Adaptive finite element for 3D time-domain airborne electromagnetic modeling based on hybrid posterior error estimation

被引:22
作者
Zhang, Bo [1 ]
Yin, Changchun [1 ]
Ren, Xiuyan [1 ]
Liu, Yunhe [1 ]
Qi, Yanfu [2 ]
机构
[1] Jilin Univ, Coll Geoexplorat Sci & Technol, Jilin, Jilin, Peoples R China
[2] Changan Univ, Coll Geol Engn & Geomat, Xian, Shaanxi, Peoples R China
基金
中国国家自然科学基金;
关键词
DIFFERENCE; EM; INVERSION; SIMULATION; MESH; WAVE;
D O I
10.1190/GEO2017-0544.1
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
Airborne electromagnetic (AEM) forward modeling has been extensively developed in past years. However, not much attention has been paid to the adaptive numerical algorithms for time-domain electromagnetic modeling. We have created an adaptive method that can generate an effective mesh for time-domain 3D AEM full-wave modeling using an unstructured finite-element method and a backward Euler scheme. For the estimation of the posterior error in the adaptive process, we use a hybrid technique based on the continuity of the normal current density for modeling the off-time channels, and on the continuity of the tangential magnetic field for the on-time channels. To improve the stability of the forward modeling and control the number of grids in the adaptive process, a random grid-selection technique is applied. We check the modeling accuracy of the algorithm by comparing our adaptive results with the semianalytical solution for a time-domain AEM system over a homogeneous half-space. Furthermore, we test the effectiveness of our algorithm for multiple-source time-domain AEM systems by analyzing the meshes generated by the adaptive method and the model results. Finally, we study the topographic effect by calculating time-domain AEM responses over a hill model with an abnormal body embedded.
引用
收藏
页码:WB71 / WB79
页数:9
相关论文
共 38 条
[1]  
Amestoy PR, 2001, LECT NOTES COMPUT SC, V1947, P121
[2]  
[Anonymous], 2014, FINITE ELEMENT METHO
[3]   Airborne electromagnetic mapping for buried quaternary sands and gravels in northeast British Columbia, Canada [J].
Best, ME ;
Levson, VM ;
Ferbey, T ;
McConnell, D .
JOURNAL OF ENVIRONMENTAL AND ENGINEERING GEOPHYSICS, 2006, 11 (01) :17-26
[4]   Fast 3-D simulation of transient electromagnetic fields by model reduction in the frequency domain using Krylov subspace projection [J].
Boerner, Ralph-Uwe ;
Ernst, Oliver G. ;
Spitzer, Klaus .
GEOPHYSICAL JOURNAL INTERNATIONAL, 2008, 173 (03) :766-780
[5]   A parallel finite-difference approach for 3D transient electromagnetic modeling with galvanic sources [J].
Commer, M ;
Newman, G .
GEOPHYSICS, 2004, 69 (05) :1192-1202
[6]   An efficient vector finite element method for nonlinear electromagnetic modeling [J].
Fisher, A. ;
White, D. ;
Rodrigue, G. .
JOURNAL OF COMPUTATIONAL PHYSICS, 2007, 225 (02) :1331-1346
[7]   FINITE-DIFFERENCE CALCULATIONS OF THE TRANSIENT FIELD OF AN AXIALLY-SYMMETRIC EARTH FOR VERTICAL MAGNETIC DIPOLE EXCITATION [J].
GOLDMAN, MM ;
STOYER, CH .
GEOPHYSICS, 1983, 48 (07) :953-963
[8]   Large-scale 3D geoelectromagnetic modeling using parallel adaptive high-order finite element method [J].
Grayver, Alexander V. ;
Kolev, Tzanio V. .
GEOPHYSICS, 2015, 80 (06) :E277-E291
[9]   A finite-volume solution to the geophysical electromagnetic forward problem using unstructured grids [J].
Jahandari, Hormoz ;
Farquharson, Colin G. .
GEOPHYSICS, 2014, 79 (06) :E287-E302
[10]   Adaptive finite-element modeling using unstructured grids: The 2D magnetotelluric example [J].
Key, Kerry ;
Weiss, Chester .
GEOPHYSICS, 2006, 71 (06) :G291-G299