The main problems in three-dimensional gravity inversion are the non-uniqueness of the solutions and the high computational cost of large data sets. To minimize the high computational cost, we propose a new sorting me...The main problems in three-dimensional gravity inversion are the non-uniqueness of the solutions and the high computational cost of large data sets. To minimize the high computational cost, we propose a new sorting method to reduce fluctuations and the high frequency of the sensitivity matrix prior to applying the wavelet transform. Consequently, the sparsity and compression ratio of the sensitivity matrix are improved as well as the accuracy of the forward modeling. Furthermore, memory storage requirements are reduced and the forward modeling is accelerated compared with uncompressed forward modeling. The forward modeling results suggest that the compression ratio of the sensitivity matrix can be more than 300. Furthermore, multiscale inversion based on the wavelet transform is applied to gravity inversion. By decomposing the gravity inversion into subproblems of different scales, the non-uniqueness and stability of the gravity inversion are improved as multiscale data are considered. Finally, we applied conventional focusing inversion and multiscale inversion on simulated and measured data to demonstrate the effectiveness of the proposed gravity inversion method.展开更多
Joint inversion based on a correlation constraint utilizes a linear correlation function as a structural constraint.The linear correlation function contains a denominator,which may result in a singularity as the objec...Joint inversion based on a correlation constraint utilizes a linear correlation function as a structural constraint.The linear correlation function contains a denominator,which may result in a singularity as the objective function is optimized,leading to an unstable inversion calculation.To improve the robustness of this calculation,this paper proposes a new method in which a sinusoidal correlation function is employed as the structural constraint for joint inversion instead of the conventional linear correlation function.This structural constraint does not contain a denominator,thereby preventing a singularity.Compared with the joint inversion method based on a cross-gradient constraint,the joint inversion method based on a sinusoidal correlation constraint exhibits good performance.An application to actual data demonstrates that this method can process real data.展开更多
Due to the induced polarization(IP)eff ect,the sign reversal often occurs in timedomain airborne electromagnetic(AEM)data.The inversions that do not consider IP eff ect cannot recover the true umderground electrical s...Due to the induced polarization(IP)eff ect,the sign reversal often occurs in timedomain airborne electromagnetic(AEM)data.The inversions that do not consider IP eff ect cannot recover the true umderground electrical structures.In view of the fact that there are many parameters of airborne induced polarization data in time domain,and the sensitivity diff erence between parameters is large,which brings challenges to the stability and accuracy of the inversion.In this paper,we propose an inversion mehtod for time-domain AEM data with IP effect based on the Pearson correlation constraints.This method uses the Pearson correlation coeffi cient in statistics to characterize the correlation between the resistivity and the chargeability and constructs the Pearson correlation constraints for inverting the objective function to reduce the non uniqueness of inversion.To verify the eff ectiveness of this method,we perform both Occam’s inversion and Pearson correlation constrained inversion on the synthetic data.The experiments show that the Pearson correlation constrained inverison is more accurate and stable than the Occam’s inversion.Finally,we carried out the inversion to a survey dataset with and without IP eff ect.The results show that the data misfit and the continuity of the inverted section are greatly improved when the IP eff ect is considered.展开更多
基金This work was supported by the Key National Research Project of China （Nos. 2017YFC0601900 and 2016YFC0303100） and the Key Program of National Natural Science Foundation of China （Nos. 41530320 and 41774125）.
文摘The main problems in three-dimensional gravity inversion are the non-uniqueness of the solutions and the high computational cost of large data sets. To minimize the high computational cost, we propose a new sorting method to reduce fluctuations and the high frequency of the sensitivity matrix prior to applying the wavelet transform. Consequently, the sparsity and compression ratio of the sensitivity matrix are improved as well as the accuracy of the forward modeling. Furthermore, memory storage requirements are reduced and the forward modeling is accelerated compared with uncompressed forward modeling. The forward modeling results suggest that the compression ratio of the sensitivity matrix can be more than 300. Furthermore, multiscale inversion based on the wavelet transform is applied to gravity inversion. By decomposing the gravity inversion into subproblems of different scales, the non-uniqueness and stability of the gravity inversion are improved as multiscale data are considered. Finally, we applied conventional focusing inversion and multiscale inversion on simulated and measured data to demonstrate the effectiveness of the proposed gravity inversion method.
基金supported by the National Key Research and Development Project of China(No:2017YFC0602201)
文摘Joint inversion based on a correlation constraint utilizes a linear correlation function as a structural constraint.The linear correlation function contains a denominator,which may result in a singularity as the objective function is optimized,leading to an unstable inversion calculation.To improve the robustness of this calculation,this paper proposes a new method in which a sinusoidal correlation function is employed as the structural constraint for joint inversion instead of the conventional linear correlation function.This structural constraint does not contain a denominator,thereby preventing a singularity.Compared with the joint inversion method based on a cross-gradient constraint,the joint inversion method based on a sinusoidal correlation constraint exhibits good performance.An application to actual data demonstrates that this method can process real data.
基金This paper was fi nancially supported by the National Natural Science Foundation of China(Nos.42030806,41774125,41904104,41804098)the Pioneer Project of Chinese Academy of Sciences(No.XDA14020102).
文摘Due to the induced polarization(IP)eff ect,the sign reversal often occurs in timedomain airborne electromagnetic(AEM)data.The inversions that do not consider IP eff ect cannot recover the true umderground electrical structures.In view of the fact that there are many parameters of airborne induced polarization data in time domain,and the sensitivity diff erence between parameters is large,which brings challenges to the stability and accuracy of the inversion.In this paper,we propose an inversion mehtod for time-domain AEM data with IP effect based on the Pearson correlation constraints.This method uses the Pearson correlation coeffi cient in statistics to characterize the correlation between the resistivity and the chargeability and constructs the Pearson correlation constraints for inverting the objective function to reduce the non uniqueness of inversion.To verify the eff ectiveness of this method,we perform both Occam’s inversion and Pearson correlation constrained inversion on the synthetic data.The experiments show that the Pearson correlation constrained inverison is more accurate and stable than the Occam’s inversion.Finally,we carried out the inversion to a survey dataset with and without IP eff ect.The results show that the data misfit and the continuity of the inverted section are greatly improved when the IP eff ect is considered.
