New Publications are available for Magnetostatics
http://dl-live.theiet.org
New Publications are available now online for this publication.
Please follow the links to view the publication.Application of wedge functions to 2D magnetostatic problems
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0027
In this paper the application of so called wedge functions is presented to solve two-dimensional simple geometries magnetostatic and electrostatic problems, e.g. rectangles of varying aspect ratio and with different values of the magnetic permeability μ. Such wedge solutions contain surface pole distributions of a certain power. The same pole distribution functions have been used in the analysis of conducting and infinite permeability structures. Here we apply such functions in a boundary integral analysis method of finite permeability.Calculation of field distribution in electromagnetic problems with random domains
http://dl-live.theiet.org/content/conferences/10.1049/cp.2011.0014
While solving a problem with random domains, discontinuities on the magnetic field distribution in the stochastic dimension appear. Numerical methods are already available to account for random domains. Some of them require the use of additional approximation functions to tackle the problem of discontinuity. In this paper, we focus on a method based the use of a random mapping and we show that in that case the discontinuities are naturally approximated and that no additional approximation function needs to be added.Applying a simple approach for evaluating the magnetic fields produced by power lines
http://dl-live.theiet.org/content/conferences/10.1049/cp.2009.0527
The paper applies a new method for evaluating magnetic fields produced far from power lines. Simple formulae are used for calculating the magnitudes of the resultant, semi- major and semi-minor axes that characterize the magnetic fields produced by various power-line configurations. We analyze the first few terms of the power series that expresses the magnetic field produced by an assembly of parallel line currents, such as those used to approximate the conductors of electric-power transmission and distribution lines. The properties of the first few terms of this series will be described. Some assumptions have been taken into consideration. Field measurements were carried out for magnetic flux densities produced at large distances by three-phase power lines. The results were compared with the calculated values. Several methods for reducing the magnetic field generated by power lines are addressed. (4 pages)Exploiting partial symmetries in 3D magnetostatic calculations using a scalar potential formulation
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080247
A 3D magnetostatic formulation in terms of the magnetic scalar potential is used to simulate a long dipole magnet. The translatory symmetric part of the model is discretised by a combination of finite elements and Legendre polynomials, which results in a substantial reduction of the model size.Transient 3D magnetic field simulations with combined space & time mesh adaptivity for lowest order WFEM formulations
http://dl-live.theiet.org/content/conferences/10.1049/cp_20080210
A scheme for the totally error controlled simulation of three-dimensional transient magnetic field problems is presented, combining lowest order Whitney finite elements with Zienkiewicz-Zhu-type gradient recovery error estimators for spatial mesh adaptivity with higher order embedded SDIRK schemes for adaptive time step selection. Numerical results are presented for a magnetoquasistatic benchmark example.Modeling and design of axial-flux permanent magnet: a new approach
http://dl-live.theiet.org/content/conferences/10.1049/cp_20060129
In this paper, a new analytic approach to the calculation of magnetic field in a slot-less, two-rotor axial-flux permanent magnet machine is presented. The analytic modeling is based on calculation of scalar and vector magnetic potentials. The magnet is modeled by magnetization vector and a two dimensional current sheet represents the armature winding. The effects of armature reaction and the harmonics of field are also considered. The simulation of magnetic field by analytic model is compared with the 2-D finite element analysis. The analytic model predicts the magnetic field within %5 compared to FEA results. Ultimately, by using the analytic model in a genetic algorithm method, an optimal design of an axial-flux permanent magnet machine is presented.Generated voltage harmonics minimization of radial permanent magnet single-phase AC linear generator
http://dl-live.theiet.org/content/conferences/10.1049/cp_20060109
Two-dimensional finite element method is used in the analysis to calculate the magnetic field parameters in a radial permanent magnet linear single-phase AC generator. Both surface and internal permanent magnet schemes are analyzed and compared according to three different parameters; the cogging force, the induced AC voltage and the THD of the induced AC voltage. Three different design structures were used to reduce the THD of the induced voltage; tooth shoe, stepped tooth and chamfered tooth. The chamfered tooth provides as low THD as 4% and 12% for surface and internal prototype schemes, respectively. The cogging force was much reduced using the stator optimal length; for both schemes, in addition to the total generator weight is reduced.The use of a genetic algorithm in the design optimisation of a brushless DC permanent magnet machine rotor
http://dl-live.theiet.org/content/conferences/10.1049/cp_20040395
