In future more electric aircraft (MEA) applications increased requirements of power levels stress the importance of low losses in converters and switching elements. In this work the losses of transistor switches are analysed and compared with the losses in magnetic amplifiers (Magamp) where amorphous alloy properties allows design of competitive devices. More Electric Aircraft technology can take advantage of this to accomplish compact, robust and reliable systems with low losses. Magamps are already used in the power supply of the Electronic Flight Control System in the Swedish "Gripen" fighter-aircraft. This paper presents a comparison between Magamp technology and an alternative Insulated Gate Bi-polar Transistor (IGBT) design. The study is concentrated on the power losses in a proposed 10 kW converter with ± 270 VDC (540 VDC) regulated output, where the fundamental difference between the switching elements is that the magnetic switch handles AC and the IGBT handles DC. To allow comparison, the copper and magnetic losses in the magnetic switch of the magnetic amplifier are considered as equivalent to the switching and conduction losses in the IGBT. The comparison is based on the sum of losses in power converters comprising six switching elements.
New soft magnetic materials made possible the use of the magnetic amplifier technology in designing competitive electric power supplies. This technology is used in the Swedish. fighter aircraft Gripen, being also attractive for future more electrical aircraft systems due to the possibility to achieve a compact and robust design. A modeling approach of a magnetic amplifier based on the magnetic hysteresis of the core material is presented here for a common amorphous magnetic alloy.
The purpose of this paper is to demonstrate the feasibility. of using a dynamic hysteresis model in simulating a magnetic amplifier (mag amp). The presented model includes static hysteresis, classical eddy currents, and excess losses. The proposed modeling approach is shown to be a feasible tool for designing mag amps.
A trend of increased Electric Power in Aircraft stresses the need of robust, low weight systems with low losses. New hard and soft magnetic materials have made it feasible to work with high speed and high frequency. Applications of new soft magnetic materials and hard magnetic materials like NeFeB magnets have enabled high density power generation systems. A new concept comprising a high speed PM-generator system and a magnetic amplifier control is presented. Magnetic amplifiers are used in the power supply of the Electronic Flight Control System in the Swedish "Gripen" fighter-aircraft. This technology is attractive in More Electric Aircraft (MEA) systems due to the possibility to achieve a compact, robust and a highly reliable system with low losses. Applications of new soft magnetic materials, such as amorphous magnetic alloys, have enabled the use of magnetic amplifier (magamp) technology in the design of competitive electric power converters. This paper presents a studied design on a 20-40 kW generator system including a +/- 270V controlled output, performed by magnetic amplifier technology. This work addresses the power generator, and the power converter. High speed PM-generators are offering high power density. The impact of operating a generation system with higher frequency and an increasing number of poles as well as the advantages with new soft magnetic materials is studied. The iron losses and the copper losses are analyzed for the generator, and the power converter.
As a contribution to the development of methods for the design and the analysis of devices based on giant magnetostrictive materials, a model for the simulation of the dynamic behaviour of the nonlinear magnetoelastic medium is presented. The coupled magnetic, magnetoelastic and mechanical equations that describe the magnetostrictive problem are solved by means of the finite element method. The thin sheets bending principle (surface splines) is used to introduce in the simulation the nonlinear properties of giant magnetostrictive materials, obtained by static characterizations
A phenomenological inherently vector hysteresis model employing simple differentials as a means of keeping track of the past history of the magnetic field is proposed. This results in a simple and computationally fairly efficient formulation for vector hysteresis of any dimension. An expression for determining model parameters from experimental data or from the Preisach function is given. The model exhibits rotational hysteresis and reduction of remanent magnetization by an orthogonal field and reduces to the classical Preisach model in one dimension. Details concerning the numerical implementation are discussed and computational examples demonstrating model properties are presented.
To obtain accurate design tools for applications involving giant magnetostrictive materials, magnetomechanical hysteresis effects should be taken into account. The problem consists of determining the magnetization and mechanical strain from the combined past history of magnetic field and mechanical stress. In this work, coupled magnetomechanical hysteresis has been modeled by using simple path #x2010;dependent differentials to accumulate the past history in functions related to the magnetization and strain through material #x2010;dependent parameters. By using anhysteretic curves and a few additional parameters to characterize a material, major, and minor loops with respect to both field and stress have been calculated for Terfenol #x2010;D and have shown good agreement with experiments.
