支持向量机在机械故障诊断中的应用研究

在机械故障诊断中,通常不具备有大量的故障样本,因此,制约了故障诊断技术向智能化方向发展。而基于统计学习理论(SLT)的支持向量机(SVM)方法正好克服了这方面的不足。统计学习理论是专门研究少样本情况下的统计规律及学习方法的理论。SLT理论和SVM方法为故障诊断技术向智能化发展提供了新的途径。该文讨论了支持向量机在故障诊断领域中应用的分类算法。并以滚动轴承的振动信号为例进行了试验论证。试验表明:SVM方法对具有少样本的故障诊断领域具有很强的适应性。

Application Research on Support Vector Machines in Mechanical Fault Diagnosis

Generally,there are no large numbers of fault samples in mechanical fault diagnosis.So,the development of intelligent diagnosis is restricted.The support vector machine(SVM)based on the statistical learning theory(SLT)can solve this problem.SLT is a statistical theory,which concerns mainly the statistic principles with small sample.SLT and SVM provide us with a new method to develop the intelligent fault diagnosis.In this paper,the classification algorithm of sup-port vector machines is discussed.Experimentation with vibration signals of rolling bearing is conducted and the results prove that the SVM has high adaptability for fault diagnosis in the case of smaller number of samples.