The paper describes the use of a genetic algorithm in the design optimisation of a surface magnet brushless DC machine. The method combines a two-dimensional magnetostatic finite element model (FEM) with a genetic algorithm (GA) to investigate the influence of the shape of the permanent magnets and their direction of magnetisation. The objective function used in the optimisation procedure is based on maximising torque for a given volume of permanent magnet, assuming the electrical loading of the machine remains unchanged. Compared to a standard radial magnetisation, the optimised rotor designs yield a significant improvement in output torque. To confirm the GA technique a gradient descent optimisation is also applied to the problem. (5 pages)Three-dimensional finite element modelling of a claw pole type car alternator
http://dl-live.theiet.org/content/conferences/10.1049/cp_20040329
This paper describes the three-dimensional finite element modelling of a claw pole type car alternator. The 3D finite element model takes into account the conducting and non-linear BH characteristics of the steel, the dynamic rotation of the rotor and the rectifying circuit to which the alternator is connected. The results also demonstrate how changes in the rotor pole shape affects the output of the alternator. (5 pages)Finite element method in magnetic field analysis of the inductive superconducting fault current limiter with an open core
http://dl-live.theiet.org/content/conferences/10.1049/cp_20040437
Finite element method in magnetic field analysis of the inductive superconducting fault current limiter with an open core is described in this paper. The V-I characteristics of SFCLs obtained by numerical analysis fulfill the utility network expectations to a great extend. The numerical analysis of magnetic field is very helpful in designing of inductive superconducting fault current limiter and obtaining the V-I characteristics of the device.Hybrid magnetostatic hydrodynamic numerical model of capture edge in OGMS separator
http://dl-live.theiet.org/content/conferences/10.1049/cp_20040456
This paper discusses the hybrid magnetostatic hydrodynamic numerical model of capture edge in OGMS separator. The operating principle of the OGMS (open gradient magnetic separation) separator consists in the selective deflection of particles, according to their magnetic susceptibility, during their passage through a strong and nonhomogeneous magnetic field generated by superconducting magnet with an appropriate configuration of coils. The superconducting OGMS separator for cleaning industrial water has been discussed. The numerical model can be used to research for the influence of the parameters of superconducting magnet on the capture edge shape in OGMS separator.Particle swarm optimisation for pareto optimal solutions in electromagnetic shape design
http://dl-live.theiet.org/content/conferences/10.1049/cp_20040468
This paper discusses particle swarm optimization (PSO) algorithm that was applied to multiobjective optimization problems with two and three objective functions. PSO is a stochastic optimization technique, which imitates the social behaviour of a bird flock flying around and sitting down on a pylon. The idea of the PSO implementation is that the flock should now sit down on the transmission line (the pareto front) and the pylons are controlling points (sampling points) for the flock. The algorithm revealed an excellent performance concerning the number of solutions of the forward problem and completed with a very reliable representation of the pareto optimal front. The pareto optimal front gives a survey of all reasonable solutions of the multiobjective problem.Comparison of different methods to estimate numerical errors in finite element problem coupled with external circuit equations
http://dl-live.theiet.org/content/conferences/10.1049/cp_20040457
The comparison of different methods to estimate numerical errors in finite element problem coupled with external circuit equations is discussed. In this paper, a new approach is proposed to discretise the electric circuit equation and to estimate the numerical errors. In the first step, both time discretisation scheme and error estimator are presented. In the second step, a comparison of the estimated numerical errors is carried out with a more classical scheme.The equivalence between cell method, TDTD and FEM
http://dl-live.theiet.org/content/conferences/10.1049/ic_20020135
Some equivalences between the cell method (CM) and other methods, like the finite difference in time domain method (FDTD) and the finite elements method (FEM), are proved. In particular the equivalence of the equations arises between CM and electrostatic 3D FEM, between CM and magnetostatic 3D FEM and between CM and FDTD in 3D electrodynamics. (2 pages)Coupled finite element - spectral element discretization for models with circular inclusions and far field domains
http://dl-live.theiet.org/content/conferences/10.1049/ic_20020160
In many technical models, large parts are source-free and consist of a homogeneous material. An analytical solution is found as linear combination of harmonic functions. In this paper, a finite set of harmonic functions is selected as basis in a spectral element method. Other model parts, suffering from complicated geometries, ferromagnetic saturation, hysteresis and eddy current effects, are discretized by finite elements. In this paper, the properties of spectral element, finite element coupled systems of equations are studied and specialized solution algorithms are developed. Although generally applicable, for simplicity, the methods are explained using a 2D magnetostatic example. (2 pages)Special hybrid FE/BE method for open magnetic field problems
