Search results

21 – 30 of 135
Article
Publication date: 9 September 2013

Steven Bielby and David A. Lowther

The conventional starting point for the design of an electrical machine (or any low-frequency electromagnetic device) is known as “sizing”. In this process, a simple magnetic…

Abstract

Purpose

The conventional starting point for the design of an electrical machine (or any low-frequency electromagnetic device) is known as “sizing”. In this process, a simple magnetic circuit is used to estimate the main geometric parameters. This does not work for many devices, particularly where eddy currents and non-linearity dominate. The purpose of this paper is to investigate an approach using a neural network trained on a large database of existing designs as a general sizing system.

Design/methodology/approach

The approach is based on a combination of a radial basis function neural network and a database of stored performances of electrical machines. The network is trained based on a set of typical performance requirements for a machines design problem. The resulting design is analyzed using finite elements to determine if the design performance is acceptable.

Findings

The number of neurons in the network was varied to determine the approximation and generalization capabilities. The finite element analysis showed that the network produced initial design parameters which resulted in an appropriate performance.

Research limitations/implications

The research has looked at only one class of machine. Further work is needed on a range of machines to determine how effective the approach can be.

Practical implications

The approach can provide a good initial design and thus can reduce overall design time significantly.

Originality/value

The paper proposes a novel, fast and effective generalized approach to sizing low frequency electromagnetic devices.

Details

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, vol. 32 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 9 September 2013

Leena Lehti, Janne Keränen, Saku Suuriniemi, Timo Tarhasaari and Lauri Kettunen

The authors aim to search for a practical and accurate way to get good loss estimates for coil filaments in electrical machines, for example transformers. At the moment including…

Abstract

Purpose

The authors aim to search for a practical and accurate way to get good loss estimates for coil filaments in electrical machines, for example transformers. At the moment including loss estimations into standard finite element computations is prohibitively expensive for large coils.

Design/methodology/approach

A low-dimensional function space for finite element method (FEM) is introduced on the filament-air interface and then extended into the filament to significantly reduce the number of unknowns per filament. Careful choice of these extensions enables good loss estimate accuracy. The result is a system matrix assembly block that can be used verbatim for all filaments, further reducing the cost. Both net current and voltage per length of the filament are readily available in the problem formulation.

Findings

The loss estimates from the developed model agree well with traditional FEM and the computation times are faster.

Originality/value

To produce accurate loss estimates in large coils, the low-dimensional function space is constricted on the filament boundaries. The proposed method enables electrical engineers to compute the ohmic losses of individual conductors.

Details

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, vol. 32 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 19 June 2007

Saku Suuriniemi, Jari Kangas and Lauri Kettunen

The paper addresses various ways of driving a magneto‐quasi‐static coupled field‐circuit problems, starting with the underlying assumptions of this problem class. It focuses on…

Abstract

Purpose

The paper addresses various ways of driving a magneto‐quasi‐static coupled field‐circuit problems, starting with the underlying assumptions of this problem class. It focuses on problem consistency, supporting both conceptual understanding, and translation into software.

Design/methodology/approach

The paper proceeds from a precisely defined problem class and analyze its consistency with homology theory.

Findings

Precise notion of “driving a problem,” extensive discussion of modeling assumptions and decisions, and classification and consistency analysis of various driving methods.

Practical implications

Helps modelers systematically pose consistent coupled field‐circuit problems. The computation of homology groups can be automated to help pose problems and detect consistency problems.

Originality/value

Starting from the basic underlying assumptions, the paper summarizes logically the application of homology to consistency analysis. The style is tutorial for modelers, with numerous particular cases.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 26 no. 3
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 7 September 2015

Tetsuji Matsuo, Jun Kawahara, Tomohiro Shimoi and Takeshi Mifune

The purpose of this paper is to examine the numerical stability of a space-time finite integration (FI) method. A symmetric correction is proposed to give an accurate constitutive…

Abstract

Purpose

The purpose of this paper is to examine the numerical stability of a space-time finite integration (FI) method. A symmetric correction is proposed to give an accurate constitutive relation at the subgrid connections.

