引用本文: | 张亚楠,魏武,武林林.基于小波包Shannon熵SVM和遗传算法的电机机械故障诊断[J].电力自动化设备,2010,(1): |
| .Motor mechanical fault diagnosis based on wavelet packet,Shannon entropy,SVM and GA[J].Electric Power Automation Equipment,2010,(1): |
|
|
摘要: |
针对电机机械多故障同时诊断问题,基于小波包、Shannon熵、支持向量机(SVM)和遗传算法,提出了一种电机机械故障诊断新方法,称之为WPSSG法或多模型融合法。该方法选择容错性强的Shannon熵作为特征参数,通过对振动信号进行基于DMeyer小波的小波包分解,提取振动信号的小波包Shannon熵为特征向量,将特征向量作为多类别SVM的输入,具有较高的去噪能力;在训练SVM时,与传统方法多采用试凑法选择参数不同,该方法采用遗传算法对SVM的参数进行全局寻优,使SVM获得最佳的分类性能,具有更高的识别准确率。采用凯斯西储大学提供的电机机械故障数据进行实验,结果证明该方法具有很好的可靠性和准确性。 |
关键词: 电机 故障诊断 小波包 Shannon熵 支持向量机 遗传算法 |
DOI: |
分类号:TM307 |
基金项目: |
|
Motor mechanical fault diagnosis based on wavelet packet,Shannon entropy,SVM and GA |
ZHANG Yanan WEI Wu WU Linlin
|
Abstract: |
A motor mechanical fault diagnosis method,WPSSG,is proposed based on wavelet packet,Shannon entropy,SVM(Support Vector Machine) and GA(Genetic Algorithm).Wavelet packet decomposition and strong fault-tolerant Shannon entropy are used to compute the characteristic vectors of vibration signals,which are then served as the input vectors of multi-class SVM classifier to achieve better denoising performance.GA is used to optimize the parameters of multi-class SVM when it is trained to achieve better classificati... |
Key words: motor fault diagnosis wavelet packet Shannon entropy support vector machine genetic algorithm |