The authors present a generalization of the classical Preisach model which handles coupled magnetic and mechanical hysteresis. Magnetostrictive materials are known to have hysteresis with respect to both magnetic field H and mechanical stress lambda;. To test the validity of the model, experiments where the two components H and lambda; have been verified in many different ways have been performed on Terfenol-D and compared to results computed from the model. Some of these results are presented. This stress-dependent model is found to have an accuracy comparable to that of the classical Preisach model
A method of expressing pseudoparticles with hysteresis within a context of irreversible thermodynamics is investigated. The state of a pseudoparticle is uniquely determined by its magnetization and its evolution is governed by entropy maximization. Hysteresis appears if the free energy is a nonconvex function of magnetization. The vectorial nature of magnetization and dependence on rate are accounted for in a systematic manner. Some basic properties are derived for quasistatic processes. In particular, it is found that in the scalar case, the magnetization is a monotonically increasing functional with respect to field and that for quasistatic processes, this implies the wiping-out property
A method for representing structures of laminated electric steel is investigated. The average behavior of laminates, including eddy current effects, is approximately represented using a rate-dependent constitutive law. The accuracy of the method is studied by comparing the homogenized description with finite difference calculations.
Magnetic hysteresis effects have been included in a finite #x2010;element description of a magnetic circuit by using the classical vector Preisach model for the constitutive relation between H and B. The influence of an external electric circuit is taken into account by adding equations derived from Faraday #x2019;s law. Computational results are presented for a magnetic circuit used in a magnetostrictive device.
In this paper, two different types of ultra-fast electromechanical actuators are compared using a multi-physical finite element simulation model that has been experimentally validated. They are equipped with a single-sided Thomson coil (TC) and a double-sided drive coil (DSC), respectively. The former consists of a spirally-wound flat coil with a copper armature on top, while the latter consists of two mirrored spiral coils that are connected in series. Initially, the geometry and construction of each of the actuating schemes are discussed. Subsequently, the theory behind the two force generation principles are described. Furthermore, the current, magnetic flux densities, accelerations, and induced stresses are analyzed. Moreover, mechanical loadability simulations are performed to study the impact on the requirements of the charging unit, the sensitivity of the parameters, and evaluate the degree of influence on the performance of both drives. Finally, it is confirmed that although the DSC is mechanically more complex, it has a greater efficiency than that of the TC.
The operational efficiency of ultra fast actuators usedas drives in high voltage direct current breakers are at best5 %. To boost their efficiency, the design of the energizing circuitis crucial. A multi-physics finite element method (FEM) modelcoupled with a SPICE circuit model that is able to predict theperformance of the actuator with an accuracy of at least 95 % hasbeen developed and verified experimentally. Several variants ofprototypes and models have been simulated, built, and tested.It was shown that one of the main problems leading to lowefficiencies is the stroke of the drive. However, there is a possibilityto increase the efficiency of the electric to mechanical energyconversion process of the studied Thomson (TC) and double sidedcoils (DSC) to a maximum of 54 % and 88 % respectively iftheir stroke is minimized. This can be done at the expense ofincreasing the complexity and the cost of the contact system bydesigning a switch with several series connected contacts that isencapsulated in a medium with a high dielectric strength. Anotherproposed solution is to design a current pulse with a rise timethat is considerably shorter than the mechanical response time ofthe system. Parametric variations of capacitances and chargingvoltages show that the TC and the DSC can achieve efficienciesup to 15 % and 23 % respectively. Regardless of the chosenmethod, the DSC has a superior efficiency compared to a TC.
The operational efficiency of ultrafast actuators used as drives in high-voltage direct-current breakers is at best 5%. To boost their efficiency, the design of the energizing circuit is crucial. A multiphysics finite-element method model coupled with a SPICE circuit model that is able to predict the performance of the actuator with an accuracy of at least 95% has been developed and verified experimentally. Several variants of prototypes and models have been simulated, built, and tested. It was shown that one of the main problems leading to low efficiencies is the stroke of the drive. However, there is a possibility to increase the efficiency of the electric to mechanical energy conversion process of the studied Thomson coil (TC) and double-sided coil (DSC) to a maximum of 54% and 88%, respectively, if their stroke is minimized. These efficiencies are idealistic, and these were obtained with clamped armature studies. The efficiency of the actuator can be increased at the expense of increasing the complexity and the cost of the contact system by designing a switch with several series-connected contacts that is encapsulated in a medium with a high dielectric strength. Another proposed solution is to design a current pulse with a rise time that is considerably shorter than the mechanical response time of the system. Parametric variations of capacitances and charging voltages show that the TC and the DSC can achieve efficiencies up to 15% and 23%, respectively. Regardless of the chosen method, the DSC has a superior efficiency compared to a TC.
One of the key enabling technologies for multi-terminal HVDCgrids is the existence of a breaker that can operate withina few milliseconds. A lot of research has been done to builddifferent ultra-fast drives to actuate the electric contacts ofthese breakers. What they all have in common is an operationalefficiency of at best 5 %. Capacitor banks are discharged throughspirally shaped flat coils to generate ultra-fast repulsive forces. Tooptimize the efficiency of the drive, the design of the energizingcircuit is crucial. The aim of this paper is to optimize theenergizing source and provide a deep explanation of the effectof the chosen capacitance and charging voltage for two actuatorconcepts, the Thomson coil (TC) and the double sided coil (DSC)for different stroke requirements. An experimentally validatedmulti-physics finite element method (FEM) simulation model is applied.
