Ebook: Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering

This book contains the papers presented at the International Symposium on Electromagnetic Fields in Mechatronics, Electrical and Electronic Engineering ISEF'05 which was held in Baiona, Spain on September 15–17, 2005. ISEF conferences have been organized since 1985 as a common initiative of Polish and European researchers who deal with electromagnetic fields applied to electrical engineering. Until now the conferences have been held every two years, either in Poland or in one of European academic centres renowned for electromagnetic research. The Royal Village of Baiona is well-known in the world for its beauty and historical background and meaning, as it was the place where Columbus landed after his travels to the New World. The venue of the conference is very close to Vigo, which is one of the most important cities in Spanish Galicia. The University of Vigo was the logistic centre of the conference.

It is the tradition of the ISEF meetings that they try to tackle quite a vast area of computational and applied electromagnetics. Moreover, the ISEF symposia aim at joining theory and practice, thus the majority of papers are deeply rooted in engineering problems, being simultaneously of high theoretical level. Continuing this tradition, we hope to touch the heart of the matter in electromagnetism.

After the selection process 211 papers were accepted for presentation at the Symposium and almost all of them were presented at the conference both orally and in the poster sessions. The papers have been divided into the following groups:

• Computational Electromagnetics

• Electromagnetic Engineering

• Coupled Field and Special Applications

• Micro- and Special Devices

• Bioelectromagnetics and Electromagnetic Hazards

• Magnetic Material Modelling

The papers which were presented at the symposium had been reviewed and assessed by the sessions' chairmen and the Editorial Board assembled for the postconference issue of ISEF'05. All the papers accepted for further publication were divided into two parts: those of more computational and those of more applicable nature. The latter ones are published here while the first, which contains less papers (26), went to COMPEL journal.

COMPEL: The International Journal for Computation and Mathematics in Electrical and Electronic Engineering, vol. 25, No. 3/2006.

The papers selected for this volume have been grouped in three chapters which cover the topics:

• Fundamental Problems

• Computational Electromagnetics

• Applied Electromagnetism

They introduce some necessary order while reading the book and somehow represent the main directions which are penetrated by researchers dealing with contemporary electromagnetics. Looking at the content of the book one may also notice that more and more researchers are engaging in investigation of electromagnetic applications, especially these connected with mechatronics, information technologies, medicine, biology and material sciences. This is readily seen when looking at a number of papers which belong to each part of the book. Computational techniques which were under development during the last three decades and are still being developed serve as good tools for discovering new electromagnetic phenomena. This means that computational electromagnetics belongs to an application area rather than to research area. Anyway, a lot of researchers are still working on this subject.

The first chapter is devoted to fundamental problems which appear in electromagnetics. The subject has been intensively developed from the very beginning of the history of electromagnetism. Some problems are still the subject of scientific discussions and are still unsolved; other problems are discussed as the background for building numerical models. There is a group of papers dealing with the analysis of material properties (magnetic, dielectric or even biological), and thus the papers have more physical than computational considerations. Such an analysis is somehow out of the scope of international conferences, like COMPUMAG, AEM or CEFC, which rather focus their attention on computational aspects. Also, there are some papers in the chapter which show analytical solutions of the electromagnetic field problems. The analytical approach is nowadays mostly not of interest for researchers, especially those of the younger generation, but this attitude seems not to be supported by the reality of electromagnetism. The analysis of electromagnetic field sometimes requires deeper insight into the structure of mathematical model and this can be done just by means of analytical approach.

The second chapter is devoted to problems widely discussed at all conferences, namely how to improve the efficiency, accuracy and ability of numerical models. Older participants of the conference and readers of the book will remember that a few decades ago the majority of papers concentrated on the problems of how to build FEM solver or how to regard nonlinearity or dynamics in computer modelling. Also, there were strong discussions about the advantages and disadvantages of such numerical methods, such as FEM, FDM, BEM, and during the conferences followers of each particular method tended to form separate groups. Nowadays, these problems are mostly behind us and the interest of researchers is focused on some specific contributions to the existing solvers.

The last chapter, the richest in papers, is a good mirror of what the ISEF conference is. As has already been mentioned, most papers deal with some applications of electromagnetic field and the stress in these papers is put on the phenomena or devices or both, and computational technique is there just as a tool to understand the phenomena, to design a device or to know what hazards can occur. It seems to be very difficult, even impossible, to find some common idea which joins the papers in some groups because, in fact, each paper deals with its own particular problem. Prospective readers are recommended to look at this chapter very carefully and find the paper of their personal interest.

At the end of these remarks let us, the Editors of the book, be allowed to express our thanks to our colleagues who have contributed to the book by peer-reviewing the papers at the conference as well as in the publishing process. We also convey our thanks to IOS Press (Publisher) for their effective collaboration in shaping this editorial enterprise. As ISEF conferences are organised biannually we do hope to keep our strong links with IOS Press in the future.

