数据融合下的腐蚀油气管道剩余寿命预测
|
摘要
针对管道样本量少、退化数据不足造成寿命预测不准确的问题,提出了基于维纳过程(Wiener Process)的贝叶斯信息融合法以实现腐蚀油气管道剩余寿命的实时预测。首先通过双应力加速退化试验获得退化数据,并结合现场实测数据建立剩余寿命预测模型;然后利用马尔科夫链蒙特卡洛(MCMC)估计未知参数;最后以某型管道为例,验证所提方法的合理性和正确性。结果表明:利用加速退化试验能大幅度缩短数据获取时间,加速退化数据和少量现场实测数据相结合,运用贝叶斯方法能够提高参数估计的精度,从而提高剩余寿命预测的准确性,该方法可应用于管道的可靠性评价中。
In view of the small sample pipeline and insufficient degradation data that could not meet the requirement of accurate prediction of residual lifetime,this paper used a Bayesian information fusion method based on Wiener process to predict the residual lifetime of the corroded pipelines. Firstly,the degenerated data were obtained by the double-stress constant accelerated degradation test,and the predict model of residual lifetime was established based on measured data. Then the Markov Chain Monte Carlo( MCMC) method was used to estimate the unknown parameters. At last,a certain type of service pipeline as an example was used to verify the rationality and correctness of the proposed method. Results showed that using accelerated degradation tests could greatly shorten the data acquisition time. The accuracy of parameter estimation was improved by Bayesian method with the combination of accelerated degradation data and a small amount of field data,and then the residual lifetime could be predicted accurately. Therefore,the method can be applied to the reliability evaluation of pipelines.
In view of the small sample pipeline and insufficient degradation data that could not meet the requirement of accurate prediction of residual lifetime,this paper used a Bayesian information fusion method based on Wiener process to predict the residual lifetime of the corroded pipelines. Firstly,the degenerated data were obtained by the double-stress constant accelerated degradation test,and the predict model of residual lifetime was established based on measured data. Then the Markov Chain Monte Carlo( MCMC) method was used to estimate the unknown parameters. At last,a certain type of service pipeline as an example was used to verify the rationality and correctness of the proposed method. Results showed that using accelerated degradation tests could greatly shorten the data acquisition time. The accuracy of parameter estimation was improved by Bayesian method with the combination of accelerated degradation data and a small amount of field data,and then the residual lifetime could be predicted accurately. Therefore,the method can be applied to the reliability evaluation of pipelines.
引文
[1]王小完,顾建荣,骆正山.基于可靠性的油气管道系统维修策略研究[J].消防科学与技术,2015,34(5):655-659.
[2]张新生,李亚云,王小完.基于逆高斯过程的腐蚀油气管道维修策略[J].石油学报,2017,38(3):356-362.
[3] GASPERIN M,JURICIC D,BOSKOSKI P,et al. The Model-based prognostics of gear health using stochastic dynamical models[J]. Mechanical Systems and Signal Processing,2011,25(2):537-548.
[4] ZHOU W,XIANG W,HONG H P. Sensitivity of system reliability of corroding pipelines to modeling of stochastic growth of corrosion defects[J]. Reliability Engineering&System Safety,2017,167:428-438.
[5] COMPARE M,BARALDI P,BANI I,et al. Development of a Bayesian multi-state degradation model for up-to-date reliability estimations of working industrial components[J].Reliability Engineering&System Safety,2017,166:25-40.
[6]张鹏,彭杨.考虑随机变量相关性的腐蚀管道失效概率[J].石油学报,2016,37(10):1 293-1 301.
[7] WANG W W,LI Y F,YANG Y J,et al. Inverse Gaussian process models for degradation analysis:a Bayesian perspective[J]. Reliability Engineering&System Safety,2014,130:175-189.
[8] WANG X. Wiener processes with random effects for degradation data[J].Journal of Multivariate Analysis,2010,101(2):340-351.
[9]骆正山,毕傲睿.腐蚀管道寿命可靠性的步降应力加速试验研究[J].中国安全科学学报,2017,27(1):59-64.
[10]任淑红,薛飞,余伟炜,等.基于性能退化的热老化可靠性剩余寿命预测方法[J].核动力工程,2013,34(5):96-99.
[11]汪亚顺,莫永强,张春华,等.双应力步进加速退化试验统计分析研究——模型与方法[J].兵工学报,2009,30(4):451-456.
[12]周绍华,胡昌华,司小胜,等.融合非线性加速退化模型与失效率模型的产品寿命预测方法[J].电子学报,2017,45(5):1 084-1 089.
[13]蔡忠义,陈云翔,李韶亮,等.考虑随机退化和信息融合的剩余寿命预测方法[J].上海交通大学学报,2016,50(11):1 778-1 783.
[14]周忠宝,厉海涛,刘学敏,等.航天长寿命产品可靠性建模与评估的Bayes信息融合方法[J].系统工程理论与实践,2012,32(11):169-174.
[15]彭卫文,黄洪钟,李彦锋,等.基于数据融合的加工中心功能铣头贝叶斯可靠性评估[J].机械工程学报,2014,50(6):185-191.
[16]潘尔顺,陈震.高可靠性产品退化建模研究综述[J].工业工程与管理,2015,20(6):1-6.
[17]汪赵新.基于Gamma过程的步进应力加速退化试验设计方法[D].长沙:国防科学技术大学,2011.
[18]司小胜,胡昌华,周东华.带测量误差的非线性退化过程建模与剩余寿命估计[J].自动化学报,2013,39(5):530-541.
[19]孙中泉,赵建印.Gamma过程退化失效可靠性分析[J].海军航空工程学院学报,2010,25(5):581-584.
