Validierung numerischer Methoden mittels vergleichender Analyse kanonischer Probleme der Elektromagnetik

Project Title
Validation of Numerical Methods Using a Comparative Analysis of Canonical Electromagnetic Problems
Principal Investigator
Project Abstract
Nowadays 3D electromagnetic (EM) simulators based on different numerical techniques such as the finite element method (FEM), the method of moments (MoM), the finite-difference time- domain technique (FDTD), the transmission-line matrix method (TLM), or the finite integration technique (FIT) are available for the analysis of a wide variety of electromagnetic problems. However it is well known that the results provided by the codes depend on many factors such as a suitable discretization of the structure under investigation or a careful positioning of absorbing boundary conditions. In this context each 3D EM simulator has its own requirements concerning the boundary size, the port type, or the shape of mesh elements. All input quantities have to be specified carefully in order to obtain accurate results. In practice it is difficult to satisfy all conditions and, hence, the simulated results could be “inaccurate”. In this case the questions arises how to validate them.

Based on the IEEE standard 1597 “Standard for Validation of Computational Electromagnetic Computer Modeling and
Guest student from Ajou Universität, Süd Korea

Simulation” various examples are investigated in this project. The goal is to find out how the different methods compare to each other when applied to different types of structures. Canonical validation problems such as dipole antenna, a loop antenna, a thick monopole antenna on a finite plate, and a rectangular cavity with apertures are considered using two commercial programs (based on FEM and FIT) and the code CONCEPT-II developed by the Institute of Electrromagnetic Theory (based on MOM). This project investigates the conditions necessary to obtain accurate results for all mentioned cases.