The magnetotelluric (MT) method has been widely used in geophysical electromagnetic (EM) exploration. However, complex geometries and anisotropic structures still pose challenges for the simulations of 3D MT problems. This study presents a nodal finite-element (FE) solution to simulate 3D MT responses in 3D conductivity structures with general anisotropy. The method is based on the \({\mathbf{A}} - \psi\) decomposition of the electric field. The computational domain is discretized into hexahedral elements. The linear system equations that result from the FE discretization are solved by iterative solvers. We designed three examples to test the performance of the algorithm in this study. For the first example, we compare our results with those of other scholars to validate the effectiveness of the procedure. For the second example, the convergence behaviors of different iterative solvers with different preconditioners are tested. For the third example, a complex model is designed to demonstrate the robustness and effectiveness of the proposed code. Numerical experiments show that the convergence rate of the iterative solver of the \({\mathbf{A}} - \psi\) method is very fast, especially at low frequencies.

大地电磁（MT）方法已广泛应用于地球物理电磁（EM）勘探。然而，复杂的几何形状和各向异性结构仍然对 3D MT 问题的模拟构成挑战。本研究提出了一种节点有限元 (FE) 解决方案，用于模拟具有一般各向异性的 3D 电导率结构中的 3D MT 响应。该方法基于\({\mathbf{A}} - \psi\)电场分解。计算域被离散化为六面体元素。由 FE 离散化产生的线性系统方程由迭代求解器求解。我们设计了三个例子来测试本研究中算法的性能。对于第一个例子，我们将我们的结果与其他学者的结果进行比较，以验证该程序的有效性。对于第二个示例，测试了具有不同预处理器的不同迭代求解器的收敛行为。对于第三个示例，设计了一个复杂的模型来证明所提出代码的稳健性和有效性。数值实验表明\({\mathbf{A}} - \psi\)方法的迭代求解器的收敛速度非常快，尤其是在低频时。