[20]彭宝华,周经伦,孙权,等.基于退化与寿命数据融合的产品剩余寿命预测[J].系统工程与电子技术,2011,380(5):111-116.
[21]赵志草,宋保维,赵晓哲.加速退化试验与加速寿命试验相结合的产品可靠性评估[J].系统工程理论与实践,2014,34(7):286-290.
[22]王浩伟,滕克难.基于加速退化数据的可靠性评估技术综述[J].系统工程与电子技术,2017,39(12):2 877-2 885.
[23]张详坡,尚建忠,陈循,等.三参数Weibull分布竞争失效场合变应力加速寿命试验统计分析[J].机械工程学报,2014,50(14):42-49.
[24]魏高乐,陈志军.基于多阶段-随机维纳退化过程的产品剩余寿命预测方法[J].科学技术与工程,2015,15(26):27-34.
[25]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):159-166.
[26]姜梅.基于Gamma模型和加速退化数据的可靠性分析方法[J].海军航空工程学院学报,2013,28(4):408-411.
[27]蔡忠义,陈云翔,张诤敏,等.非线性步进加速退化数据的可靠性评估方法[J].北京航空航天大学学报,2016,42(3):576-582.
[28]蔡忠义,陈云翔,项华春,等.融合先验加速退化与外场退化信息的可靠性评估方法[J].系统工程与电子技术,2016,38(4):970-976.
[2]张新生,李亚云,王小完.基于逆高斯过程的腐蚀油气管道维修策略[J].石油学报,2017,38(3):356-362.
[3] GASPERIN M,JURICIC D,BOSKOSKI P,et al. The Model-based prognostics of gear health using stochastic dynamical models[J]. Mechanical Systems and Signal Processing,2011,25(2):537-548.
[4] ZHOU W,XIANG W,HONG H P. Sensitivity of system reliability of corroding pipelines to modeling of stochastic growth of corrosion defects[J]. Reliability Engineering&System Safety,2017,167:428-438.
[5] COMPARE M,BARALDI P,BANI I,et al. Development of a Bayesian multi-state degradation model for up-to-date reliability estimations of working industrial components[J].Reliability Engineering&System Safety,2017,166:25-40.
[6]张鹏,彭杨.考虑随机变量相关性的腐蚀管道失效概率[J].石油学报,2016,37(10):1 293-1 301.
[7] WANG W W,LI Y F,YANG Y J,et al. Inverse Gaussian process models for degradation analysis:a Bayesian perspective[J]. Reliability Engineering&System Safety,2014,130:175-189.
[8] WANG X. Wiener processes with random effects for degradation data[J].Journal of Multivariate Analysis,2010,101(2):340-351.
[9]骆正山,毕傲睿.腐蚀管道寿命可靠性的步降应力加速试验研究[J].中国安全科学学报,2017,27(1):59-64.
[10]任淑红,薛飞,余伟炜,等.基于性能退化的热老化可靠性剩余寿命预测方法[J].核动力工程,2013,34(5):96-99.
[11]汪亚顺,莫永强,张春华,等.双应力步进加速退化试验统计分析研究——模型与方法[J].兵工学报,2009,30(4):451-456.
[12]周绍华,胡昌华,司小胜,等.融合非线性加速退化模型与失效率模型的产品寿命预测方法[J].电子学报,2017,45(5):1 084-1 089.
[13]蔡忠义,陈云翔,李韶亮,等.考虑随机退化和信息融合的剩余寿命预测方法[J].上海交通大学学报,2016,50(11):1 778-1 783.
[14]周忠宝,厉海涛,刘学敏,等.航天长寿命产品可靠性建模与评估的Bayes信息融合方法[J].系统工程理论与实践,2012,32(11):169-174.
[15]彭卫文,黄洪钟,李彦锋,等.基于数据融合的加工中心功能铣头贝叶斯可靠性评估[J].机械工程学报,2014,50(6):185-191.
[16]潘尔顺,陈震.高可靠性产品退化建模研究综述[J].工业工程与管理,2015,20(6):1-6.
[17]汪赵新.基于Gamma过程的步进应力加速退化试验设计方法[D].长沙:国防科学技术大学,2011.
[18]司小胜,胡昌华,周东华.带测量误差的非线性退化过程建模与剩余寿命估计[J].自动化学报,2013,39(5):530-541.
[19]孙中泉,赵建印.Gamma过程退化失效可靠性分析[J].海军航空工程学院学报,2010,25(5):581-584.
[20]彭宝华,周经伦,孙权,等.基于退化与寿命数据融合的产品剩余寿命预测[J].系统工程与电子技术,2011,380(5):111-116.
[21]赵志草,宋保维,赵晓哲.加速退化试验与加速寿命试验相结合的产品可靠性评估[J].系统工程理论与实践,2014,34(7):286-290.
[22]王浩伟,滕克难.基于加速退化数据的可靠性评估技术综述[J].系统工程与电子技术,2017,39(12):2 877-2 885.
[23]张详坡,尚建忠,陈循,等.三参数Weibull分布竞争失效场合变应力加速寿命试验统计分析[J].机械工程学报,2014,50(14):42-49.
[24]魏高乐,陈志军.基于多阶段-随机维纳退化过程的产品剩余寿命预测方法[J].科学技术与工程,2015,15(26):27-34.
[25]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):159-166.
[26]姜梅.基于Gamma模型和加速退化数据的可靠性分析方法[J].海军航空工程学院学报,2013,28(4):408-411.
[27]蔡忠义,陈云翔,张诤敏,等.非线性步进加速退化数据的可靠性评估方法[J].北京航空航天大学学报,2016,42(3):576-582.
[28]蔡忠义,陈云翔,项华春,等.融合先验加速退化与外场退化信息的可靠性评估方法[J].系统工程与电子技术,2016,38(4):970-976.