http://dl-live.theiet.org/content/conferences/10.1049/ic_20020169
A special hybrid method was developed combining the finite element method (FEM) with a special boundary element method (BEM) to treat open boundary axisymmetric electromagnetic field problems. The FEM, applied for the field description in an inner region bounded by a fictitious spherical surface, is appropriate to take into account complex shaped devices in the presence of heterogeneous, anisotropic or even nonlinear media. The new BEM, used to take into account the field in the open homogeneous outer region, was based on an analytical treatment in spherical coordinates, consisting on an associated Legendre function series development. The method was applied to the static magnetic field evaluation due to the current flowing trough a main coil in the air. The analytical treatment, for the external field, reduces time and memory consumption for the computer evaluation, provides effective error control, and allows the field determination everywhere. Results were compared with those obtained either from experimental or by using other numerical methods. (2 pages)Enhanced accuracy of 2D finite element field quantities by a local post-process
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960179
Energy conversion in electromagnetic energy transducers takes place in the air gap only. Numerical field computation techniques of general application range are used for the design and optimisation of those electromagnetic devices. To predict the operational behaviour of such devices, particular attention has to be paid to the computation of the air gap values of the flux densities and the magnetic field strength. Ongoing publications and research activities on force computations in electromagnetic devices using different approaches indicate the importance of this field. This paper is focused on the practical application of the static electromagnetic field solution of the Laplace equation in a local post-process to increase the accuracy of an existing solution obtained by the standard two-dimensional finite element method using triangular types of elements. Advantages and drawbacks are discussed.Regularization of near singularity in FE-BE coupling analysis of three-dimensional static magnetic fields
http://dl-live.theiet.org/content/conferences/10.1049/cp_19960173
This paper describes a coupled finite element (FE) boundary element (BE) method for three-dimensional magnetostatic field computations. Finite regions consisting of magnetic materials and current sources are modeled by the FE method while the infinite vacuum region surrounding the FE region is analyzed by the BE method. Scalar potentials are chosen as unknown functions in both regions. The current vector potential induced by a source current is expressed by edge elements, which allows one to reduce cancellation errors in the high permeability region, and to give simple boundary conditions for the potentials on the interface between the FE and BE regions. The magnetic field in the BE region, computed by differentiating the the potential, is accurately computed by a technique that regularizes the near singularity of the kernels in the integral equations.An equivalent electric network model for nonlinear 3-D magnetostatic field analysis
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940017
An equivalent electric network model for nonlinear magnetostatic field analysis is presented. The electric circuit allows a simple analysis of the magnetic field distribution in the presence of homogeneous and inhomogeneous materials and allows an easy linkage of lumped parameter circuits and field equations. The model can be used for the analysis of direct and inverse problems, furthermore, an estimate of the B-H curve of a nonlinear material can be obtained. The method has been tested comparing the calculated results with linear and nonlinear benchmark problems. Then the characteristics of the method are shown analysing the magnetic field due to an electromagnet with a large air gap.Permanent magnet modelling by edge element and facet element (discussion on the force distribution in permanent magnet)
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940040
The permanent magnet is modeled by two dual formulations (h-oriented and b-oriented formulations) by using respectively edge element and facet element in three dimensions. The virtual work principle is applied for the force calculation in the permanent magnet. The calculation of local forces is discussed.Application of sensitivity analysis to the calculation of electromagnetic field quantities
http://dl-live.theiet.org/content/conferences/10.1049/cp_19940057
This paper investigates the application of sensitivity analysis to the calculation of electromagnetic field quantities, such as force. This is particularly important in the design of electrical devices. The electromagnetic field is described by means of a finite element network model which is similar to the classical finite element method. Tellegen's theorem is then applied to the network model to assess the sensitivity analysis. The method is simple requiring only one calculation of the electromagnetic field, for any small perturbation of the system. An algorithm and a 2D magnetostatic problem are discussed.Automated <i xmlns="http://pub2web.metastore.ingenta.com/ns/">B</i>–<i xmlns="http://pub2web.metastore.ingenta.com/ns/">H</i> curve identification algorithm combining field simulation with optimisation methods and exploiting parallel computation
http://dl-live.theiet.org/content/journals/10.1049/iet-smt.2011.0109
