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
PA-A1: Numerical techniques 1
Time:
Tuesday, 23/May/2023:
1:50pm - 3:20pm

Session Chair: Dr. Hidenori Sasaki, Hosei University, Japan

Presentations
ID: 452 / PA-A1: 1
Topics: Mathematical Modelling and Formulations, Numerical Techniques, Novel Computational Methods for Machines and Devices
Keywords: Core losses, lookup-tables, state-space model, synchronous machine

Lookup-Table Based Simulink Models for Synchronous Machines Including Core Losses

Amr A. Abbas1, Joonas Vesa1, Hao Chen2, Yujing Liu2, Paavo Rasilo1

1Tampere University, Finland; 2Chalmers University, Sweden

This paper presents a fast and accurate phase domain model for synchronous machines taking into consideration the machine geometry, material non-linearities and core losses. A state-space model is first constructed in the well-known manner of storing the solutions of multiple static finite element (FE) simulations into lookup-tables (LUTs) to express the stator flux linkages as functions of the state variables, i.e., the currents and the rotor position. A novel approach is presented for constructing a precomputed LUT for the core loss as a function of the state variables and their time derivatives, so that the loss can be directly interpolated when time-stepping the state-space model.

PA-A1-1-452.pdf


ID: 321 / PA-A1: 2
Topics: Numerical Techniques, Multi-Physics and Coupled Problems
Keywords: Eddy currents, Finite element analysis, Magnetic confinement, Magnetohydrodynamics

An Improved 3D MHD Equilibrium Formulation

Oszkar Biro1, Pietro Testoni2, Alfredo Portone2

1TU Graz, Austria; 2F4E, Barcelona, Spain

An improved fully three-dimensional finite element formulation of magnetohydrodynamic plasma equilibrium is introduced. The method takes full account of the component of the current density parallel to the magnetic flux density and explicitly includes the mechanical pressure as a variable. It is shown that a previous formulation without these properties fails. In the new approach, the reluctivity function describing the proportionality between the parallel component of the current density and the flux density is solved for using nodal finite elements.

PA-A1-2-321.pdf


ID: 142 / PA-A1: 3
Topics: Numerical Techniques
Keywords: Approximation algorithms, Integral equations, Linear systems, Numerical analysis, Electrostatic processes

Algebraic Partition Construction Method for Hierarchical Matrices

Akihiro Ida

Japan Agency for Marine-Earth Science and Technology, Japan

A hierarchical matrix (H-matrix) is an approximated form of a dense matrix derived from an integral equation with a degenerate kernel. H-matrix construction is achieved by finding a proper permutation and partition for the matrix such that most of submatrices become numerically low-rank. In this paper, an algebraic method for the H-matrix construction is proposed. The proposed method requires only values of the original dense matrix, unlike the conventional method, which requires geometrical information. We confirm through numerical experiments on electrostatic field analyses that the proposed algebraic method achieves high compressibility close to that of the conventional method using auxiliary geometrical information.

PA-A1-3-142.pdf


ID: 140 / PA-A1: 4
Topics: Numerical Techniques
Keywords: finite element method, inverse scheme, magnetic material, numerical analysis

Determination of Local Magnetic Material Properties using an Inverse Scheme

Andreas Gschwentner, Klaus Roppert, Manfred Kaltenbacher

Graz University of Technology, Austria

In this paper, we postulate a concept for determining local magnetic material properties for electrical steel sheets. The basic idea involves the development of a sensor-actuator system capable of locally magnetizing the electrical steel sheets and measuring the resulting magnetic flux density. Based on this measured data, the magnetic permeability is determined by applying an inverse scheme based on solving the magneto-static partial differential equations with the finite element (FE) method. In a first step, the measurement data is generated artificially by a reference simulation and the magnetic permeability is locally restricted to a linear isotropic behavior. First result of the investigations are presented.

PA-A1-4-140.pdf


ID: 519 / PA-A1: 5
Topics: Numerical Techniques
Keywords: Finite element analysis, Iterative methods, Linear systems, Superconducting thin films.

