摘要
隐半马尔科夫模型在进行系统状态估计及寿命预测时,其状态转移概率矩阵是固定值,得到的剩余寿命预测值呈阶梯状变化,与系统的实际剩余寿命值之间存在着较大的误差.针对上述问题,提出了具有时变状态转移概率矩阵的隐半马尔科夫模型,根据系统的3种典型退化状态分析,给出3种不同的状态转移系数.与初始状态转移矩阵相结合,得到随时间变化的状态转移矩阵.提高系统在当前健康状态下的剩余持续时间估计精度,最终得到更为准确的总体剩余寿命预测值.结果表明,基于时变状态转移概率矩阵的隐半马尔科夫模型相比传统的隐半马尔科夫模型,可显著提高剩余寿命预测的准确性.
In system state recognition and prognostics,state transition probability matrix of hidden semi-Markov model(HSMM)is constant and the predicted life value shows stepladder change,which is different from the actual residual life of the system.To solve this problem,an HSMM with time varying state transition probability matrix was proposed.Based on the analysis of three typical degradation states of the system,three different state transition coefficients were given.Combined with initial state transition matrix,a time varying state transition matrix was obtained,the estimation accuracy of residual life of the system under current healthy state was increased,and a more accurate overall residual life prediction value can be obtained.Experiment results show that,compared with traditional HSMM,HSMM based on time varying state transition probability matrix can increase the accuracy of residual life prediction and can be used in life prediction with high precision.
引文
[1]LEE Jay,WU Fang-ji,ZHAO Wen-yu.Prognostics and health management design for rotary machinery systems-reviews,methodology and applications[J].Mechanical Systems and Signal Processing,2014,42(1/2):314-334.
[2]杨宇,王欢欢,曾鸣,等.基于变量预测模型的模式识别方法在滚动轴承故障诊断中的应用[J].湖南大学学报:自然科学版,2013,40(3):36-40.YANG Yu,WANG Huan-huan,ZENG Ming,et al.Application of pattern recognition approach based on VPMCD in roller bearing fault diagnosis[J].Journal of Hunan University:Natural Sciences,2013,40(3):36-40.(In Chinese)
[3]MOORE M D,SAVIC M I.Speech reconstruction using ageneralized HSMM(GHSMM)[J].Digital Signal Processing,2004,14(1):37-53.
[4]CAREY B,DAN M,TARIK A.Condition-based maintenance of machines using hidden markov models[J].Mechanical Systems and Signal Processing,2000,14(4):597-612.
[5]YU Shun-zheng.Hidden semi-markov models[J].Artificial Intelligence,2010,174(2):215-243.
[6]DONG Ming,HE David.A segmental hidden semi-Markov model(HSMM)-based diagnostics and prognostics framework and methodology[J].Mechanical Systems and Signal Processing,2007,21(5):2248-2266.
[7]DONG Ming,HE David.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.
[8]CHEN A,WU G S.Real-time health prognosis and dynamic preventive maintenance policy for equipment under aging Markovian deterioration[J].International Journal of Production Research,2007,45(15):3351-3379.
[9]RABINER L R.A tutorial on hidden Markov models and selected applications in speech recognition[C]//Proceedings of the IEEE.1989,77:257-286.
[10]LEVINSON S E.Continuously variable duration hidden Markov models for automatic speech recognition[J].Computer Speech and Language,1986,1(1):29-45.
[11]RUSSELL M J,MOORE R K.Explicit modeling of state occupancy in hidden Markov models for automatic speech recognition[C]//Acoustics,Speech and Signal Processing,IEEE International Conference on ICASSP'85.1985,10:5-8.
[12]王宁,孙树栋,李淑敏.基于DD-HSMM的设备运行状态识别与故障预测方法[J].计算机集成制造系统,2012,18(8):1861-1868.WANG Ning,SUN Shu-dong,LI Shu-min.Equipment state recognition and fault prognostics method based on DDHSMM model[J].Computer Integrated Manufacturing Systems,2012,18(8):1861-1868.(In Chinese)