基于遗传编程和支持向量机的故障诊断模型.pdfVIP

第 38 卷 　第 3 期 西　安　交　通　大　学　学　报 Vol. 38 　№3 2004 年 3 月 　 Mar. 2004 J OU RNAL OF XI′AN J IAO TON G UN IV ERSIT Y 基于遗传编程和支持向量机的故障诊断模型 李良敏 , 屈梁生 (西安交通大学机械工程学院 , 710049 , 西安) 摘要 : 提出了一种基于遗传编程和支持向量机的故障诊断模型. 该模型利用遗传编程对传统的时域指标进 行特征选择和提取 ,得到更能反映信号本质的特征 ,与其他特征组合后作为识别特征输入多类支持向量机 , 实现了对机器不同类型故障的识别. 实验结果表明 , 同传统时域指标相比 ,经过遗传编程选择和提取的特征 对轴承的故障具有更好的识别能力 ,进而提高了多类支持向量机的分类准确性. 关键词 : 故障诊断 ;支持向量机 ;遗传编程 ;滚动轴承 ( ) 中图分类号: TH17 　文献标识码 : A 　文章编号 : 0253 - 987X 2004 03 - 0239 - 04 Fault Detection Based on Genetic Programming and Support Vector Machines L i L ian gmin , Q u L ian gshen g ( School of Mechanical Engineering , Xi ′an Jiaotong University , Xi ′an 710049 , China) Abstract : A new classification model based on genetic pro gramming and support vector machine for machine fault diagnosis was proposed. In this model , genetic pro gramming constructs and selects features from original feature set . The selected features form input feature set of support vector machines. Then multiclass support vector machine is applied to detect abnormal cases from normal ones. Experiments of rolling bearings fault detec tion are carried out to test the performance of this model. Practical results show that the compound features gen erated by genetic pro gramming possess better recognition ability than the initial time domain features do. The classification ability of multiclass support vector machine is improved after feature extraction and selection. Keywords : f ault detection ; sup port vector m achines ; genetic pro gram min g ; rolling bearing 　　机械故障诊断就其本质来