Xose M. Lopez-Fernandez, Chairman of the Organising Committee

Andrzej Krawczyk, Scientific Secretary

Sławomir Wiak, Chairman of the ISEF Symposium

In this paper, co-evolutionary augmented lagrangian method (CEALM) is applied to the constrained optimization problem of electromagnetic devices. To verify the performance of the CEALM, it is applied to the superconducting magnetic energy storage (SMES) optimization benchmark problem. As a result, accurate global solution can be found with less computational efforts. However, the conventional CEALM has a problem that the convergence speed decreases as the solution approaches near a global optimum. In this paper, to solve the problem, an improved scheme for CEALM is proposed. The effectiveness of the improved CEALM will be proven by the comparison with the conventional one.

System consisting of infinite long metal cylinder and two infinite long conductors perpendicular to its axis, which conduct sinusoidally alternating (both in relation to time and spatial coordinate) exciting current is considered in the paper. Aim of the present work is estimate of influence of displacement currents on electrodynamics field's distribution in close zone.

Having defined the applicative limits of the duality rule of E. Colin Cherry, the criteria concerning the extension to the case of time varying networks is analyzed. A topological criterion emerges that, while confirming the efficacy already seen for the analysis of transformers, is suited for being adapted in a unified interpretation involving rotating machinery.

The reduction of the magnetic field emitted by industrial sources, such as MV/LV substations, has been faced by means of computation and experimentation. The proper selection of parameters and dimensions of the magnetic screens is critical for their efficiency. In this paper, the influence of the thickness is analyzed varying it for different plane shape ferromagnetic and conductive materials (Fe, Al and Cu). A typical three phase conductor distribution has been used as magnetic source. The results of the measurements validate those of the simulations. The graphics presented show the limitation phenomenon appearing in the efficiency of the different materials which demonstrates that shielding effectiveness is not lineally improved by an increment in screen thickness. For each of the materials there is one optimal value for the thickness which is shown and analyzed in this paper.

In this paper an analytic approach to deal with vector diffusion problems into hysteretic materials is presented. A first procedure attempts to find a solution to the nonlinear problem approximating the differential magnetic permeability tensor by a set of polynomials; the second method reduces the tensor nonlinear terms to equivalent optimized linear coefficients. Scalar hysteresis cycles are reported to better show how the methods work and a hypothetic anisotropy is chosen to carry out preliminary results and discussions for vector problems.

The paper presents semi-anlytical method of the calculation of SAR (Specific Absorption Rate) coefficient and such approach enables investigating of the internal structure of the coefficient. It is possible to trace the influence of electromagnetic parameters, such as frequency and polarisation on the value of SAR. The biological model is an isotropic lossy dielectric spheroid. The excitation is a uniform plane wave. The correctness of this approach has been verified by an experiment made on spheroidal phantom.

The aim of this paper is to describe a software tool dedicated to the design of electromechanical devices. It is based on a serial implementation of an analytical and a numerical procedure. The first one, from datasheet specification give a raisonnable preliminary structure. The second one uses a finite element calculation, associated to a sensitivity analysis and optimization method. For the sensitivity analysis, useful fractional factorial designs are performed to identify the most influential factors and to determine an initial configuration for the optimization procedure.

Our goal is to describe a comprehensive procedure to determine the radiation of a mobile phone take into consideration the influence of the presence of a human head hard by. In the course of our calculations we have been concerned chiefly with the physical effect, namely the influence of the head on radiation characteristic and efficiency. We applied variational principle method and supplied boundary conditions with using R-functions.

The evolution in the field of modeling software, especially in numerical simulation, leads to the complication of its organization. Thus, the needs of data structure optimization are also growing progressively. This paper describes an approach to what may be called implementing business rules in an object-oriented model applied to a multi-physics numerical simulation software. Thanks to this proposition, the complexity of data model manipulations may be reduced.

The effects of the kinetics of polarization switching in ferroelectrics, taking into account the spatio-temporal fluctuations of the polarization field, given by the Langevin and Fokker-Planck equations, are studied. A Fokker-Planck equation is derived, which describes the polarization reversal in terms of Fourier amplitudes. A monotone and stable exponential-type finite volume difference scheme is elaborated to solve this equation numerically in a spatially homogeneous case, where the equation is one-dimensional. The temporal behaviour of the polarization probability distribution is calculated in presence of a sinusoidal field and white noise. The resulting hysteresis of the mean polarization is discussed for different noise intensities.

