摘要
针对设备退化过程中异常数据下的剩余有效寿命预测问题,提出了一种基于动态的期望最大化算法(EM)-分段隐半马尔可夫模型(SHSMM)预测方法。基于SHSMM的理论框架,采用期望最大化参数自适应估计算法估计模型中的未知参数;基于WGM(1,1)模型,提出动态前向后向灰色填充算法处理样本中的异常数据,并利用健康预测过程预测设备的剩余有效寿命;通过实例分析对模型进行评价和验证。结果表明,提出的设备健康预测方法能有效解决异常数据的问题。
Aiming at the problem of remaining useful life prognosis under abnormal data during equipment degradation,this paper developed a prognostic method based on dynamic expectation maximization(EM)-segmented hidden semi-Markov model(SHSMM). Firstly,based on the SHSMM model framework,it used the expectation maximization algorithm to estimate the unknown parameters of the model. Secondly,to process the anomaly data in the samples,it proposed a dynamic forward-backward gray-fill algorithm based on WGM(1,1),and it carried out the equipment health prognosis. Finally,it used a case study to evaluate the performance of the model. The results show that the proposed method could effectively solve the problem of abnormal data.
引文
[1]胡昌华,施权,司小胜,等.数据驱动的寿命预测和健康管理技术研究进展[J].信息与控制,2017,46(1):72-82.(Hu Changhua,Shi Quan,Si Xiaosheng,et al. Data-driven life prediction and health management:state of the art[J]. Information and Control,2017,46(1):72-82.)
[2] Peng C Y,Tseng S T. Mis-specification analysis of linear degradation models[J]. IEEE Trans on Reliability,2009,58(3):444-455.
[3] Carr M J,Wangbc W. An approximate algorithm for prognostic modelling using condition monitoring information[J]. European Journal of Operational Research,2011,211(1):90-96.
[4] Kharoufeh J P,Cox S M,Oxley M E. Reliability of manufacturing equipment in complex environments[J]. Annals of Operations Research,2013,209(1):231-254.
[5] Dong M,He D,Banerjee P,et al. Equipment health diagnosis and prognosis using hidden semi-Markov models[J]. International Journal of Advanced Manufacturing Technology,2006,30(7):738-749.
[6] Dong M,He D. Hidden semi-Markov model-based methodology for multi-sensor equipment health diagnosis and prognosis[J]. European Journal of Operational Research,2007,178(3):858-878.
[7] Peng Y,Dong M. A prognosis method using age-dependent hidden semi-Markov model for equipment health prediction[J]. Mechanical Systems&Signal Processing,2011,25(1):237-252.
[8]刘勤明,李亚琴,吕文元,等.基于自适应隐式半马尔可夫模型的设备健康诊断与寿命预测方法[J].计算机集成制造系统,2016,22(9):2187-2194.(Liu Qinming,Li Yaqin,Lyu Wen yuan,et al.Equipment health diagnostics and prognostics method based on adaptive HSMM[J]. Computer Integrated Manufacturing Systems,2016,22(9):2187-2194.)
[9] Fang Xiaolei,Zhou Rensheng,Gebraeel N. An adaptive functional regression-based prognostic model for applications with missing data[J]. Reliability Engineering&System Safety,2015,133(1):266-274.
[10]武昭融,李秀君,李梦晨,等.基于衰变-Markov模型的沥青路面性能预测研究[J].上海理工大学学报,2016,38(2):187-191.(Wu Zhaorong,Li Xiujun,Li Mengchen,et al. Pavement performance prediction of highway based on the decay-Markov prediction model[J].Journal of University of Shanghai for Science and Technology,2016,38(2):187-191.)
[11]Daroogheh N,Baniamerian A,Meskin N,et al. Prognosis and health monitoring of nonlinear systems using a hybrid scheme through integration of PFs and neural networks[J]. IEEE Trans on Systems Man&Cybernetics Systems,2017,47(8):1990-2004.
[12]Zhao Fuqiong,Xie Mingjiang,Tian Zhigang,et al. Integrated equipment health prognosis considering crack initiation time uncertainty and random shock[J]. Chinese Journal of Mechanical Engineering,2017,30(6):1383-1395.
[13]冯磊,王宏力,司小胜,等.基于半随机滤波—期望最大化算法的剩余寿命在线预测[J].航空学报,2015,36(2):555-563.(Feng Lei,Wang Hongli,Si Xiaosheng,et al. Real-time residual life prediction based on semi-stochastic filter and expectation maximization algorithm[J]. Acta Aeronautica et Astronautica Sinica,2015,36(2):555-563.)
[14]陈雄姿,于劲松,唐荻音,等.基于贝叶斯LS-SVR的锂电池剩余寿命概率性预测[J].航空学报,2013,34(9):2219-2229.(Chen Xiongzi,Yu Jinsong,Tang Diyin,et al. Probabilistic residual life prediction for lithium-ion Batteries based on Bayesian LS-SVR[J]. Acta Aeronautica et Astronautica Sinica,2013,34(9):2219-2229.
[15]曾绍华,魏延,唐远炎.剔除支持向量回归中异常数据算法[J].重庆大学学报:自然科学版,2012,35(12):120-132.(Zeng Shaohua,Wei Yan,Tang Yuanyan. Algorithm of removing outliers in SVR[J]. Journal of Chongqing University:Natural Science Edition,2012,35(12):120-132.)
[16]徐艺文,徐宁彬,庄重文,等.面向群智感知车联网的异常数据检测算法[J].湖南大学学报:自然科学版,2017,44(8):145-151.(Xu Yiwen,Xu Ningbin,Zhuang Zhongwen,et al. An algorithm of abnormal data detection for internet of vehicles based on crowdsensing[J]. Journal of Hunan University:Naturnal Science,2017,44(8):145-151.)
[17]Si Xiaosheng,Hu Changhua,Zhou Donghua. Nonlinear degradation process modeling and remaining useful life estimation subject to measurement error[J]. Acta Automatica Sinica,2013,39(5):530-541.