摘要
针对飞机强度试验中异常信号被背景噪声淹没、提取信号特征困难这一问题,提出了结合相关系数的集合经验模态分解方法(EEMD)降噪方法。首先对信号进行EEMD降噪,然后根据相关系数筛选出用于重构信号的IMF分量,提取特征值,最后运用支持向量机进行分类辨识。通过与几种降噪法进行比较,结果表明,结合相关系数的EEMD降噪方法优于其它降噪方法,更适用于充满噪声的全尺寸飞机强度试验中。
Aiming at the problems that the abnormal sound signal is buried in the background noises in aircraft strength test and it is difficult to obtain available abnormal sound signal feature,the noise reduction method based on ensemble empirical mode decomposition（ EEMD） and correlation coefficient is proposed. Firstly,abnormal sound signal is decomposed by EEMD and a set of intrinsic mode function components are obtained. Filtering the components according to the correlation coefficient,the selected IMF components are used to reconstruct the signal. Lastly,energy feature extracted from a number of IMFs could serve as input vectors of support vector machine.Practical examples show that the method has obvious advantages compared with other noise reduction methods and can be used in aircraft strength test with the strong background noise.
出处
《工程与试验》
2016年第3期9-13,74,共6页
Engineering and Test
基金
航空基金(20150981006,20140937001)资助
关键词
集合经验模态分解方法
降噪
相关系数
特征提取
支持向量机
ensemble empirical mode decomposition
noise reduction
correlation coefficient
feature extraction
support vector machine