The paper shows the calculation mode of mathematical model parameters starting from catalogue data and presents the simplified mathematical model where are introduced the catalogue data as well as calculated parameters. This model is valuable for slow process. These parameters of the mathematical model can be determinated as well by experimental tests being structured in way to permit the determination of all the parameters. The stator resistance winding, the field winding one, the synchronous reactances longitudinal and quadrature unsaturated have been determinated by classic methods; for the determination of the other parameters one has employed the static methods (fixed rotor) – for the direct-axis d, the system is a quadripole permitting by a supplying of a terminal couple two kinds of tests by short-circuiting or by no load operation of the two other terminals.

An essential step in the development of quantum Hall devices for ac metrological standards is the frequency characterisation of multiconductor resistances [1]. The needed resistance values would require the use of rather long wires; a proposed solution consists in folding the wire into multiconductor geometries. In this work we present an accurate modeling of a quadrifilar resistance in the range 0–10 kHz, using a transmission-line approach. Capacitances were obtained with a version of the Boundary Element Method optimised to treat very different length scales; inductances, eddy currents and skin effects are computed through analytical formulae paying special attention to the singularities occurring in the problem.

In this paper it is presented some of the authors experience in teaching computational electromagnetics to students of university level in an electromagnetic laboratory. It is described the practical case of an E core electromagnet used to give “hands on experience” on finite element modelling of electromagnetic devices.

In this paper, a microwave imaging technique for reconstructing the shape of two-dimensional perfectly conducting scatterers by means of the particle swarm optimization algorithm is proposed. The reconstruction is based on scattered field simulated measurements derived by transverse magnetic illuminations. Two different implementations of the particle swarm optimization algorithm, namely the synchronous and the asynchronous one, are considered. Furthermore, the robustness of the algorithm with respect to different initial particle populations is investigated. Finally, the performance of the technique in the case of noisy scattered field data is examined.

In this paper a new dynamical hysteresis model is introduced that based on the second derivates of the measured reversal curves. The advantages of the model are its easy identification, past memory representation and numerical simplicity. The accommodation property of the model can be ensued from the construction philosophy of the model. The feasibility of the extrapolation is defined by the boundary conditions. The convenience of this model is its efficiency in reduction of the calculation time. The parameters can be determined from the measured data points. Reported results prove the accuracy of the model.

Finite element modelling is used to investigate and optimise the radial stiffness of a levitated disc thereby improving its radial stability. The disc is to be used as a bearing for high-speed micro-machines such as micro-motors and micro-turbine generators. Particle Swarm Optimisation is used to find the optimal geometry and excitation which yield the best radial stiffness while maintaining strong and uniform axial support.

The calculation of the power losses in a ferromagnetic is often executed with the utilisation of analytic dependence. This dependence was marked on the basis of the Maxwell equations. So, it accepts the ferromagnetic material as devoid of the magnetic domains. The domains behaviour in external magnetic field is imitated with help of the magnetic permeability. However, when in a real ferromagnetic two field harmonics coexists (the fundamental harmonic which establish the magnetic work conditions and the higher order harmonic with small amplitude), the computed power loss values can be unreal. The authors of this paper show the dependence of the power losses (the power losses caused by higher order harmonic) in relation with the temporary shape of the wall (the temporary shape is connected with the temporary value of the fundamental harmonic field integrated over a cross section).

The numerical simulation of electromagnetic devices with motion by the finite-element method requires some method to couple the fixed and moving parts. Improvements in automatic meshing techniques have enabled a gap remeshing method to be included in the finite-element analysis package OPERA-3d/CARMEN to achieve this coupling. This method allows devices, such as motors, with complex air gap geometries to be modelled and has some advantages over other techniques which can be employed. The gap remeshing approach is demonstrated for an axial flux permanent magnet motor where the open circuit back e.m.f. characteristics of the motor are presented.

In this paper a contribution to the improvement of accuracy/speed in induction motor optimization is presented. It is based on the reduction of the number of variables involved by first finding similarities between different geometrical shapes in the magnetic circuit. Therefore, a prior filtering of the “shapes” to be computed is performed, allowing the use of accurate optimization algorithms with a reduced number of specimens to be tested. This allows the use of more accurate optimization procedures that conventionally have been limited in their scope due to the time required to simulate every single motor (for instance, finite element analysis).

The great development of Large Variable Speed Wind Turbine Generators has caused a new interest in the Brushless Doubly-Fed Induction Machines (B.D.F.I.M) because they could be a feasible alternative to Wound Rotor Induction Machines (W.R.I.M). The B.D.F.I.M. has two different stator windings and a rotor cage that make it more robust than the W.R.I.M. The operation of the B.D.F.I.M. is possible due to the two stator magnetic fields couple each other through the rotor cage which has a special configuration. This paper studies these coupling mechanisms comparing simulation results, obtained with a Finite Element (FE) Model, and experimental result, obtained with a Prototype of B.D.F.I.M.