Conference Agenda

Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).

 
 
Session Overview
Session
OB1: Multi-scale modelling and homogenization
Time:
Wednesday, 24/May/2023:
9:00am - 10:40am

Session Chair: Prof. Arnulf Kost, Technische Universität Berlin, Germany
Session Chair: Prof. Anouar Belahcen, Aalto University, Finland

Presentations
9:00am - 9:20am
ID: 462 / OB1: 1
Topics: Static and Quasi-Static Fields, Multi-Scale Modelling and Homogenization
Keywords: Eddy currents, foil windings, homogenization

Homogenization of Foil Windings with Globally Supported Polynomial Shape Functions

Jonas Bundschuh1,2, Yvonne Späck-Leigsnering1,2, Herbert De Gersem1,2

1Institut für Teilchenbeschleunigung und Elektromagnetische Felder (TEMF), TU Darmstadt, 64293 Darmstadt, Germany; 2Graduate School of Excellence Computational Engineering, 64293 Darmstadt, Germany

The large number of foils in foil windings and their small dimensions require many mesh elements in conventional finite element simulations. This renders models quickly computationally infeasible. This paper formulates a homogenization approach for foil windings by approximating the voltage distribution in the foil winding domain with globally supported polynomials. This way, the small-scale structure in the foil winding domain is not resolved by the finite element mesh. The method is validated successfully for a pot inductor example.

OB1-1-462.pdf


9:20am - 9:40am
ID: 460 / OB1: 2
Topics: Numerical Techniques, Multi-Scale Modelling and Homogenization
Keywords: Finite element analysis, homogenization, multi-turn coil, parasitic capacitance

Homogenized Finite Element Analysis for Multi-Turn Coils Considering Parasitic Capacitances

Yuki Sato1, Daisuke Ishii1, Hirokazu Matsumoto1, Hajime Igarashi2

1Aoyama Gakuin University, Japan; 2Hokkaido University, Japan

This paper presents a novel homogenized finite element method that can consider parasitic capacitances of the multi-turn coils. In this method, the multi-resonant peaks can be identified only by solving the magneto-static FE analysis coupled with the resonant circuit. The proposed method is applied to the multi-turn coil model so that the impedance characteristics including the multi-resonant peaks can be represented. Moreover, the computing cost is shown to be much lower than that of the conventional full-wave finite element analysis.

OB1-2-460.pdf


9:40am - 10:00am
ID: 302 / OB1: 3
Topics: Numerical Techniques, Multi-Scale Modelling and Homogenization
Keywords: domain decomposition, model order reduction, soft magnetic materials

High Frequency Modeling of Soft Magnetic Materials with Local Model Order Reduction

Antero Marjamäki, Rainer Schneckenleitner, Reda Elkhadrawy, Paavo Rasilo

Tampere University, Finland

A full-wave electromagnetic field equation in a cross-section of a ferrite core, modelled as an array of grains, is solved with finite-element method utilizing local model order reduction techniques. With traditional finite-element method and high number of grains the resulting system becomes prohibitively resource intensive. The reduced approach greatly decreases the amount of degrees of freedom and computational time in the resulting system.

OB1-3-302.pdf


10:00am - 10:20am
ID: 204 / OB1: 4
Topics: Static and Quasi-Static Fields, Numerical Techniques, Multi-Scale Modelling and Homogenization
Keywords: multiscale finite element method MSFEM, eddy currents, error estimator

Error Estimation for Higher Order MSFEM for the Eddy Current Problem in Laminated Materials

Markus Schöbinger, Karl Hollaus

TU Wien, Austria

The multiscale finite element method has been shown to be able to solve the eddy current problem in laminated materials in a computationally cheap way. Instead of modelling each iron sheet in the finite element mesh, the whole material is treated as a bulk medium by the mesh. The local behavior of the solution is preserved by the use of suitable polynomial micro-shape functions. Multiscale approaches have been developed and shown to work for 2D, 3D and also 2D/1D methods. Both the A-formulation and the T-formulation can be used in the multiscale setting. However, up to now, error estimators have only been developed for multiscale approaches of lower dimensionality and low order. This work introduces an error estimatior for the full three dimensional eddy current problem using both formulations and higher order multiscale approaches. Due to its underlying theory it provides a reliably upper bound for the error without introducing any generic constants. The estimator is further shown to correctly estimate the error also in a local sense, which allows for its use in adaptive mesh refinement.

OB1-4-204.pdf


10:20am - 10:40am
ID: 311 / OB1: 5
Topics: Wave Propagation
Keywords: beamforming, electromagnetic propagation, random media, scattering invariant mode (SIM), statistical analysis

Investigation of Scattering Invariant Modes for Electromagnetic Wave Propagation in Random Vegetation Models

Olivér Csernyava1, József Pávó1, Zsolt Badics1,2

1Budapest University of Technology and Economics, Hungary; 2Tensor Research, LLC, USA

Scattering invariant modes are investigated in random media for the microwave regime, resulting in low-loss propagation. A 2D model of an agroforest, constructed of long cylinders, is utilized for radio link or radar use cases in rural areas. The robustness of the constructed scattering invariant mode is quantified by the invariance to the properties of the forest (material parameter, trunk position, and diameter). The statistical characterization method of Polynomial Chaos Expansion is used, providing sensitivity to the forest parameters. In this work, the scattering invariant mode is calculated and validated by analytic and numerical methods. As a novelty, this paper uses an application example to perform a feasibility study of the scattering invariant modes and their sensitivity analysis on the stochastic forest parameters.

OB1-5-311.pdf