滚动轴承技术故障诊断的支持向量机方法研究

针对当前故障诊断中几种常用方法的不足,首次提出将支持向量机方法应用于滚动轴承技术故障诊断。该文提出的两种算法其核心均是利用支持向量机方法对样本进行分类。支持向量机方法基于小学习样本条件下,通过寻求结构风险最小,以期获得良好的分类效果和泛化能力。两种实验结果表明,在选用合适核函数及参数条件下,支持向量机具有学习速度快、诊断正确率高的优良性能,这一结论表明了该文所提出方法的优越性。

A Novel SVM Approach to the Technique State Diagnosis of the Trundle Bearing

Based on analyzing the shortages of several approaches in fault diagnosis,a Support Vector Machine(SVM)method is proposed and applied to the technique state diagnosis of the trundle bearing for the first time.The cores of the two algorithms presented in this paper are to classify the samples with SVM method.The SVM method is based on seeking on the Structural Risk Minimization by few learning samples supporting,and it has important feature such as good generalization and classification performance,etc.The experimental results show that the SVM approach has learning fast and a good diagnosis performance under the condition of the appropriate kernel function and its parameters being selected,this conclusions indicate the proposed approach has the superiority.