3D MT Anisotropic Inversion Based on Unstructured Finite-element Method

2021 ◽  
Vol 26 (1) ◽  
pp. 49-60
Author(s):  
Xiaoyue Cao ◽  
Xin Huang ◽  
Changchun Yin ◽  
Liangjun Yan ◽  
Bo Zhang
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Hessian Matrix ◽  
Synthetic Data ◽  
Forward Modeling ◽  
Explicit Calculation ◽  
Computational Time ◽  
Earth Model ◽  
Earth Structures ◽  
Element Method

The conventional 3D magnetotelluric (MT) forward modeling and inversions generally assume an isotropic earth model. However, wrong results can be obtained when using an isotropic model to interpret the data influenced by the anisotropy. To effectively model and recover the earth structures including anisotropy, we develop a 3D MT inversion framework for a triaxial anisotropic model. We use the unstructured finite-element method for our forward modeling. This offers more possibility to simulate more complex underground geology and topography. To solve the inverse modeling problem, we use a limited-memory quasi-Newton algorithm (L-BFGS) with a parallel direct solver for optimization that avoids the explicit calculation of the Hessian matrix and saves the memory and computational time. We validate our algorithm via numerical experiments on both synthetic data and MT survey data from the US Array project.

Finite element method for analysis of frozen earth structures

1987 ◽  
Vol 13 (2) ◽  
pp. 121-129
Author(s):  
Sweanum Soo ◽  
Robert K. Wen ◽  
Orlando B. Andersland
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Earth Structures ◽  
Element Method

An efficient iterative algorithm for the natural convection equations based on finite element method

2018 ◽  
Vol 28 (3) ◽  
pp. 584-605 ◽  
Author(s):  
Lei Wang ◽  
Jian Li ◽  
Pengzhan Huang
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Natural Convection ◽  
Iterative Method ◽  
Error Correction ◽  
Computational Time ◽  
Content Type ◽  
Natural Convection Problem ◽  
New Iterative Method ◽  
Element Method

Purpose This paper aims to propose a new highly efficient iterative method based on classical Oseen iteration for the natural convection equations. Design/methodology/approach First, the authors solve the problem by the Oseen iterative scheme based on finite element method, then use the error correction strategy to control the error arising. Findings The new iterative method not only retains the advantage of the Oseen scheme but also saves computational time and iterative step for solving the considered problem. Originality/value In this work, the authors introduce a new iterative method to solve the natural convection equations. The new algorithm consists of the Oseen scheme and the error correction which can control the errors from the iterative step arising for solving the nonlinear problem. Comparing with the classical iterative method, the new scheme requires less iterations and is also capable of solving the natural convection problem at higher Rayleigh number.

scholarly journals Computational Finite Element Method (FEM) forward modeling workflow for transcranial Direct Current Stimulation (tDCS) current flow on MRI-derived head: Simpleware and COMSOL Multiphysics tutorial

2019 ◽  
Author(s):  
Ole Seibt ◽  
Dennis Truong ◽  
Niranjan Khadka ◽  
Yu Huang ◽  
Marom Bikson
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Direct Current ◽  
Transcranial Direct Current Stimulation ◽  
Computational Models ◽  
Current Flow ◽  
Forward Modeling ◽  
Fem Model ◽  
Direct Current Stimulation ◽  
Element Method

AbstractTranscranial Direct Current Stimulation (tDCS) dose designs are often based on computational Finite Element Method (FEM) forward modeling studies. These FEM models educate researchers about the resulting current flow (intensity and pattern) and so the resulting neurophysiological and behavioral changes based on tDCS dose (mA), resistivity of head tissues (e.g. skin, skull, CSF, brain), and head anatomy. Moreover, model support optimization of montage to target specific brain regions. Computational models are thus an ancillary tool used to inform the design, set-up and programming of tDCS devices, and investigate the role of parameters such as electrode assembly, current directionality, and polarity of tDCS in optimizing therapeutic interventions. Computational FEM modeling pipeline of tDCS initiates with segmentation of an exemplary magnetic resonance imaging (MRI) scan of a template head into multiple tissue compartments to develop a higher resolution (< 1 mm) MRI derived FEM model using Simpleware ScanIP. Next, electrode assembly (anode and cathode of variant dimension) is positioned over the brain target and meshed at different mesh densities. Finally, a volumetric mesh of the head with electrodes is imported in COMSOL and a quasistatic approximation (stead-state solution method) is implemented with boundary conditions such as inward normal current density (anode), ground (cathode), and electrically insulating remaining boundaries. A successfully solved FEM model is used to visualize the model prediction via different plots (streamlines, volume plot, arrow plot).

Evaluation of Relative Permeability Models in CO2/Brine System Using Mixed and Hybrid Finite Element Method

2016 ◽  
Vol 819 ◽  
pp. 401-405
Author(s):  
J.S. Pau ◽  
William K.S. Pao ◽  
Suet Peng Yong ◽  
Paras Qadir Memon
Keyword(s):  
Finite Element Method ◽  
Finite Element ◽  
Relative Permeability ◽  
Carbon Capture ◽  
Carbon Capture And Storage ◽  
Computational Time ◽  
Permeability Model ◽  
Hybrid Finite Element Method ◽  
Hybrid Finite Element ◽  
Element Method

The requirement to reduce 40% carbon emission in 2020 has lead Malaysia to adopt the carbon capture and storage (CCS) technology in 2009. In this research, the pressure and transport differential equation for CO2 – brine phases flow is discretized using mixed and hybrid finite element method (MHFEM) which ensures the local continuity of the finite elements. Result shows that CO2 flow radially outward from the injection well. Three relative permeability models are investigated and it was find out that the simplified relative permeability model (SRM) has reduced the computational time by 8.3 times (when compare to Brooks and Corey model) but it is accurate for 1 year preliminary prediction. For longer period of prediction, classical Brooks and Corey and van Genuchten models shall be used.

Sign in / Sign up

Export Citation Format

Share Document