基于CEEMD的爆破振动信号自适应去噪

针对爆破振动信号短时、突变的特点,提出了基于CEEMD的爆破振动信号自适应降噪方法。为解决信噪分量分界点的判定问题,该方法提出依据CEEMD分解得到的噪声分量和有用信号分量自相关函数的波峰宽度特性自适应地判定信噪分量的分界点的方法。对IMF分量消噪阈值函数进行了改进,新阈值函数具有连续和高阶可导性。实验分析表明:采用新阈值函数的CEEMD自适应去噪方法能够有效地去除实测爆破振动信号中夹杂的噪声,并且很好地保留了爆破振动信号的突变细节,去噪效果良好。

Adaptive de-noising of Vibration Signal Based on CEEMD auto-correlation properties

Considering the characteristics of the blasting vibration signal that it is short-time and mutation, adaptive noise reduction method for the blasting vibration signal was presented based on the CEEMD. To solve the problem of the judgement of cut-off point between signal and noise component, the method put forward that signal-to-noise weight cut-off point method can be adaptively decided on the basis of the peak width features of autocorrelation function of the decomposed noise and useful signal components by CEEMD. The IMF component de-noising threshold function was improved, the new threshold function is continuous and has high order. Experimental analysis show that CEEMD adaptive de-noising method with new threshold function can effectively remove the inclusions in the measured blasting vibration signal, and well retain the blasting vibration signal mutation in detail, the denoising effect is good.