| SVM技术与ANN方法对旋转机械故障诊断性能的比较 |
| |
| 引用本文： | 张金泽,单甘霖.SVM技术与ANN方法对旋转机械故障诊断性能的比较[J].电光与控制,2006,13(3):72-74. |
| |
| 作者姓名： | 张金泽 单甘霖 |
| |
| 作者单位： | 军械工程学院光学与电子工程系,石家庄,050003;军械工程学院光学与电子工程系,石家庄,050003 |
| |
| 摘 要： | 论述了基于支持向量机故障诊断技术的基本原理；介绍了传统的基于人工神经网络的故障诊断方法；以旋转机械故障诊断为例对两种诊断方法进行了比较，实验结果表明，与神经网络相比，基于支持向量机的故障诊断方法在训练速度、诊断精度、可靠性等方面都表现出了优越的诊断性能。
|
| 关 键 词： | 支持向量机 神经网络 故障诊断 |
| 文章编号： | 1671-637X（2006）03-0072-03 |
| 收稿时间： | 2005-05-17 |
| 修稿时间： | 2005-05-172005-07-07 |
Comparison of fault diagnosis performances based on SVM and ANN |
| |
| ZHANG Jin-ze,SHAN Gan-lin.Comparison of fault diagnosis performances based on SVM and ANN[J].Electronics Optics & Control,2006,13(3):72-74. |
| |
| Authors: | ZHANG Jin-ze SHAN Gan-lin |
| |
| Affiliation: | Ordnance Engineering College, Shijiazhuang 050003, China |
| |
| Abstract: | The basic theory of fault diagnosis based on support vector machine is discussed. And then, the traditional fault diagnosis method based on ANN is introduced. Taking the fault diagnosis of turbine generator set for example, a comparison of the two methods is made. The experiment result show that the fault diagnosis method based on SVM has predominant performance on the aspects of training speed, diagnosis precision and reliability. |
| |
| Keywords: | support vector machine(SVM) artificial neural network(ANN) fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
