Journal of Changshu Institute of TechnologyIssue(2)：70-75,6.
基于奇异值分解和支持向量机的故障诊断方法研究
A Study of Fault Diagnosis Method Based on Singular Value Decomposition and Support Vector Machine
刘吕亮 1徐斌1
作者信息
- 1. 常德职业技术学院 汽车工程系，湖南 常德 415000
- 折叠
摘要
Abstract
In view of the weak fault signal of the equipment and the serious noise, the method of equipment fault diagnosis based on singular value decomposition (SVD) and support vector machine (SVM) is proposed in this paper. The cluster analysis is used to pretreat equipment status signals in order to remove signal anomalies and improve the accuracy of the signal. The matrix is constructed after pretreatment by choosing the right win⁃dow length and the singular value as the signal features. SVM is used for the identification and classification of signal characteristics in order to avoid the uncertainty of output by SVM, and record the training output as a de⁃cision table. Verified by experiments based on SVD and SVM, fault diagnosis can be performed reliably, accu⁃rately, and quickly.关键词
故障诊断/奇异值分解/支持向量机/聚类Key words
fault diagnosis/singular value decomposition/support vector machine/clustering分类
机械制造引用本文复制引用
刘吕亮,徐斌..基于奇异值分解和支持向量机的故障诊断方法研究[J].Journal of Changshu Institute of Technology,2015,(2):70-75,6.
