基于CEEMDAN的塌落触地振动信号最优降噪光滑模型算法

Optimal Denoising Smooth Model Algorithm for Blasting Collapse Ground Vibration Signal based on CEEMDAN

为了更精确的提取原始爆破振动信号特征信息,针对EMD分解产生的模态混叠及EEMD分解完备性较差且计算复杂问题,采用CEEMDAN分解的固有模态分量构建不同的带通滤波器,建立基于CEEMDAN的爆破振动信号最优光滑降噪算法,并且根据降噪均方误差指标、曲线曲折度指标建立最优光滑降噪算法的判断准则。通过构建仿真信号并以含噪信号与降噪误差比为指标分别对EMD、EEMD及CEEMDAN的最优光滑降噪算法效果进行对比分析。结果表明:CEEMDAN分解得到的算法A 4信噪比最小,验证了本文算法的科学性和有效性。将其应用于实测的水塔塌落触地振动信号进行降噪处理和时频分析,效果较好。

In order to extract the characteristic information of the original blasting vibration signal more accurately,the modal aliasing and EEMD decomposition generated by EMD decomposition are poorly completed and computationally complex.The optimal smoothing noise reduction algorithm based on CEEMDAN is established with different band-pass filters constructed using the inherent modal components of CEEMDAN decomposition,and the criterion of optimal smoothing noise reduction algorithm is established according to the mean square error index and curve tortuosity index.By constructing the simulation signal and comparing the effects of the noisy signal and the noise reduction error ratio to the optimal smoothing noise reduction algorithms of EMD,EEMD and CEEMDAN,the results show that the noise-signal ratio from algorithm A4 of CEEMDAN decomposition is the smallest,which proves the scientificity and validity of the algorithm of this paper.The application to the measured water tower collapse grounding vibration signal for noise reduction processing and time-frequency analysis shows that the method is very good.