The study proposes an automated procedure for linking an identification algorithm implemented in a general-purpose environment (MatLab) with a commercial finite-element code for magnetic field analysis (VF-Opera). The method is used to identify the <strong xmlns="http://pub2web.metastore.ingenta.com/ns/"><i>B</i></strong>–<strong xmlns="http://pub2web.metastore.ingenta.com/ns/"><i>H</i></strong> curves of anisotropic magnetic laminations in the direction normal to the sheet surface with the aid of a multiprocessor computer to benefit from parallel computing. The proposed approach, based on a non-deterministic algorithm, makes it possible to determine the entire <strong xmlns="http://pub2web.metastore.ingenta.com/ns/"><i>B</i></strong>–<strong xmlns="http://pub2web.metastore.ingenta.com/ns/"><i>H</i></strong> curve and is therefore an advancement on the commonly used assumption of constant permeability.Monte Carlo simulation of an radio frequency identification system with moving transponders using the partial element equivalent circuit method
http://dl-live.theiet.org/content/journals/10.1049/iet-map.2009.0414
When designing an radio frequency identification system it is important to take both the position and the movement of the transponders into account. In this study, a simulation method that enables the description of a complete RFID system including moving and rotating transponders as well as a complex, industrial environment is presented. By using the partial element equivalent circuit method to calculate the magnetic field generated by the reader antenna and describing the transponders using a magnetic dipole, it is possible to use the Monte Carlo method to describe the dynamic behaviour of the complete system. The method is used in this study to describe the difference in performance between two different reader antennas and these results are also compared to measurements of similar systems operating in an industrial environment. The difference in performance between the two systems is similar in both the simulations and the measurements. A small discrepancy was seen between the results from the simulations and the measurements which is for the most part because of the limited read rate of the RFID systems used in the measurements.Cut-off space of a gyromagnetic planar disk resonator with a triplet of stubs with up and down magnetisation
http://dl-live.theiet.org/content/journals/10.1049/iet-map.2011.0491
A switchable circulator may be realised by internally or externally switching or latching a suitable ferrite resonator at the junction of three transmission lines or waveguides. This study investigates the split frequencies and therefore the quality factor of a geometry consisting of a half-wave-long cylindrical cavity and another which consists of a tri-toroidal structure. The model adopted in this study divides the surface of the resonator into up and down uniformly magnetised but unequal regions. The boundary, between the two regions, which is taken as the location of a single or a pair of switching wires, and the split counter-rotating frequencies of a typical resonator is theoretically obtained. This is done by investigating the split cut-off numbers of the related planar circuit with top and bottom electric walls and a magnetic sidewall. The magnetostatic model adopted in this study assumes equal fluxes in the up and down magnetised regions but unequal flux densities.Four- and eight-piece Halbach array analysis and geometry optimisation for maglev
http://dl-live.theiet.org/content/journals/10.1049/ip-epa_20045059
A systematic analysis approach is presented for four- and eight-piece Halbach arrays’ magnetic field and the geometric optimisation for magnetic levitation (Maglev) applications, which utilise the non-ideal array's magnetic field harmonics. The field analysis results obtained using scalar potential and Fourier series methods are confirmed with FEM results. The geometric optimisation is based on the maximisation of the ratio of the square of flux to a unit area magnet weight. The results of the systematic analysis of magnetic field and the geometric optimisation for practical Halbach arrays can be used as a technical framework for further Maglev system designs.Magnetic field tunable 75–110 GHz dielectric phase shifter
http://dl-live.theiet.org/content/journals/10.1049/el.2010.3124
A magnetic field tunable W-band phase shifter based on dielectric resonance in barium hexaferrite has been designed and characterised. A phase shift of 60° with low losses is demonstrated at low bias magnetic field.Magnetic polarisability of thick multiple rectangular apertures
http://dl-live.theiet.org/content/journals/10.1049/el_19991234
The magnetostatic potential distribution through multiple rectangular apertures in a thick conducting plane is analysed. Series solutions are given for the transverse and longitudinal magnetic polarisability of multiple rectangular apertures with various depth and shapes. The differences between electric and magnetic cases are discussed.Elemental field distributions in corrugated structures with large-amplitude gratings
http://dl-live.theiet.org/content/journals/10.1049/el_20031072
A unified and robust method has been introduced to compute magnetostatic and electrostatic fields for nonperiodic excitations in periodically corrugated anisotropic structures. The method extends a previously published concept of modified phased-periodic Green's functions for solving flat-surface problems. Convergent results have been obtained for corrugations with aspect ratios exceeding 15. Novel elemental field distributions are presented as examples.Field distributions in semiconductor junction circulator
http://dl-live.theiet.org/content/journals/10.1049/el_20000743
