基于重加权L1的ATpV正则化叠前反演方法OA北大核心CSTPCD

Seismic prestack inversion can accurately obtain various parameters of underground reservoir media,and is one of the important technologies in oil and gas exploration and development.Due to unknown natural fators,seismic inversion is a typical ill-conditioned problem,and a regularization constraint objective function is usually used to alleviate the ill-conditioned inversion problem.However,the regularization constraint ignores the amplitude information of the stratigraphic boundary,and the reweighting method can overcome this problem well and restore the sparsity better.A pre-stack three-parameter in-version method for ATpV regularization based on reweighted L1(ATpV-L1 method)is proposed.The reweighted L1 method is combined with ATpV regularization for the first time and introduced into the pre-stack inversion.The Alternating direction multiplier algorithm(ADMM)is used to establish the inversion framework,and the objective function is optimized in blocks,which effectively improves the convergence speed.The manuscript first introduces the ATpV-L1 method,and establishes the pre-stack inversion objective function based on ATpV-L1.Then the theoretical model data is used to compare the inversion results of the new method and the ATpV regularization method,and the effect of this method is verified.Finally,the actual data is used for experimental analysis,which further verifies the inversion accuracy of the new method and verifies the feasibility of the method.The experimental results show that the proposed method can effectively recover the sparsity of the inversion results and improve the inversion accuracy.