2013, 111(2): 126-138.
[3] Karandikar J M, Kim N H, Schmitz T L. Prediction of remaining useful life for fatigue-damaged structures using Bayesian inference[J]. Engineering Fracture Mechanics, 2012, 96(4): 588-605.
[4] Berger J O. Statistical decision theory and Bayesian analysis[M]. New York: Springer-Verlag, 1985: 111-198.
[5] Wang X. Wiener processes with random effects for degradation data[J]. Journal of Multivariate Analysis, 2010, 101(2): 340-351.
[6] Peng B H, Zhou J L, Pan Z Q. Bayesian method for reliability assessment of products with Wiener process degradation[J]. System Engineering-Theory & Practice, 2010, 30(3): 543-549. (in Chinese) 彭宝华, 周经伦, 潘正强. Wiener过程性能退化产品可靠性评估的Bayes方法[J]. 系统工程理论与实践, 2010, 30(3): 543-549.
[7] Wang X L, Cheng Z J, Guo B. Residual life forecasting of metalized film capacitor based on wiener process[J]. Journal of National University of Defense Technology, 2011, 33(4): 146-151. (in Chinese) 王小林, 程志君, 郭波. 基于维纳过程金属化膜电容器的剩余寿命预测[J]. 国防科技大学学报, 2011, 33(4): 146-151.
[8] Jin G, Matthews D E, Zhou Z. A Bayesian framework for on-line degradation assessment and residual life prediction of secondary batteries in spacecraft[J]. Reliability Engineering and System Safety, 2013, 113(4): 7-20.
[9] Ahmad K E, Jaheeen Z F. Approximate bayesian estimators[J]. Computers and Mathematics with Applications, 1994, 29(12): 39-47.
[10] Whitmore G A. Normal-gamma mixtures of inverse Gaussian distribution[J]. Scandinavian Journal of Statistics, 1986, 13(1): 211-220.
[11] Onar A, Padgett W J. Accelerated test models with the inverse Gaussian distribution[J]. Journal of Statistical Planning and Inference, 2000, 89(1): 119-133.
[12] Seshadri V. The inverse Gaussian distribution[M]. New York: Oxford University Press, 1993: 67-99.
[13] Ntzoufras I. Bayesian modeling using WinBUGS[M]. New York: John Wiley & Sons, 2009: 12-55.
[14] Zhou Y Q, Weng Z X, Ye X T. Study on accelerated factor and condition for constant failure mechanism[J]. Systems Engineering and Electronics, 1996, 18(1): 55-67. (in Chinese) 周源泉, 翁朝曦, 叶喜涛. 论加速系数与失效机理不变的条件(Ⅰ)—寿命型随机变量的情况[J]. 系统工程与电子技术, 1996, 18(1): 55-67.
[15] Zhang C H, Chen X, Yang Y M. Research on environmental factors under common life distributions[J]. Structure & Environment Engineering, 2001, 4(7): 8-11. (in Chinese) 张春华, 陈循, 杨拥民. 常见寿命分布下环境因子的研究[J]. 强度与环境, 2001, 4(7): 8-11.
[16] Yang Y H, Zhou Y Q. Theoretical foundation of accelerated life(Ⅰ)[J]. Journal of Propulsion Technology, 2001, 22(4): 276-278. (in Chinese) 杨宇航, 周源泉. 加速寿命试验的理论基础(Ⅰ)[J]. 推进技术, 2001, 22(4): 276-278.
[17] Grace A W, Wood I A. Approximating the tail of the Anderson-Darling distribution[J]. Computational Statistics and Data Analysis, 2012, 56(12): 4301-4311.
[18] Lin R J, Chen W H, Liu J, et al. Research on feasibility of accelerated degradation test for aerospace electrical connector[J]. Journal of Engineering Design, 2010, 17(4): 317-320. (in Chinese) 林瑞进, 陈文华, 刘娟, 等. 航天电连接器加速性能退化试验可行性研究[J]. 工程设计学报, 2010, 17(4): 317-320.
[19] Zhang G L, Cai J Y, Liang Y Y, et al. Research on reliability assessment method of electronic equipment based on multi-stress ADT[J]. Acta Aeronautica et Astronautica Sinica, 2013, 34(12): 2815-2822. (in Chinese) 张国龙, 蔡金燕, 梁玉英, 等. 电子装备多应力加速退化试验技术及可靠性评估方法研究[J]. 航空学报, 2013, 34(12): 2815-2822.
/
| 〈 |
|
〉 |
CJA