结合马氏距离与隶属度函数的轴承性能退化研究

Research performance degradation of rolling bearing with mahalanobis distance and degree of membership function

提出一种融合自回归模型(auto-regressive model,AR)、马氏距离(mahalanobis distance,MD)与隶属度函数的滚动轴承的性能退化评估方法。利用自回归模型对轴承全寿命周期数据进行特征提取,将提取的自回归系数及残差作为特征参数。经过归一化处理后,分别用无故障样本与故障样本特征输入马氏距离得到两个距离指标,再输入隶属度函数中,得到轴承退化指标(degradation index,DI),描绘性能退化曲线,并对信号进行包络谱分析,验证初始故障位置。利用美国辛辛那提大学的轴承全寿命周期数据验证该方法的有效性和实用性。

This paper presents a degeneration state recognition method of rolling bearings combined with auto-re-gressive model (AR), mahalanobis distance (MD) and degree of membership function. The paper ex-tracts the feature of the full life cycle of rolling bearings using auto-regressive model and the auto-re-gressive coefficients and residuals are taken as the characteristic parameters. After the features are nor- malized, two distance indicators are obtained by inputting the Mahalanobis distance model with the nor-mal and failure sample features. Then input the membership function to obtain the bearing degradation index ( DI) and describe the performance degradation curve. And envelope spectrum analysis is per-formed to verify the initial fault location. The validity and practicability of the method is verified by the full life cycle of rolling bearings of the University of Cincinnati, USA.