基于支持向量机的发电机匝间转子绕组短路故障诊断

摘要

具有结构风险最小化原则的支持向量机(SVM)对于小样本决策具有较好的学习推广性,并且故障样本的不足在一定程度上制约了基于知识的方法在故障诊断中的运用.针对这一问题,提出了利用支持向量机的方法对匝间转子绕组短路故障诊断方法.该方法利用小波分析对探测线圈测得感应电动势进行处理构造特征向量,然后输入到支持向量机的多故障分类器中进行故障识别.实验数据表明该方法是可行、有效的,并且在小样本的情况下,较BP神经网络有更好的分类效果.%Support vector machine (SVM) with minimization of the structure risk principle has good learning extension for small sample decision, and lack of fault samples restricted application of fault diagnosis.Aiming at this problem, this paper puts forward a fault diagnosis methods based on SVM to solve generator rotor winding inter turn short circuit The paper puts forward the energy band wavelet analysis method to extract the eigenvalues of fault diagnosis, and put the data into the SVM to identification.The experimental data show that the method is feasible and effective, and in the case of small sample, SVM have a better classification effect than a BP neural network.