A method of realizing pml in dgtd by using mixed triangular prism-tetrahedral mesh

Abstract

The invention discloses a method for realizing PML in DGTD by using a mixed triangular prism-tetrahedron grid, comprising the following steps: S1, selecting a calculation model, designing a PML absorbing layer, and respectively establishing the PML area and non-PML area in the calculation domain Maxwell's equations; S2, perform grid division to ensure that the grids on the primary and secondary surfaces of the periodic boundary match; S3, expand the grids of the electric field intensity vector and magnetic field intensity vector in different regions using different basis functions; S4, Use the DGTD algorithm to perform space discretization first and then time discretization to obtain its time discretization equation; S5, integrate the unit matrix, and obtain the required unit matrix according to the conditions; S6, perform time iteration to obtain the value of the electric field and magnetic field at each moment. The invention divides the grid in the PML area into triangular prisms to control the calculation accuracy from two dimensions, has good adaptability to the layered characteristics of the PML, and has better solution accuracy and faster solution speed.

Description

technical field [0001] The invention belongs to the technical field of three-dimensional electromagnetic field numerical solution, in particular to a method for realizing PML in DGTD by using a mixed triangular prism-tetrahedron grid. Background technique [0002] With the development of computational electromagnetism, methods such as finite element and finite difference have been widely used to deal with related problems. However, due to the limitations of finite difference and finite element method itself, the time-domain discontinuous Galerkin method (DGTD) was proposed abroad. Deals with some problems in computational electromagnetism. Compared with finite elements and finite differences, DGTD can be applied to more complex scenarios such as complex electromagnetic environments, large-scale multi-scale problems, etc., and DGTD can well support parallel computing. Based on this, DGTD is used to deal with computational electromagnetics. The problem will greatly speed up t...

AI Technical Summary

Problems solved by technology

However, the design of the perfect matching layer is complex and the calculation is cumbersome. At the same time, the mesh division of the PML area and the selection of the area size have a great impact on the accuracy and calculation amount.

[0004] In order to make the PML area have a better absorption effect, the current method is generally to encrypt the grid, but with the grid encryption, the calculation amount will increase sharply. For the current tetrahedral grid, although it can be applied to any different The structure of regular shape, but because of its own too many unknowns will make the calculation amount larger than the case of regular grid division, it cannot efficiently and accurately solve the PML truncation problem in computational electromagnetics

Images