变精度粗糙集和支持向量机在轴承故障诊断中的应用

Application of Variable Precision Rough Set and SVM in Mechanical Fault Diagnosis of Rolling Bearing

针对故障诊断中样本缺乏以及样本数据中难免存在噪声等问题,提出了变精度粗糙集与支持向量机杂合的故障诊断方法:先用变精度粗糙集理论提取故障诊断的特征,获得最优决策系统,在此基础上设计了SVM多分类器进行故障诊断。轴承故障诊断的仿真结果验证了变精度粗糙集理论与支持向量机杂合的诊断方法的可行性。

Considering the lack of sample and noise problems which inevitably exists in the fault diagnosis information, the paper puts forward a fault diagnosis method based on variable precision rough set and support vector machine. Firstly, fault diagnosis features are extracted to obtain the the optimal decision system by using variable precision rough set theory. Then the SVM multi classifier are designed for fault diagnosis. The actual bearing fault diagnosis results verify the feasibility of the proposed fault diagnosis method based on variable precision rough set theory and SVM.