摘要
滚动轴承作为旋转机械最重要的零部件之一,其可靠性和寿命直接影响着机器的可靠性和寿命,为解决滚动轴承可靠性难以估计的问题,提出一种基于概率核主成分分析(Probabilistic Kernel Principal Component Analysis,PKPCA)和Logistic回归模型(Logistic Regression Model,LRM)的滚动轴承可靠性评估方法.首先提取轴承的时域、频域和时频域特征值组成高维混合域特征集,并引入相对特征值降低轴承个体差异;然后用PKPCA挑选能够表征轴承退化状态的特征值作为Logistic回归模型的协变量;最后用Logistic回归模型对滚动轴承可靠性进行评估.通过IMS滚动轴承全寿命试验,验证了该方法的有效性.
Rolling bearing is one of the most important components of the rotating machinery,its reliability directly affects the reliability and life of the machine.In order to solve the problem which is difficult to estimate the reliability of rolling bearing,this paper presents a method based on probabilistic kernel principal component analysis(PKPCA)and logistic regression model(LRM)to evaluate the reliability of rolling bearing.Firstly,the time domain,frequency domain and time-frequency domain features of rolling bearing are extracted to form the high dimensional hybrid domain feature set,and the relative eigenvalue is introduced to reduce the individual differences of bearings;Then the principal components which can accurately reflect the performance degradation of rolling bearing are obtained by PKPCA as the covariates of logistic regression model;Finally,the covariates are brought into logistic regression model to obtain the reliability of the rolling bearing.The results verified by IMS full life test of rolling bearing show that the method can accurately evaluate the reliability of rolling bearings.
作者
王萌
王奉涛
WANG Meng;WANG Feng-tao(School of Mechanical Engineering,Dalian University of Technology,Liaoning Dalian116023,China)
出处
《机械设计与制造》
北大核心
2020年第6期138-141,共4页
Machinery Design & Manufacture
基金
国家自然科学基金—基于改进COX模型的航空发动机关键部件寿命预测理论与方法研究(51375067)
基于模型与数据驱动的航空液压管路系统振动故障诊断方法研究(51775257)。
关键词
概率核主成分分析
轴承
性能退化
混合域
LOGISTIC回归模型
小波包样本熵
Probabilistic Kernel Principal Component Analysis(PKPCA)
Rolling Bearing
Reliability Evaluation
Hy-brid Domain
Logistic Regression Model
Sample Entropy of Wavelet Packet
