摘要
故障的特征提取是故障诊断尤其是动态信号分析中经常面临的首要问题,文中通过对小波包正交变换理论的分析,给出了基于小波包的特征提取方法,并采用相对熵标准对其分类能力进行优化。通过对火炮系统的仿真试验,显示了这种方法的有效性,具有一定的工程应用价值。
In pattern recognition or classification, the extraction of fault features is very important in analyzing the dynamic signal. This article is mainly concerned with extracting effective features from the recognized or classified signals by selecting wavelet basis via given training sample sets, The results of a practical diagnosis indicate that the value features are extracted by using the approach, and the way has certain value in engineering applications.
出处
《弹箭与制导学报》
CSCD
北大核心
2006年第3期359-361,共3页
Journal of Projectiles,Rockets,Missiles and Guidance
关键词
小波包
故障
特征提取
火炮
wavelet packets
fault
feature extraction
