Use of state circuit solver and model order reduction for solving inductive and capacitive unstructured PEEC formulation - CNRS - Centre national de la recherche scientifique Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Use of state circuit solver and model order reduction for solving inductive and capacitive unstructured PEEC formulation

Kouceila Alkama
  • Fonction : Auteur
  • PersonId : 1125084

Résumé

Computational strategies improvements for the inductive and capacitive unstructured PEEC formulation are presented in order to address efficiently a large frequency range of electromagnetic problems. Good accuracy on results is ensured thanks to the use of an adaptive Gauss integration procedure while multi-threaded Adpatative Multi Level Fast Multipole Method (AMLFMM) matrix compression algorithm allows to speed-up far interactions computation. This article focuses on the following two points: on the one hand, allowing good convergence for iterative linear systems at any frequency by the choice of an efficient preconditioner coupled with a suitable PEEC circuit solver and, on the other hand, the use of Model Order Reduction (MOR) techniques for solving multi frequency problems, allowing fast computing time. An example illustrates the performances of these approaches.
Fichier principal
Vignette du fichier
Compumag_2023_GM_FinalUpload.pdf (800.96 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04115619 , version 1 (02-06-2023)

Identifiants

  • HAL Id : hal-04115619 , version 1

Citer

Gérard Meunier, Jean-Michel Guichon, Kouceila Alkama, Olivier Chadebec. Use of state circuit solver and model order reduction for solving inductive and capacitive unstructured PEEC formulation. 24th International Conference on the Computation of Electromagnetic Fields (Compumag'2023), May 2023, Kyoto, Japan. ⟨hal-04115619⟩
23 Consultations
31 Téléchargements

Partager

Gmail Facebook X LinkedIn More