基于油液光谱数据的离合器剩余寿命随机过程预测  

Stochastic Process Prediction of Clutch Remaining Life Based on Oil Spectral Data

作　　者：张江 崔俊杰[1] 郑长松[2] 刘勇[1] 刘亚军 沈剑 ZHANG Jiang;CUI Jun-jie;ZHENG Chang-song;LIU Yong;LIU Ya-jun;SHEN Jian(College of Mechatronics Engineering,North University of China,Taiyuan 030051,China;School of Mechanical Engineering,Beijing Institute of Technology,Beijing 100081,China;Inner Mongolia North Heavy Industry Group Co.,Ltd.,Baotou 014000,China)

机构地区：[1]中北大学机电工程学院,山西太原030051 [2]北京理工大学机械与车辆学院,北京100081 [3]内蒙古北方重工业集团有限公司,内蒙古包头014000

出　　处：《光谱学与光谱分析》2022年第8期2631-2636,共6页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金项目(62003314);山西省自然科学基金项目(201901D111131);传动国防科技重点实验室基金项目(141009AT03091H);山西省研究生创新项目(2021Y577)资助。

摘　　要：基于油液光谱数据的湿式离合器剩余寿命预测对于综合传动装置的状态监测及可靠性具有显著影响。针对油液光谱数据随机性较高以及现有方法的性能指标单一且误差偏大的问题,利用二元Wiener过程的预测实时性和准确的优点,开展离合器剩余寿命预测研究。首先,结合湿式离合器寿命试验,通过对全寿命周期光谱数据的补换油修正处理,提取出离合器剩余寿命预测的指示元素Cu和Pb及失效阈值;其次,通过Matlab Copula函数分析指示元素的相关关系特性,由此推导出剩余寿命的关联函数;再次,根据Inverse Gaussian原理,建立上述两种指示元素的一元、二元wiener过程的性能退化数学模型;最后,利用极大似然估计法进行参数估计,并采用一元、二元性能退化数学模型分别对被试离合器的剩余寿命开展了随机过程预测研究。通过预测结果与试验结果的对比,在150~240 h范围内,二元wiener过程剩余寿命预测的偏差为6%~22%;相较一元wiener过程,剩余寿命预测的准确率提升9%以上。研究结果表明,二元wiener过程模型及其预测方法用于湿式离合器的剩余寿命预测,具有预测实时性强且预测精度高的优点。同时,该方法可推广到设备状态在线监测及剩余寿命预测等相关领域。The residual life prediction of wet clutch based on oil spectrum data significantly impacts on the condition monitoring and reliability of integrated transmission device.Aiming at the problems of high randomness of oil spectral data and single performance index and large error of existing methods,the prediction of clutch remaining life is carried out using the advantages of real-time and accuracy of binary Wiener process.Firstly,combined with the wet clutch life test,the indicator elements Cu and Pb and the failure threshold of the remaining life prediction of the clutch are extracted through the oil supplement and change correction of the spectral data of the whole life cycle;Secondly,the correlation characteristics of indicator elements are analyzed by MATLAB copula function,and the correlation function of residual life is derived;Thirdly,according to the inverse Gaussian principle,the performance degradation mathematical models of the unary and binary Wiener processes of the above two indicator elements are established;Finally,the maximum likelihood estimation method is used to estimate the parameters,and the univariate and binary performance degradation mathematical models are used to predict the remaining life of the tested clutch.By comparing the predicted results with the experimental results,the deviation of residual life prediction of binary Wiener process is 6%~22%in the range of 150~240 h;Compared with the univariate Wiener process,the accuracy of residual life prediction is improved by more than 9%.The results show that the binary Wiener process model and its prediction method have the advantages of real-time solid prediction and high prediction accuracy.At the same time,this method can be extended to related fields such as on-line monitoring of equipment status and residual life prediction.

关 键 词：油液光谱 湿式离合器 二元wiener过程 剩余寿命预测 

分 类 号：U463.12[机械工程—车辆工程]