分层隐Markov模型在设备状态识别中的应用研究  

Research on Gearbox State Recognition Based on Hierarchical Hidden Markov Model

作　　者：滕红智[1,2] 贾希胜[1] 赵建民[1] 张星辉[1] 王正军[1] 葛家友 

机构地区：[1]军械工程学院 [2]68129部队

出　　处：《中国机械工程》2011年第18期2175-2181,共7页China Mechanical Engineering

基　　金：总装备部重点预研基金资助项目(9140A27020308JB34)

摘　　要：与传统的隐Markov模型(HMM)相比较而言,应用分层隐Markov模型(HHMM)对设备进行状态识别有诸多优点,而且能以概率的形式更为精确地计算识别结果。针对模型参数随着设备状态的增加呈指数倍增这一问题,引入动态贝叶斯网络这一新的方法,由于该方法可以有效地降低模型的计算复杂度并缩短推理时间,所以将HHMM表达为动态贝叶斯网络,利用预处理的振动信号对设备的健康状态进行识别;针对现有状态分类方法的局限性,提出了基于K均值算法和交叉验证方法相结合的状态数优化方法;以齿轮箱全寿命实验为依据,对该模型实现状态识别的基本框架和计算过程进行了研究,研究结果为复杂设备的状态识别提供了新的思路。HHMM has many advantages for state recognition and more accurately calculates recog- nition results in the form of probability, in comparison with traditional hidden Markov model （HMM）. Model parameters increased exponentially with the increasing equipment state. In view of this, dynamic Bayesian network was introduced, which can effectively reduce the computational com- plexity and decrease the inference time. Accordingly, HHMM was expressed as dynamic Bayesian network, which identified health status by utilizing vibration signals of pretreatment. In order to a- void the limitations of the current state classifications,the optimization of the condition numbers was proposed,on the basis of combination of K--means algorithm and cross--validation. It also investiga- ted the basic framework for HHMM state recognition and calculation process based on full life test for gearbox, which provides a new way for state recognition of complex equipment.

关 键 词：分层隐Markov模型 状态识别 动态贝叶斯网络 状态数优化 

分 类 号：TH17[机械工程—机械制造及自动化]