摘要
提出了一种基于小波包特征熵和支持向量机相结合的故障分类方法,利用小波包分解提取信号的特征熵,然后将得到的特征熵向量输入支持向量机进行故障识别;通过对某型飞机液压系统试验中获取不同的故障特征数据进行分类,结果表明,该方法能利用较少的故障特征得到较高的诊断精度,与BP神经网络相比,采用支持向量机进行故障分类可以获得更高的识别率,表明该方法是有效、可行的。
A fault classification method based on wavelet packet characteristic entropy and support vector machines is proposed.It adopts wavelet packet to decompose the monitored signal and extracts wavelet packet-characteristic entropy,then takes those vectors as fault samples to SVM（support vector machines） for fault classification.Take the result of aircraft hydraulic system for example;fault identification by using SVM is compared with that by using BP neural network.The experimental result shows that the SVM has higher identification accuracy than BP neural network.The results prove the method is efficient and feasible.
出处
《计算机测量与控制》
CSCD
北大核心
2010年第9期1989-1991,1995,共4页
Computer Measurement &Control
基金
航空科学基金项目资助(2008ZC03005)
关键词
小波包分解
特征熵
支持向量机
故障分类
wavelet packet decomposition
characteristic entropy
support vector machines
fault classification
