CEEMDAN 小波包联合降噪的优化方法

针对实测隧道爆破振动信号降噪效果不理想的问题，引入多尺度排列熵的概念，用来筛选含噪明显的本征模态分量，并结合SG（savitzky-golay）平滑滤波方法提出了一种完备的自适应噪声经验模态分解（CEEMDAN）和小波包联合降噪的优化方法，通过信噪比(SNR)、均方根差(RMSE)、相关系数、自相关系数对仿真实验和实测信号处理结果进行对比分析。结果表明，CEEMDAN 小波包联合降噪优化方法提高了信号降噪效果，且有效保留了原始信号中的特征信息，可以应用到类似爆破振动信号的降噪处理中。

Optimization Method of CEEMDAN-Wavelet Packet Joint Noise Reduction

Aiming at the problem that the noise reduction outcome of vibration signal in tunnel blasting is not ideal, multi-scale permutation entropy was introduced to filter out the eigenmode components with obvious noise. Combined with SG (savitzky-golay) smoothing filtering method, a complete adaptive noise empirical mode decomposition (CEEMDAN) and wavelet packet joint denoising optimization method was proposed. Results of simulation experiments and measured signal processing were compared and analyzed by signal-to-noise ratio (SNR), root mean square error (RMSE), correlation coefficient, and autocorrelation coefficient. It shows that the method improves the noise reduction outcome and effectively retains the characteristic information in the original signal. It can be applied to the noise reduction processing of similar blasting vibration signals.