| 基于EMD多模态特征融合支持向量机的故障诊断 |
| |
| 引用本文： | 沈志熙,黄席樾,马笑潇.基于EMD多模态特征融合支持向量机的故障诊断[J].控制与决策,2009,24(6). |
| |
| 作者姓名： | 沈志熙 黄席樾 马笑潇 |
| |
| 作者单位： | 1. 重庆大学,自动化学院,重庆,400030
2. 新西兰况得实仪器有限公司,北京代表处,北京,100016 |
| |
| 摘 要： | 针对非平稳时间序列信号,提出一种基于经验模态分解(EMD)的特征提取和多模态特征融合支持向量机的故障诊断方法,首先对原始信号进行EMD分解,选择能量最大的几个基本模式分量(IMF)并提取其小波包特征,获得对每个IMF独立的特征子集;然后在每个IMF特征子集中训练SVM弱分类器,并根据各特征子集对应的IMF能量权重进行加权融合,获得故障状态的强分类器,将该方法应用于6135型柴油机振动信号故障诊断中,实验结果表明了其可行性和有效性.
|
| 关 键 词： | 故障诊断 非平稳时间序列 经验模态分解 基本模式分量 支持向量机 |
Fault diagnosis method based on empirical mode decomposition and support vector machine |
| |
| SHEN Zhi-xi,HUANG Xi-yue,MA Xiao-xiao.Fault diagnosis method based on empirical mode decomposition and support vector machine[J].Control and Decision,2009,24(6). |
| |
| Authors: | SHEN Zhi-xi HUANG Xi-yue MA Xiao-xiao |
| |
| Affiliation: | 1.College of Automation;Chongqing University;Chongqing 400030;China;2.Beijing Representative Office;New Zealand Commtest Instruments Ltd;Beijing 100016;China |
| |
| Abstract: | For non-stationary time series signal,a fault diagnosis method based on empirical mode decomposition(EMD)and multi-features fusion support vector machine(SVM)is proposed.Firstly,EMD is employed to decompose the original signal,the several energy-dominating intrinsic mode functions(IMFs)are chosen and the wavelet packet features of each of IMFs are extracted respectively to form several independent feature sub-spaces.Then,a series of SVM-based weak classifiers are trained in each of feature sub-spaces respec... |
| |
| Keywords: | Fault diagnosis Non-stationary time series Empirical mode decomposition Intrinsic mode function Support vector machine |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