Design/methodology/approach

A scheme for the numerical stability analysis of the space-time FI method is presented, where the growth rate of instability is evaluated by a numerical eigenvalue analysis formulated from an explicit time-marching scheme.

Findings

The 3D and 4D subgrid schemes using the space-time FI method are conditionally stable, where a symmetric correction does not induce numerical instability. The staircase-type 4D space-time subgrid allows a larger time-step than the straight-type subgrid.

Originality/value

The numerical stability of space-time FI method is proven by an eigenvalue analysis, which provides 3D and 4D stable subgrid schemes.

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: 1 September 2003

Piergiorgio Alotto, Christian Magele, Werner Renhart, Andreas Weber and Gerald Steiner

Uncertainties in the design variables of non‐linear engineering optimization problems are often neglected. That could result in considerable deterioration of the target function…

Abstract

Uncertainties in the design variables of non‐linear engineering optimization problems are often neglected. That could result in considerable deterioration of the target function value of an implemented design compared with the computed optimal solution. This effect can be reduced with robust optimization, where it is tried to achieve robust designs by actively embedding the uncertainties and robustness measures in the optimization process. A methodology for robust optimization of non‐linear problems is presented, including practical methods for the solution of such programs. The benefits of the approach are discussed in a numerical field calculation example.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 22 no. 3
Type: Research Article
ISSN: 0332-1649

Keywords

Abstract

Purpose

The paper aims to illustrate a numerical technique to calculate fields and inductances of rotating electrical machines.

Design/methodology/approach

The technique is based on an integral formulation of the nonlinear magnetostatic model in terms of the unknown magnetization. The solution is obtained by means of a Picard-Banach iteration whose convergence can be theoretically proved.

Findings

The proposed method has been used to build a model of a large turbine generator. In particular, the influence of end effects on flux linkages has been computed. It has been demonstrated that the 2D solution underestimates the flux linkages as well as the no load voltage of 2 per cent, while the leakage fluxes are computed by the 2D solution with errors as high as 20 per cent.

Originality/value

The method is advantageous in comparison to standard methods.

Details

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, vol. 32 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 7 September 2015

Patrick Dular, Mauricio V. Ferreira da Luz, Patrick Kuo-Peng and Laurent Krähenbühl

The purpose of this paper is to develop a subproblem finite element method for progressive modeling of lamination stacks in magnetic cores, from homogenized solutions up to…

Abstract

Purpose

The purpose of this paper is to develop a subproblem finite element method for progressive modeling of lamination stacks in magnetic cores, from homogenized solutions up to accurate eddy current distributions and losses.

Design/methodology/approach

The homogenization of lamination stacks, subject to both longitudinal and transversal magnetic fluxes, is first performed and is followed by local correction subproblems in certain laminations separately, surrounded by their insulating layers and the remaining laminations kept homogenized. The sources for the local corrections are originally defined via interface conditions to allow the coupling between homogenized and non-homogenized portions.

Findings

The errors proper to the homogenization model, which neglects fringing effects, can be locally corrected in some selected portions via local eddy current subproblems considering the actual geometries and properties of the related laminations. The fineness of the mesh can thus be concentrated in these portions, while keeping a coupling with the rest of the laminations kept homogenized.

Research limitations/implications

The method has been tested on a 2D case having linear material properties. It is however directly applicable in 3D. Its extension to the time domain with non-linear properties will be done.

Originality/value

The resulting subproblem method allows accurate and efficient calculations of eddy current losses in lamination stacks, which is generally unfeasible for real applications with a single problem approach. The accuracy and efficiency are obtained thanks to a proper refined mesh for each subproblem and the reuse of previous solutions to be locally corrected only acting in interface conditions. Corrections are progressively obtained up to accurate eddy current distributions in the laminations, allowing to improve the resulting global quantities: the Joule losses in the laminations, and the resistances and inductances of the surrounding windings.

