摘要
机组的振动水平是表征电厂稳定安全最重要的标志之一。本文利用支持向量机的智能方法对机组的轴系故障进行诊断,在小样本集上取得了100%的分类精度。在此基础上,还引入部分噪声数据,统计其分类性能,展示了支持向量机的容错能力。最后分析了支持向量机方法在轴系振动故障振动的优势和缺陷,引入模糊输出支持向量机进行了改进,给设备维修提供了更多的参考信息。
The vibration level of the unit is one of the most important parameters of power plant's safety and steady.In this paper,support vector machines method is applied to fault diagnosis of shaft vibration.The accuracy of classification on a small sample set is 100%.Then,some different levels of noise are added to the samples,which show the fault tolerance of support vector machines.At last,the advantages and disadvantages of support vector machines are analyzed.The fuzzy output support vector machine is used.The improved method can give more information for further decision.
出处
《节能技术》
CAS
2007年第5期423-425,469,共4页
Energy Conservation Technology
关键词
模糊支持向量机
轴系振动
故障诊断
fuzzy output support vector machines
rotor vibration
fault diagnosis