1. FFT-PEEC: A Fast Tool From CAD to Power Electronics Simulations
- Author
-
Schanen, J.-L, Torchio, Riccardo, Lucchini, Francesco, Schanen, Jean-Luc, Chadebec, Olivier, Meunier, Gerard, Laboratoire de Génie Electrique de Grenoble (G2ELab ), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA), and Universita degli Studi di Padova
- Subjects
Electromagnetics ,Discretization ,Computer science ,Fast Fourier transform ,Capacitive Effects ,Computational modeling ,FFT ,Geometry ,Magnetic resonance imaging ,Magnetic separation ,Non-Linear Materials ,PEEC ,Solid modeling ,Tools ,02 engineering and technology ,Computational science ,Power electronics ,0202 electrical engineering, electronic engineering, information engineering ,Electrical and Electronic Engineering ,Partial element equivalent circuit ,020208 electrical & electronic engineering ,[SPI.NRJ]Engineering Sciences [physics]/Electric power ,Solver ,Data compression - Abstract
International audience; A fast and general Partial Element Equivalent Circuit (PEEC) method based on the Fast-Fourier-Transform (FFT) is proposed for the first time. The numerical tool only requires common CAD data input files (e.g. .stl format), then the discretization process is performed automatically by means of a fast voxelization technique based on ray intersection, thus drastically reducing the human effort required to setup the model. The method allows for considering at the same time inductive and capacitive effects, and is focused on power electronics applications where propagation effects can be neglected, whereas all the other electromagnetic phenomena are considered. Specifically, the proposed method is particularly suited for problems where both electric and magnetic fields are equally important and therefore quasistatic approximations do not apply. An ad-hoc preconditioner which significantly speeds-up the solver is also proposed and, thanks to the FFT, both memory and computation time are significantly reduced, without the need of applying data compression. Both linear and non-linear materials are considered by the proposed FFT-PEEC method. Sample implementation of the method is made publicly available.
- Published
- 2022