基于SLPP和CHMM的轴承健康评估与预测研究  

Bearing Health Assessment and Prediction by Supervised Local Preserving Projection and Coupled Hidden Markov Model

作　　者：陈庆[1] 刘韬[1] 伍星[1,2] CHEN Qing;LIU Tao;WU Xing(Faculty of Mechanical and Electrical Engineering,Kunming University of Science and Technology,Kunming 650500,China;Yunnan Vocational College of Mechanical and Electrical Technology,Kunming 650201,China)

机构地区：[1]昆明理工大学机电工程学院,云南昆明650500 [2]云南机电职业技术学院,云南昆明650201

出　　处：《昆明理工大学学报（自然科学版）》2022年第3期67-74,共8页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：国家自然科学基金项目(52065030,51875272);云南省重点研发计划项目(202102AC080002XX)。

摘　　要：为了更加有效地评估和预测轴承健康状态,更早地发现轴承异常,提出了一种基于有监督局部保持投影(SLPP)和耦合隐马尔可夫模型(CHMM)的轴承健康评估与预测方法.首先,针对轴承信号特点,选择11个时域和频域特征进行提取;其次,利用SLPP方法对提取的高维特征进行高质量特征约减;最后,结合CHMM获取性能指标并对轴承健康状态进行评估和预测.通过轴承全寿命加速疲劳实验分析表明,该方法能够对轴承健康状态进行准确而有效的评估预测;在特征约减效果上,与PCA相比,SLPP刻画轴承状态变化过程的能力明显更强;在评估效果上,与HMM相比,CHMM能更早地发现轴承异常.A methodology based on supervised local preserving projection(SLPP)and coupled hidden Markov model(CHMM)was presented in order to effectively assess and forecast the degradation of bearing performance and to find abnormal bearing as soon as possible.At first,eleven characteristics of time domain and frequency domain were extracted according to the characteristics of bearing signals.And then high-dimensional feature was converted to low dimensional space using SLPP.At last,the pre-trained CHMM was employed to assess and predict the performance degradation of bearings quantitatively.A bearing accelerated life experiment was performed to validate the feasibility and validity of proposed method.In terms of feature reduction effect,SLPP has significantly stronger ability to describe bearing degradation process than PCA.In the evaluation effect,CHMM can detect bearing abnormalities earlier than HMM.

关 键 词：轴承 机械故障诊断 健康评估与预测 有监督局部保持投影 耦合隐马尔可夫模型 

分 类 号：TH133.3[机械工程—机械制造及自动化]