基于褶积模型的地球物理反演模型增强

地球物理数据在采集和处理过程中,由于存在噪声、模型误差、以及数据离散化误差等系统误差,导致了异常体边界模糊和模型分辨率降低等一些不可避免的不良系统退化效应的产生.本文提出了一种新的地球物理反演模型增强方法,通过消除反演估计模型中的系统误差,压制模型中的不良系统退化效应,增强反演模型的分辨率.文章从理论上分析了数据中存在的系统误差对模型求解的影响,提出了一个新的系统误差褶积退化模型,并根据该模型提出了一种基于混合范数总变分正则化的盲反褶积模型增强算法.最后,文章通过1D线性反演增强试验和2D大地电磁反演增强试验,验证了所提出的地球物理系统退化模型的正确性,以及盲反褶积增强算法的有效性.试验结果表明,方法可以有效地提高反演参数模型的分辨率.

The geophysical model enhancement based on the convolution model

Systematic errors, such as noise, errors in forward and inversion models, and discretization error are inevitably present in acquisition and processing processes of geophysical data. They induce systematic degradation, such as blurring and resolution decreasing, to the geophysical models. This article proposes a new enhancement method for geophysical inversion models. Through reducing the systematic error, the method removes the blurring degradations and increases the model resolution. We analyze the influence of systematic errors on the model estimating process, and a new systematic error convolution degradation model is proposed. A mix-norm TV regularization deconvolution algorithm is also proposed to enhance the inversion model according to the proposed convolution model. Finally, we use the 1D linear inversion and 2D magnetotelluric inversion enhancement experiments to prove the correctness of the theoretical basis and the effectiveness of the proposed blind deconvolution enhancement algorithm. The results show that the proposed algorithms can effectively enhance the resolution of inversion models.