摘要
针对神经网络故障诊断存在的诸多问题,提出了基于最小二乘支持向量机的机车轴承故障诊断方法。采用最小二乘支持向量机方法建立多类故障分类器,对输入的特征向量信息进行分类,完成故障诊断功能。仿真证明了最小二乘支持向量机在小样本情况下比神经网络具有更强的泛化能力,用于故障诊断时在识别准确率和抗干扰能力方面有明显的优势。
For the problems of neural network for fault diagnosis,a method of fault diagnosis for locomotive bearing based on least squares support vector machine is proposed.The multi-class fault classifier is built based on least squares support vector machine.The inputing feature vectors are classed by the classifier and the function of classifier is completed.The simulation result not noly shows that the generalized ability of least squares support vector is better than neural network in the case of small samples ...
出处
《电气传动自动化》
2009年第6期14-16,35,共4页
Electric Drive Automation
关键词
最小二乘支持向量机
故障诊断
多类故障分类
机车轴承
least squares support vector machine
fault diagnosis
multi-class fault classification
locomotive bearing
