| 支撑向量机在航空发动机故障诊断中的应用研究 |
| |
| 引用本文： | 许将军,侯宽新,高丽霞. 支撑向量机在航空发动机故障诊断中的应用研究[J]. 自动化技术与应用, 2011, 30(4): 69-71 |
| |
| 作者姓名： | 许将军 侯宽新 高丽霞 |
| |
| 作者单位： | 1. 中国民航飞行学院航空工程学院,四川广汉,618307
2. 中国民航飞行学院广汉分院,四川,广汉618307 |
| |
| 摘 要： | 航空发动机故障样本有限,利用传统的统计识别方法故障诊断,正确率不高.支撑向量机能解决小样本的故障分类识别问题.研究Support Vector Machine(简称SVM)核函数对识别精度的影响,并把SVM与最大似然法、马氏距离法,最小距离法进行比较,结果表明SVM核函数对故障识别正确率影响不大,基于SVM的航空发动机...
|
| 关 键 词： | 支撑向量机 核函数 故障诊断 |
Applied Research of Support Vector Machines in the Fault Diagnosis of Engine |
| |
| XU Jiang-jun,HOU Kuan-xin,GAO Li-xia. Applied Research of Support Vector Machines in the Fault Diagnosis of Engine[J]. Techniques of Automation and Applications, 2011, 30(4): 69-71 |
| |
| Authors: | XU Jiang-jun HOU Kuan-xin GAO Li-xia |
| |
| Affiliation: | XU Jiang-jun1,HOU Kuan-xin,GAO Li-xia1(1.College of Aeronautical Engineering Civil Aviation Flight University of China,Guanghan 618307 China,2.Civil Aviation Flight University of China,Guanghan Branch,Guanghan China) |
| |
| Abstract: | As lack of fault samples of engine,the accuracy of traditional classification methods is always unsatisfactory.Support vector machine（SVM）is a good method for solving limited sample problem.the influences of different kernel functions on classification accuracy are researched,and the classification results by SVM are compared with that of other methods（such as Maximum likelihood method 、Minimum Distance method、Mahalanobis Distance）.Experiment results show that the SVM method has better recognition accuracy than traditional algorithms,and the recognition accuracy is almost identical for different kernel functions. |
| |
| Keywords: | support vector machines kernel function fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