Fast Solver for Symmetric Linear System in Shielding Current Analysis of HTS Thin Film: Improvement of H-Matrix-Based Acceleration Technique

Ayumu Saitoh

Yamagata University, Japan

The H-matrix-based acceleration technique has been improved, and the modified acceleration technique is incorporated into the symmetric linear-system solver in the shielding current analysis of the HTS film. The results of computations show that the modified solver is slightly faster than the conventional one as long as the judgment parameter does not exceed a certain limit. From the above results, the modified solver might be a useful tool for executing the shielding current analysis.

PA-A1-5-519.pdf


ID: 131 / PA-A1: 6
Topics: Numerical Techniques
Keywords: Eddy current field, finite element analysis, magnetic hysteresis, play model, predictor-corrector scheme.

Hysteretic Finite Element Analysis Using a Semi-implicit Method Combined with the Predictor-Corrector Scheme

Miwa Tobita, Tetsuji Matsuo

Graduate School of Engineering, Kyoto University, Japan

A semi-implicit method for solving the time evolution of nonlinear equations in hysteretic finite element analysis is proposed in this study. The magnetic field is updated with a linear approximation to avoid iterations, and the differential reluctivity is corrected using the predictor-corrector scheme to improve and evaluate accuracy. The error evaluation was also investigated by observing the difference between the predictor and corrector.

PA-A1-6-131.pdf


ID: 374 / PA-A1: 7
Topics: Numerical Techniques
Keywords: finite element analysis, reduced order systems, electrical machines, equivalent circuits

Application of Reduced-Order Models to the Analysis of 3-Dimensional Electrical Machines

Amir Akbari, David Lowther

McGill University, Canada

3-D analysis is a crucial tool for designing and optimizing electrical machines, especially those with significant 3-D effects such as an axial flux permanent magnet (AFPM) machine. Building high-fidelity surrogate models of such topologies requires 3-D FEA. However, where the computational budget is limited for large or medium-scale parametric machine optimization, replacing 3-D FEA with models that confidently and quickly compute the 3-D performance can be extremely helpful. To resolve the 3-D issues, this paper attempts to build effective high-speed reduced order models with an accuracy approaching a full 3-D model. A hybrid modeling approach is presented that couples the 2½-D modeling approaches with simple and effective magnetic circuits (MEC) models incorporating the 3-D effects. The proposed approach provides a satisfactory compromise regarding computation time and accuracy.

PA-A1-7-374.pdf


ID: 384 / PA-A1: 8
Topics: Numerical Techniques
Keywords: eddy currents, finite element method

Extended Finite Element Method for Skin and Proximity Effects in Round Wires

Shingo Hiruma

Kyoto University, Japan

In this study, the extended finite element method is used to calculate eddy currents due to skin and proximity effects in round wires. The curved boundary of the round wires is expressed by the exact boundary mapping in the finite element analysis. In addition, the Bessel functions are used as enrichment functions, which are the general solutions of the radial direction of the two-dimensional (2D) Helmholtz equation in the cylindrical coordinate system. Owing to the enrichment functions, we can use a mesh without considering the skin depth at high frequencies. The method is applied to the 2D eddy current problem, and it is demonstrated that the error of the Joule losses is significantly reduced compared to the conventional method.

PA-A1-8-384.pdf


ID: 246 / PA-A1: 9
Topics: Numerical Techniques
Keywords: Differential equations, Electromagnetic waveguides, Meshless Method, Numerical analysis

Application of Meshless Local Radial Point Interpolation Method on a Two-dimensional Vector Eigenvalue Problem

URSULA DO CARMO RESENDE, MARCIO FRAGA ANDRADE

CEFET-MG, Brazil

In this work, the meshless local radial point interpolation method is applied on a 2D vector eigenvalue problem. The method is entirely nodal based, and each node is associated with a vector basis that allows direct enforcement of essential boundary conditions. Unlike traditional methods, the problem itself is described by a mixed formulation, in which, the vector wave equation and the divergence-free constraint are coupled by using a Lagrange multiplier. The complete proposed technique provides a novel approach to the solution of vector problems in computational electromagnetism. The numerical results are compared with analytical ones.

PA-A1-9-246.pdf