Magnetic field gradient tensor technique provides abundant data for delicate inversion of subsurface magnetic susceptibility distribution. Large scale magnetic data inversion imaging requires high speed and accuracy f...Magnetic field gradient tensor technique provides abundant data for delicate inversion of subsurface magnetic susceptibility distribution. Large scale magnetic data inversion imaging requires high speed and accuracy for forward modeling. For arbitrarily distributed susceptibility data on an undulated surface, we propose a fast 3D forward modeling method in the wavenumber domain based on(1) the wavenumber-domain expression of the prism combination model and the Gauss–FFT algorithm and(2) cubic spline interpolation. We apply the proposed 3D forward modeling method to synthetic data and use weighting coefficients in the wavenumber domain to improve the modeling for multiple observation surfaces, and also demonstrate the accuracy and efficiency of the proposed method.展开更多
A fast and high precision spatial domain algorithm is presented for forward modeling of two-dimensional(2D)body gravity anomalies of arbitrary shape and density distribution.The new algorithm takes advantage of the co...A fast and high precision spatial domain algorithm is presented for forward modeling of two-dimensional(2D)body gravity anomalies of arbitrary shape and density distribution.The new algorithm takes advantage of the convolution properties of the expression for 2D gravity anomalies,uses a rectangular cell as a grid subdivision unit,and then 2D bodies with irregular cross-sections are approximated by a combination of 2D bodies with a rectangular cross section.The closed-form expression is used to calculate the gravitational anomalies of the combination of 2D bodies with a rectangular cross section.To improve computing effi ciency,the new algorithm uses a fast algorithm for the implementation of the Toeplitz matrix and vector multiplication.The synthetic 2D models with rectangular and circular cross-sections and constant and variable densities are designed to evaluate the computational accuracy and speed of the new algorithm.The experiment results show that the computation costs less than 6 s for a grid subdivision with 10000×10000 elements.Compared to the traditional forward modeling methods,the proposed method significantly improved computational effi ciency while guaranteeing computational accuracy.展开更多
Fully recognize various problems in the observed magnetotelluric(MT)data is the precondition of inverse solutions.In the paper,according to the geomorphological conditions of the observational MT stations in the Guang...Fully recognize various problems in the observed magnetotelluric(MT)data is the precondition of inverse solutions.In the paper,according to the geomorphological conditions of the observational MT stations in the Guangxi area,we constructed several different kinds of models to conduct a three-dimensional forward simulation of the MT fi eld using the vector fi nite element method(FEM).First,the variation rule and differences of apparent resistivityρxy andρyx in the xy and yx modes were studied and analyzed,and then the geoelectric information refl ected by the change of apparent resistivityρxx andρyy were discussed.Final,the responses of typical geological structures that cause a static shift problem were presented.The synthetic examples showed thatρxy andρyx were relevant to the layout of the survey line,for instance,ρxy had different values along the west-east profi le compared with that of the southnorth profi le,Moreover,ρxx andρyy could subtly show the abnormal body-host rock interface,which could be used to restrict the anomalous domain in the inversion process.In addition to the scale and depth of the top surface of the anomalous body,the widespread rivers and hills,can simulate static shift.Hence,to reduce the infl uence of static shift on MT data,a reasonable distance between a station and rivers or hills should be considered in accordance with the scale of rivers or hills.展开更多
基金Project(2021JJ30577) supported by Hunan Provincial Natural Science Foundation of ChinaProject(X202110555425) supported by College Students Innovation and Entrepreneurship Training Program of University of South China。
基金supported by the National Special Plan for the 13th Five-Year Plan of China(No.2017YFC0602204-10)Independent Exploration of the Innovation Project for Graduate Students at Central South University(No.2017zzts176)+3 种基金National Natural Science Foundation of China(Nos.41574127,41404106,and 41674075)Postdoctoral Fund Projects of China(No.2017M622608)National Key R&D Program of China(No.2018YFC0603602)Natural Science Youth Fund Project of the Hunan Province,China(No.2018JJ3642)
文摘Magnetic field gradient tensor technique provides abundant data for delicate inversion of subsurface magnetic susceptibility distribution. Large scale magnetic data inversion imaging requires high speed and accuracy for forward modeling. For arbitrarily distributed susceptibility data on an undulated surface, we propose a fast 3D forward modeling method in the wavenumber domain based on(1) the wavenumber-domain expression of the prism combination model and the Gauss–FFT algorithm and(2) cubic spline interpolation. We apply the proposed 3D forward modeling method to synthetic data and use weighting coefficients in the wavenumber domain to improve the modeling for multiple observation surfaces, and also demonstrate the accuracy and efficiency of the proposed method.
基金This work is jointly sponsored by the National Natural Science Foundation of China(No.41404106)the Scientific Research Startup Fund for Doctoral Program of Guilin University of Technology,Guangxi Natural Science Foundation Program(No.2018GXNSFBA138049)Guangxi Natural Science Foundation Program for Innovation Research Team(No.2016GXNSFGA380004).
文摘A fast and high precision spatial domain algorithm is presented for forward modeling of two-dimensional(2D)body gravity anomalies of arbitrary shape and density distribution.The new algorithm takes advantage of the convolution properties of the expression for 2D gravity anomalies,uses a rectangular cell as a grid subdivision unit,and then 2D bodies with irregular cross-sections are approximated by a combination of 2D bodies with a rectangular cross section.The closed-form expression is used to calculate the gravitational anomalies of the combination of 2D bodies with a rectangular cross section.To improve computing effi ciency,the new algorithm uses a fast algorithm for the implementation of the Toeplitz matrix and vector multiplication.The synthetic 2D models with rectangular and circular cross-sections and constant and variable densities are designed to evaluate the computational accuracy and speed of the new algorithm.The experiment results show that the computation costs less than 6 s for a grid subdivision with 10000×10000 elements.Compared to the traditional forward modeling methods,the proposed method significantly improved computational effi ciency while guaranteeing computational accuracy.
基金This paper was partially supported by the National Natural Science Foundation of China(No.41674075 and 41904123)the Natural Science Foundation of Guangxi Province(No.2016GXNSFGA380004)and the High Level Innovative Team and Excellent Scholar Plan of Guangxi High Education Institution.
文摘Fully recognize various problems in the observed magnetotelluric(MT)data is the precondition of inverse solutions.In the paper,according to the geomorphological conditions of the observational MT stations in the Guangxi area,we constructed several different kinds of models to conduct a three-dimensional forward simulation of the MT fi eld using the vector fi nite element method(FEM).First,the variation rule and differences of apparent resistivityρxy andρyx in the xy and yx modes were studied and analyzed,and then the geoelectric information refl ected by the change of apparent resistivityρxx andρyy were discussed.Final,the responses of typical geological structures that cause a static shift problem were presented.The synthetic examples showed thatρxy andρyx were relevant to the layout of the survey line,for instance,ρxy had different values along the west-east profi le compared with that of the southnorth profi le,Moreover,ρxx andρyy could subtly show the abnormal body-host rock interface,which could be used to restrict the anomalous domain in the inversion process.In addition to the scale and depth of the top surface of the anomalous body,the widespread rivers and hills,can simulate static shift.Hence,to reduce the infl uence of static shift on MT data,a reasonable distance between a station and rivers or hills should be considered in accordance with the scale of rivers or hills.
