机构地区: 中南大学信息科学与工程学院
出 处: 《铁路计算机应用》 2007年第9期52-54,共3页
摘 要: 通过对机车轴承振动信号的分析处理,提出基于支持向量机(SVM)的故障诊断方法,提取反映轴承运行状态的无量纲系数作为故障的特征向量,并以此作为输入来建立支持向量机分类器,利用SVM网络的智能性来判断机车轴承的工作状态和故障类型。实验结果表明,提出的方法在小样本的情况下仍能准确、有效地对机车轴承的工作状态和故障类型进行分类,实现机车轴承故障的智能诊断。 By analyzing and processing the vibration signals of locomotive bearing, a fault diagnosis approach based on support vector machine was presented. The non-dimension coefficients which represented operating state of the locomotive bearing were extracted to construct the characteristic vectors. It was taken the vectors as the inputs of the SVM network to definite the fault classifier of the support vector machine. The condition and fault pattern of the locomotion bearing could be identified with the intellectual ability of SVM network. Practical examples showed that the proposed approach which could classify the condition and fault pattern of the locomotion bearing accurately and effectively even when the number of samples was small. It was implemented the intelligent diagnosis to locomotive bearing faults.