| 基于有监督核函数主元分析的故障状态识别 |
| |
| 引用本文： | 王新峰,邱静,刘冠军.基于有监督核函数主元分析的故障状态识别[J].测试技术学报,2005,19(2):200-203. |
| |
| 作者姓名： | 王新峰 邱静 刘冠军 |
| |
| 作者单位： | 湖南长沙国防科技大学,机电工程研究所,湖南,长沙,410073;湖南长沙国防科技大学,机电工程研究所,湖南,长沙,410073;湖南长沙国防科技大学,机电工程研究所,湖南,长沙,410073 |
| |
| 基金项目： | 国防科技行业预研基金资助项目(41319040202) |
| |
| 摘 要： | 介绍一种有监督核函数主元分析算法(SupervisedKernelPrincipalComponentAnalysis,SKPCA),通过将样本类内差和类间差融入总体方差中,从而达到更好的分类目的.在齿轮故障诊断实验中,采用SKPCA提取故障信号的非线性特征,实验结果表明SKPCA相比KPCA前两个主元贡献更为集中,故障识别结果更为理想.
|
| 关 键 词： | 核函数主元分析(KPCA) 有监督核函数主元分析(SKPCA) 特征提取 非线性特征 状态识别 |
| 文章编号： | 1671-7449（2005）02-0200-04 |
| 收稿时间： | 2004-09-16 |
| 修稿时间： | 2004年9月16日 |
Machine Fault Condition Recognition Based on Supervised Kernel PCA |
| |
| WANG Xin-feng,QIU Jing,Liu Guan-jun.Machine Fault Condition Recognition Based on Supervised Kernel PCA[J].Journal of Test and Measurement Techol,2005,19(2):200-203. |
| |
| Authors: | WANG Xin-feng QIU Jing Liu Guan-jun |
| |
| Abstract: | A Supervised Kernel Principal Component Analysis (SKPCA) is proposed in the paper, which integrates between and within class variances into KPCA, and the classification performances can be enhanced. In gear fault diagnosis example, SKPCA was applied to extract the nonlinear feature from the fault feature set, the result shows contribution of the first and second principle components are more concentrated by SKPCA than by SPCA, and SKPCA has better classification performance than KPCA. |
| |
| Keywords: | Kernel Principal Component Analysis(KPCA) Supervised Kernel Principal Component Analysis(SKPCA) feature extraction nonlinear feature condition recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
