首页 | 本学科首页   官方微博 | 高级检索  
     

基于支持向量机的旋转机械非线性故障诊断研究
引用本文:王秉仁,刘兆阳,张家伟,田丽洁.基于支持向量机的旋转机械非线性故障诊断研究[J].煤矿机械,2005(2):122-123.
作者姓名:王秉仁  刘兆阳  张家伟  田丽洁
作者单位:华北电力大学,机械工程学院,河北,保定,071003
摘    要:故障样本不足是制约故障诊断技术向智能化方向发展的主要原因之一 ,支持向量机(SVM)是一种基于统计学习理论 (SLT)的机器学习算法 ,它能在训练样本很少的情况下达到很好的分类效果 ,从而为故障诊断技术向智能化发展提供了新的途径。介绍了支持向量机分类算法 ,探讨了该算法在故障诊断领域中的应用 ,并利用不同的核函数与BP神经网络分类方法进行了对比研究。结果表明 ,SVM方法在小样本情况下的分类效果优于BP神经网络。

关 键 词:统计学习理论(SLT)  支持向量机(SVM)  故障诊断  神经网络
文章编号:1003-0794(2005)02-0122-02
修稿时间:2004年11月18

Researched Non-Linear Fault Diagnosis of Rotating Machine on Support Vector Machine
WANG Bing_ren,LIU Zhao_yang,ZHANG Jia_wei,TIAN Li_jie.Researched Non-Linear Fault Diagnosis of Rotating Machine on Support Vector Machine[J].Coal Mine Machinery,2005(2):122-123.
Authors:WANG Bing_ren  LIU Zhao_yang  ZHANG Jia_wei  TIAN Li_jie
Abstract:Shortage of fault samples is one of the main reasons that restricting the developing of fault diagnosis. The support vector machine (SVM) is a machine-learning algorithm based on the statistical theory (SLT), which has desirable classification ability even if with fewer samples, SVM provides us a new method to develop the intelligent fault diagnosis. In this paper, the classification algorithm of support vector machine and its application in fault diagnosis are discussed. The result of fault diagnosis by using different kernel function is compared with that by using BP neural network, which shows that the SVM has higher classification ability than BP neural network in the case of fewer samples.
Keywords:statistical learning theory  support vector machine  fault diagnosis  neural network
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号