Search results

1 – 3 of 3
Article
Publication date: 7 September 2015

Paul Handgruber, Simon Schernthanner, Oszkár Bíró, Andrej Stermecki and Georg Ofner

The purpose of this paper is to study the effects of inverter supply on the iron loss characteristics of slip-ring induction machines. Pulse width modulated (PWM) voltage supply…

Abstract

Purpose

The purpose of this paper is to study the effects of inverter supply on the iron loss characteristics of slip-ring induction machines. Pulse width modulated (PWM) voltage supply on the stator side, as well as a doubly fed operation mode with rotor-sided inverter, are investigated.

Design/methodology/approach

An inverter fed machine model is coupled to previously developed eddy current and hysteresis loss models. The eddy current model is based on a finite element method and considers the three-dimensional (3D) eddy current distribution in the steel sheets. The hysteresis losses are computed by a static Preisach vector model.

Findings

It is found that under stator-sided inverter supply the eddy current losses do significantly increase when compared to sinusoidal feeding, contributing to a total loss increase of 10-15 percent. In doubly fed operation, the additional losses are generally lower owing to the winding topology of the studied machine.

Research limitations/implications

The analyses presented are restricted to single PWM pattern only. The influences of different PWM parameters remain to be investigated in future.

Practical implications

Regarding practical applications, the reduced additional losses in doubly fed configurations can be considered as a further advantage when competing against other topologies available.

Originality/value

The 3D eddy current model is applied for the first time to quantify the effects of inverter supply. Furthermore, the presented studies on the iron losses in doubly fed operation are original and of practical value for designers and researches.

Details

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, vol. 34 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 14 September 2010

Andrej Stermecki, Oszkár Bíró, Kurt Preis, Siegfried Rainer, Klaus Krischan and Georg Ofner

The purpose of this paper is to define a time‐efficient numerical procedure for the extraction of load‐dependent equivalent circuit (EC) parameters of induction machines. The…

Abstract

Purpose

The purpose of this paper is to define a time‐efficient numerical procedure for the extraction of load‐dependent equivalent circuit (EC) parameters of induction machines. The parameters are determined for every operating point, thus their variation due to skin effect and material saturation under arbitrary load condition is taken into consideration.

Design/methodology/approach

Two methods are presented and compared. The first one is based on the numerical simulation of the standard measurement process, yielding an EC with constant parameters. A time‐harmonic finite element analysis is applied in the second method to calculate the load‐dependent EC parameters. Material linearization and the superposition principle for the magnetic flux are employed to define the leakage inductances.

Findings

A distinct load dependence of all EC parameters has been proven as well as the clear disparity between stator and rotor leakage inductances. These effects can only be taken accurately into account by the EC obtained by the second numerical procedure proposed.

Originality/value

The presented method successfully overcomes typical problems of the measurement process and of the standard numerical procedure for EC parameter estimation, thus the obtained EC parameters are load‐dependent while the physical interpretation of the variables and parameters remains straightforward. Hence, the paper of the internal machine variables is enabled.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 29 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 25 July 2019

Stephan Willerich and Hans-Georg Herzog

The use of gradient-based methods in finite element schemes can be prevented by undefined derivatives, which are encountered when modeling hysteresis in constitutive material…

Abstract

Purpose

The use of gradient-based methods in finite element schemes can be prevented by undefined derivatives, which are encountered when modeling hysteresis in constitutive material laws. This paper aims to present a method to deal with this problem.

Design/methodology/approach

Non-smooth Newton methods provide a generalized framework for the treatment of minimization problems with undefined derivatives. Within this paper, a magnetostatic finite element formulation that includes hysteresis is presented. The non-linear equations are solved using a non-smooth Newton method.

Findings

The non-smooth Newton method shows promising convergence behavior when applied to a model problem. The numbers of iterations for magnetization curves with and without hysteresis are within the same range.

Originality/value

Mathematical tools like Clarke's generalized Jacobian are applied to magnetostatic field problems with hysteresis. The relation between the non-smooth Newton method and other methods for solving non-linear systems with hysteresis like the M(B)-iteration is established.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering , vol. 38 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

1 – 3 of 3