Hilbert-Huang变换去除可控震源谐波畸变

Removal of harmonic distortions in vibroseis data using the Hilbert-Huang transformation

可控震源地震勘探采集信号中常常带有谐波畸变,去掉高阶谐波而又保留有用信号对提高地震勘探分辨率具有重要意义。采用Hilbert-Huang方法去除可控震源采集信号谐波畸变。对可控震源信号进行经验模态分解(EMD)和Hilbert变换,得到Hilbert-Huang谱,在时频域识别并去除高阶谐波,再用Hilbert反变换得到去除高阶谐波后的本征模态函数(IMF),叠加后可得到不含高阶谐波的信号,而保留与信号发射频率一致的基波。去掉高阶谐波后再互相关,互相关中高阶谐波也相应去除,相关函数主瓣更突出,旁瓣减少,提高了计算延时和计算波速的准确性。

Vibroseis seismic signals often include harmonic distortions.The higher order harmonics need to be removed while maintaining the desired signal to improve the seismic resolution.The Hilbert-Huang method can be used to remove harmonic distortions from vibroseis signals.The vibroseis signals are processed using empirical mode decomposition（EMD） and Hilbert transforms to get the Hilbert-Huang spectrum.After recognizing and removing the higher order harmonics in the time-frequency domain,and an inverse Hilbert transform is used to get the intrinsic mode functions（IMF） without harmonics.Then addiction is used to obtain the signals without higher order in a fundamental wave with a frequency consistent with the original signal.A cross-correlation after removal of the higher order harmonics shows that the higher order harmonics in cross correlation are correspondingly removed,so that the main-lobe of the correlation function is prominent and the side-lobe is small,so this signal can be used to more accurate calculate the time delay and velocity.