半航空瞬变电磁数据小波降噪方法研究

Research on wavelet denoising method for semi-airborne transient electromagnetic data

半航空瞬变电磁法野外勘探时,实测信号易受外界电磁噪声影响,导致后续反演效果不佳.常规降噪方法在处理非平稳性和相关性问题有一定局限性,因此,本文研究优化了Birge-Massart中经验系数α,提出一种基于Birge-Massart阈值策略的小波降噪法.通过正演生成理论含噪数据加入不同噪声,经本文方法降噪后信噪比(SNR)与均方根误差(RMSE)分别提升38%和降低75%,与其他阈值降噪方法相比,SNR与RMSE分别提升51%和降低59%.野外数据经本方法降噪后,反演效果准确解释了实地情况.说明此方法能灵活地分离噪声,提升后期反演和解释的精确性,是一种有效实用的半航空瞬变电磁数据降噪方法.

During the flight detection process of the semi-airborne transient electromagnetic method, the measured signal is susceptible to the influence of external electromagnetic noise, which causes the subsequent inversion effect to be poor. Conventional noise reduction methods have certain limitations in dealing with non-stationary and correlation problems. Therefore, this paper studies and optimizes the empirical coefficients in Birge-Massart, and proposes a wavelet noise reduction method based on the Birge-Massart threshold strategy. The theoretical noisy data is generated by forward modeling and different noises are added. The Signal-to-Noise Ratio(SNR) and Root Mean Square Error(RMSE) are increased by 38% and 75% respectively after denoising by the method in this paper. Compared with other threshold noise reduction methods, SNR and RMSE were increased by 51% and decreased by 59%, respectively. After the field data is denoised by this method, the inversion effect accurately explains the field situation. It shows that this method can flexibly separate the noise and improve the accuracy of later inversion and interpretation. It is an effective and practical method for noise reduction of semi-aeronautical transient electromagnetic data.