基于支持向量机的变压器故障分类

A Power Transformer Fault Diagnosis Method Based on Support Vector Machine

基于支持向量机的变压器故障诊断分类方法,通过有限的学习样本,建立变压器故障特征与其溶解气体之间的关系。利用获得的故障变压器数据建立了故障分类器,通过对样本的分类输出检验,验证了该故障诊断方法的可行性。

Support Vector Machines （SVM） is a machine-learning algorithm based on statistical learning theory. The method for power transformer fault diagnosis based on SVM is proposed in this paper. The principle and algorithm of this method are introduced. Through a finite learning sample the relation is established between the transformer fault signature and the quantity of its dissolved gas. A fault classifier is constructed by using the dissolved gas data of the fault transformer. The testing results show that this method can successfully be applied to the diagnosis of gear faults.