变分模态分解和熵理论在超声信号降噪中的应用

发布时间：2018-06-05 09:10

本文选题：超声检测 + 降噪　；参考：《中国工程机械学报》2017年04期

【摘要】：针对超声检测信号中结构噪声难以去除的问题,提出了一种变分模态分解(Variational Mode Decomposition,VMD)和小波能量熵阈值(Wavelet Energy Entropy Threshold,WEET)联合降噪的算法.分析了含噪系统熵增的特性以及结构噪声在不同时间段的分布特征,提出了用小波能量熵表征信号的含噪状态,并以小波能量熵最大子区间的小波系数参与计算各个尺度层的阈值.对仿真及实测信号进行处理,结果表明,该方法(VMD-WEET)能很好地抑制超声回波信号中存在的白噪声及结构噪声,还原了准确的波形特征,验证了其有效性.
[Abstract]:In order to solve the problem that structural noise is difficult to be removed in ultrasonic signal detection, a new algorithm for noise reduction is proposed, which is based on variational Mode decomposition (VMD) and wavelet Energy Entropy Energy Entropy decomposition (Wet) combined with wavelet energy entropy threshold. The characteristics of entropy increase in noisy system and the distribution of structural noise in different time periods are analyzed. The wavelet energy entropy is used to characterize the noisy state of the signal. The wavelet coefficients of the maximum sub-interval of wavelet energy entropy are used to calculate the threshold of each scale layer. The simulated and measured signals are processed. The results show that the proposed method can effectively suppress the white noise and structural noise in the ultrasonic echo signal, restore the accurate waveform features, and verify its effectiveness.
【作者单位】：华北电力大学能源动力与机械工程学院;中国航发北京航科发动机控制系统科技有限公司;
【基金】：中央高校基本科研业务费资助项目(2014MS118)
【分类号】：TG115.285

【相似文献】