一种基于EMD与SVM的转子系统故障诊断方法.pdfVIP

26 卷 16 期 中 国 电 机 工 程 学 报 Vol.26 No.16 Aug. 2006 2006 年8 月 Proceedings of the CSEE 2006 Chin.Soc.for Elec.Eng. 文章编号：0258-8013 (2006) 16-0162-06 中图分类号：TH 113 文献标识码：A 学科分类号：47040 一种基于经验模式分解与支持向量机的 转子故障诊断方法 于德介，陈淼峰，程军圣，杨 宇 (湖南大学机械与汽车工程学院，湖南省 长沙市 410082) A Fault Diagnosis Approach for Rotor Systems Based on Empirical Mode Decomposition Method and Support Vector Machines YU De-ji e, CHEN Miao-feng, CHENG Jun-sheng, YANG Yu (College of Mechanical and Automotive Engineering, Hunan University, Changsha 410082, Hunan Province, China) ABSTRACT: A fault diagnosis approach for rotor systems 方法能有效地应用于转子系统的故障诊断。 based on EMD method and support vector machine is proposed. 关键词：经验模态分解；自回归模型；支持向量机；故障诊 The EMD method is used to decompose the vibration signal of 断；转子系统 a rotor system into a number of intrinsic mode function 0 引言 components and then the autoregressive mode(AR) model of each IMF component is established. The main auto-regressive 转子系统故障诊断过程 本上可以分 3 个步 parameters and the variances of remnant are regarded as the 骤：①故障信号的采集；②故障特征的提取；③状 feature vectors. Then, the support vector machines used as fault 态识别和故障诊断。在故障特征提取中，传统的转 classifiers are established to identify the condition and fault 子系统故障诊断技术通过分析并提取振动信号时 pattern of the rotor system. Practical examples show that the 域或频域波形的特征量，建立判别函数如神经网络 proposed approach can be applied to the rotor system fault