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: 11 September 2009

Leandro dos Santos Coelho and Piergiorgio Alotto

The purpose of this paper is to show, on a widely used benchmark problem, that normative knowledge concepts can be incorporated into particle swarm optimization (PSO) algorithms…

Abstract

Purpose

The purpose of this paper is to show, on a widely used benchmark problem, that normative knowledge concepts can be incorporated into particle swarm optimization (PSO) algorithms in order to improve their search ability.

Design/methodology/approach

Normative knowledge concepts are used within the framework of PSO algorithms in order to influence the cognitive and social components of the particle behaviour.

Findings

It is shown that the proposed algorithm can significantly improve the performance of PSO on the selected benchmark problem, in terms of both best and average solutions.

Research limitations/implications

Although the chosen benchmark is considered to be representative of typical electromagnetic problems, different test cases may give less satisfactory results.

Practical implications

The proposed approach appears to be an efficient general purpose stochastic optimizer for electromagnetic design problems.

Originality/value

This paper introduces the use of normative knowledge concepts to control the cognitive and social components of PSO algorithms.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 28 no. 5
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 19 June 2007

C.W. Trowbridge and H.R.McK. Hyder

To investigate the effectiveness of a simple analytic model for the calculation of electrostatic fields and optical properties of particle beams in accelerator tubes.

Abstract

Purpose

To investigate the effectiveness of a simple analytic model for the calculation of electrostatic fields and optical properties of particle beams in accelerator tubes.

Design/methodology/approach

The analytic model is derived by superimposing the solution obtained from two co‐axial circular apertures. Potentials and fields derived from the model are compared to solutions based on finite element analysis together with first, third, and fifth order optical parameters and aberration. More realistic models are also compared.

Findings

It is shown that the model is capable of capturing many of the ion optical effects produced in small electrostatic accelerator tubes. Even for more realistic tube models the basic optical parameters, first and third order, are sufficiently accurate to allow a preliminary design to be achieved.

Originality/value

The work demonstrates that the first stages of the optical design of accelerator tubes can be rapidly achieved by analytic means which then provide good starting parameters for a fuller optimisation carried out by advanced numerical methods. The results also serve as a benchmark for numerical methods.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 26 no. 3
Type: Research Article
ISSN: 0332-1649

Keywords

Article
Publication date: 19 June 2007

Daniel Watzenig, Gerald Steiner, Anton Fuchs, Hubert Zangl and Bernhard Brandstätter

The investigation of the influence of the modeling error on the solution of the inverse problem given uncertain measured data in electrical capacitance tomography (ECT).

Abstract

Purpose

The investigation of the influence of the modeling error on the solution of the inverse problem given uncertain measured data in electrical capacitance tomography (ECT).

Design/methodology/approach

The solution of the nonlinear inverse problem in ECT and hence, the obtainable accuracy of the reconstruction result, highly depends on the numerical modeling of the forward map and on the required regularization. The inherent discretization error propagates through the forward map, the solution of the inverse problem, the subsequent calculation of process parameters and properties and may lead to a substantial estimation error. Within this work different finite element meshes are compared in terms of obtainable reconstruction accuracy. In order to characterize the reconstruction results, two error measures are introduced, a relative integral error and the relative error in material fraction. In addition, the influence of the measurement noise given different meshes is investigated from the statistical point of view using repeated measurements.

Findings

The modeling error, the degree of regularization, and measurement uncertainties are the determining and limiting factors for the obtainable reconstruction accuracy of electrical tomography systems. The impact of these key influence factors on the calculation of process properties given both synthetic as well as measured data is quantified. Practical implications – The obtained results show that especially for measured data, the variability in calculated parameters strongly depends on the efforts put on the forward modeling, i.e. on an appropriate finite element mesh size. Hence, an investigation of the modeling error is highly recommended when real‐world tomography problems have to be solved.

Originality/value

The results presented in this work clearly show how the modeling error as well as inherent measurement uncertainties influence the solution of the inverse problem and the posterior calculation of certain parameters like void fraction in process tomography.

Details

COMPEL - The international journal for computation and mathematics in electrical and electronic engineering, vol. 26 no. 3
Type: Research Article
ISSN: 0332-1649

Keywords

21 – 30 of 135