The axial magnetic field distribution within a three-port semiconductor junction circulator at several frequencies is computed, and the similarity to previously published results for the axial electric field in a ferrite junction circulator illustrates the gyrotropic duality between the semiconductor and the ferrite.Magnetising-field saturation and dynamic behaviour of induction machines. Part 2: Stability limits of a voltage-fed induction motor and of a self-excited induction generator
http://dl-live.theiet.org/content/journals/10.1049/ip-b.1983.0002
In Part 1 of the paper, an improved induction-machine model was derived, making more accurate allowances for the magnetising field saturation. This model yields satisfactory results, as proved experimentally. In this Part, two interesting applications of this saturated model are discussed. The first application deals with the stability behaviour of the voltage-fed induction motor. The stability region in the parameter space is shown to be much larger when calculated by means of the saturated model. This result competely agrees with experience. The second application deals with the self-excited induction generator. The necessity to use the improved model for the prediction of the dynamic behaviour is pointed out, because the classical model leads to unrealistic predictions; moreover, only the improved model allows the minimum flux level for which self-excitation still may occur to be determined. Also, the theoretically predicted stability limit is shown to agree with experiment.Simulation of transient and steady-state magnetisation characteristics with hysteresis
http://dl-live.theiet.org/content/journals/10.1049/piee.1977.0270
Exterior finite elements for 2-dimensional field problems with open boundaries
http://dl-live.theiet.org/content/journals/10.1049/piee.1977.0267
Electric- and magnetic-field problems with boundaries at infinity are treated in finite-element terms by constructing an element to model an extremely large annulus surrounding the region of interest. A simple recursion technique is employed to generate the matrix representing the annular region. All nodes are eliminated from the external element except those on its inner surface, so that the final matrix is no larger than that required to describe the region of interest only. The method is simpler to program and requires less computing effort than boundary-integral techniques. It has been tested by solving several 2-dimensional magnetostatic and electrostatic problems and comparing the results with analytic solutions. The method can be applied to any 2-dimensional field problem bounded by a large empty region in which the field satisfies Laplace's equation.Reply: Simulation of transient and steady-state magnetisation characteristics with hysteresis
http://dl-live.theiet.org/content/journals/10.1049/piee.1977.0271
Multigrid transmission-line modelling (TLM) method for solving electromagnetic field problems
http://dl-live.theiet.org/content/journals/10.1049/el_19911115
The development of a novel approach to solve electromagnetic problems in three dimensions is presented. The method described as the ‘multigrid TLM method’ uses different grid sizes to model space thus allowing the modelling of small features using a fine mesh of grid points, whereas the rest of space is modelled using a coarse mesh. The advantage of this method is that it offers simultaneously very high space resolution and modest computer storage and run-time requirements. The basis of the method is described briefly, together with typical results and conclusions.Comment: Active-<i xmlns="http://pub2web.metastore.ingenta.com/ns/">RC</i>-circuit synthesis for the simulation of a grounded inductor
http://dl-live.theiet.org/content/journals/10.1049/el_19850010
Reply: Active-<i xmlns="http://pub2web.metastore.ingenta.com/ns/">RC</i>-circuit synthesis for the simulation of a grounded inductor
http://dl-live.theiet.org/content/journals/10.1049/el_19850011
Comment: Fictitious potential for 3-D magnetic flux plots
http://dl-live.theiet.org/content/journals/10.1049/el_19860242
Magnetostatic problems in multiply connected regions: some properties of the curl operator
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1988.0028
The two main versions of the boundary-value problem of magnetostatics (the one with <i xmlns="http://pub2web.metastore.ingenta.com/ns/">h</i> normal to the wall, the one with a tangential magnetic field) are studied in connection with a deep underlying mathematical structure: the existence of a decomposition of the space of square summable vector fields in five closed subspaces. This structure encompasses classical Helmholtz decomposition in its different guises. It is best conceived in terms of differential forms, but, for pedagogical reasons, this point of view has been left aside in the present text, which only makes use of classical notions of applied functional analysis.Fast numerical method for calculation of electric and magnetic fields based on potential-flux duality
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1985.0017
By discretising the field into slices and tubes, upper and lower bounds to the energy of a system are obtained. Each slice carries the same flux and each tube has the same potentials at its ends. This makes the best possible use of the input data and avoids the necessity for solving a set of simultaneous equations, which is a characteristic feature of the usual finite-element method. Savings of computer time of one order of magnitude can be achieved in small problems and several orders of magnitude are likely in large problems.Computer package for calculating electric and magnetic fields exploiting dual energy bounds
