摘要
提出一种基于特征评估和特征加权FCM算法的滚动轴承故障诊断方法.对原始振动信号提取时域、频域和小波包归一化能量特征,组成联合特征.然后对联合特征进行评价,计算类可分性评价指标.根据该指标大小选取敏感特征,进行特征加权模糊聚类分析,实现对轴承故障状态的自动识别.特征评估克服了传统方法在特征选择上的盲目性,特征加权提高了分类准确率.实例表明,该算法不仅可以可靠识别不同类型的滚动轴承故障,而且可以识别不同程度的故障.
A new method for rolling bearings fault diagnosis was proposed, which was based on feature evaluation and feature weighted Fuzzy C-Means （FCM） algorithm. The time domain features, frequency domain features, and wavelet packet energy features were extracted from the original signals, and these features constructed the combined features. The feature evaluation method was applied to calculate the class separability criterion of the features, and the corresponding sensitive features were selected according to the criterion. Then these selected features were used to identify different fault conditions of bearings. The experimental results demonstrate that the proposed method not only can identify different types of fault, but also can identify different degrees of fault.
出处
《武汉理工大学学报（交通科学与工程版）》
2010年第1期72-75,共4页
Journal of Wuhan University of Technology（Transportation Science & Engineering）
基金
国家自然科学基金项目资助(批准号:50275089)
关键词
特征评价
特征加权
聚类分析
故障诊断
小波包
feature evaluation
feature weight
cluster analysis
fault diagnosis
wavelet packet