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1988.0023
A computer graphics package, exploiting dual energy bounds for calculation of circuit parameters (<i xmlns="http://pub2web.metastore.ingenta.com/ns/">R, L, C</i>), has been recently developed in Southampton and implemented on a small desk-top computer. Simple pre-and postprocessing and considerable savings of computer time in comparison with other techniques are the most distinctive features of the method. The paper gives a brief description of this new software.Unified approach to problems in electromagnetism
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1984.0007
The development of a structure linking the various potentials and field vectors of the electric and magnetic fields is presented. The approach shows that it is usually possible to find alternative formulations, for a given problem, in a relatively simple manner. These alternative formulations, when correctly chosen, pose problems in dual or complementary form such that bounded solutions are possible. Such properties are particularly attractive when numerical solutions are being considered. The proposed structure is general, incorporating both electrostatic and magnetostatic systems as well as time varying field quantities.Magnetostatics with scalar potentials in multiply connected regions
http://dl-live.theiet.org/content/journals/10.1049/ip-a-2.1990.0038
Differential forms and electromagnetism in 3-dimensional Euclidean space <i xmlns="http://pub2web.metastore.ingenta.com/ns/">R</i><sup xmlns="http://pub2web.metastore.ingenta.com/ns/">3</sup>
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1986.0025
The paper develops the exterior algebra of differential forms as applied to electromagnetic theory in 3-dimensional Euclidean space (<i xmlns="http://pub2web.metastore.ingenta.com/ns/">R</i><sup xmlns="http://pub2web.metastore.ingenta.com/ns/">3</sup>). The synthesis afforded by the algebraic formulation allows the construction of diagrams which immediately relate the electromagnetic equations (Maxwell's and those of electrostatics and magnetostatics). The resulting framework is both pedagogically enlightening and valuable for practical applications.Complementary solutions of linear and nonlinear magnetostatic problems
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1988.0068
Complementary magnetostatic formulations and the associated bounds are derived systematically from an error based on the mismatch in the constitutive relationship. Two-dimensional finite-element results are presented for linear and nonlinear problems. Combined presentation of complementary computational results is shown to yield significant information on their comparative numerical behaviour, the discrepancies between them and discretisation errors.Use of simplex co-ordinates in dual finite-element calculations of Poissonian magnetic field problems
http://dl-live.theiet.org/content/journals/10.1049/ip-a-2.1990.0054
A method based on properties of simplex co-ordinates is proposed to be used in dual finite-element calculations in magnetostatics. The method considers the source's distribution within the field region and the interelement boundary conditions. The provision of upper and lower bounds, in connection with the obtained high rate of convergence, makes the method extremeley powerful and flexible. The simplicity and applicability of the method is shown by an example, where some previous results are also discussed.Dual finite-element calculations for static electric and magnetic fields
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1983.0019
The finite-element method is widely used for the solution of field problems but the method, as generally applied, suffers from the fact that it is not known how close the solution is to the actual value. This uncertainty can be reduced by providing a dual finite-element method which is so arranged that both methods together provide upper and lower bounds to the correct solution. It is found that the double approach also promises economies in the computation. The paper examines the physical basis of the dual method and applies it to Laplacian and Poissonian problems.Electromagnetism and semantics
http://dl-live.theiet.org/content/journals/10.1049/ip-a-2.1989.0048
Methods for the solution of open-boundary electromagnetic-field problems
http://dl-live.theiet.org/content/journals/10.1049/ip-a-1.1988.0024
A review of methods suitable for the solution of open-boundary field problems is given. Emphasis is placed on algorithms that can be used in conjunction with the finite-element method. The requirements for any algorithm to model an exterior region is outlined, as well as a description of some of the implications of the possible methods. It is hoped that users of such models will then appreciate the assumptions made in any method and thus its limitations, and so be able to choose the most appropriate one for a particular application.Magnetostatics with scalar potentials in multiply connected regions
http://dl-live.theiet.org/content/journals/10.1049/ip-a-2.1989.0008
Magnetostatic problems in multiply connected regions are studied with the use of magnetic scalar potentials. Particular emphasis is given to the significance of the nontrivial topological structure of the region, for the mathematical formulation. Practical rules are given for the necessary ‘cuts’ that are essential in numerical approaches to such problems.Comment: Bipolar magnetometer sensor similar to fluxgate device but having equal input and output sinusoidal frequencies and zero Barkhausen noise
http://dl-live.theiet.org/content/journals/10.1049/el_19931014