In this paper, an ultra-fast single-sided Thomson based actuator is studied. The actuator is comprised of a flat spiral-shaped coil with a conductive armature in its proximity. This armature is mechanically loaded with a uniform mass distribution over its cross section. The energizing source consists of a capacitor bank that is discharged through the actuator coil resulting in a high magnetic pressure within fractions of a millisecond. The coil is dimensioned to withstand the temperature rise.
An experimentally validated multi-physical finite element model is used to perform simulations by varying the mechanical load to explore the performance of the actuator topology. The obtained currents, induced forces, stresses, and accelerations of the armature are then analyzed in an attempt to develop scaling techniques that can predict for example velocity and efficiency. Finally, the results of the scaling techniques are presented and compared to each other.
In this paper, a multi-physics computational tool has been developed to accurately model and build high performance ultra-fast actuators. The research methodology is based on a finite element method model coupled with a circuit model. Electromagnetic, thermal, mechanical, and algebraic equations are implemented in Comsol Multiphysics and verified with laboratory experiments of a built prototype. A simplified model is preferred as long as its underlying assumptions hold. However, in the presence of large current and force densities, nonlinearities such as deformations may occur. Such phenomena can only be captured by the use of the developed comprehensive multi-physics simulation model. Although this model is computationally demanding, it was shown to have an accuracy of at least 95% when compared with experiments.
In this paper, a simulation of an ultra-fast electromechanical drive was performed by using a two-dimensional axi-symmetric multi-physical finite element model. The aim of this paper is to primarily show that the following model can be used to simulate and design those actuators with good accuracy, secondly, to study the behavior and sensitivity of the system and thirdly, to demonstrate the potential of the model for industrial applications. The simulation model is coupled to a circuit and solves for the electro-magnetic, thermal, and mechanical dynamics utilizing a moving mesh. The actuator under study is composed of a spiral-shaped coil and a disk-shaped 3mm thick copper armature on top. Two numerical studies of such an actuator powered by 2640 J capacitor banks were performed. It is shown that forces up to 38 kN can be achieved in the range of 200 μs. To add credibility, a benchmark prototype was built to validate this experimentally with the use of a high speed camera and image motion analysis.
In recent years, ultra-fast actuators have become key elements in the development of high voltage direct current (HVDC) breakers for multiterminal grids which represent a huge progress in modern power transmission [1]. After fulfilling their operation these actuators need to be decelerated using controllable forces to avoid deforming vital components incorporated in the system. In this paper, a dedicated damper is proposed based on a magnet array that induces eddy currents in a composite metal tube resulting in an efficient braking response. Several topologies are investigated by simulations and experiments. The theory behind eddy current damping is explained in [2]. The main requirements for such dampers are reliability, robustness, and ease of construction. The expected durability of these kind of dampers is longer than the breaker itself which guarantees extremely good reliability within HVDC systems.
High-voltage direct current (HVdc) breakers are the key components in the realization of multiterminal HVdc grids. In the presence of fault current, these breakers should be able to deliver impulsive forces to swiftly open the metallic contacts. After the acceleration phase, the moving armature should be decelerated using controllable forces to avoid plastically deforming fragile components integrated in the system. In this paper, finite-element method-based simulation models, complimented with small-scale and large-scale experimental prototypes, were utilized to benchmark different damping topologies. It was found that a Halbach-based configuration can deliver a damping force that is almost two and a half times larger than its sequel. Its sequel, composed of vertically stacked oppositely oriented magnets, is easier to assemble and is also capable of generating a considerable damping force. Finally, it has been shown that both these schemes, inserted in a composite tube, have a potential to be used as dampers in HVdc breakers.
Exciting nonlinear circuits with modulated signals will generate crossmodulation frequencies. In ordinary simulation techniques this makes time stepping methods a necessity, In the case of EMC-testing where the carrier frequency is as high as 2 GHz and the modulating frequency only 1 kHz, the simulation time will be extremely long. This paper presents Time Domain Frequency Separation (TDFS), a method,where the high and low frequency behavior is calculated separately to decrease the computation time. In a simulation made with Saber the calculation took approximately fourteen hours, and with TDFS in Matlab two minutes.
In calculations and simulations regarding magnetic materials, it is important to have a have an accurate model of the hysteresis loop. The major loop, in particular, is used in many simulations. However, it is generally not possible to measure the true major loop, and it must therefore be approximated using a minor loop. There are several methods available for approximating magnetization curves, but they are primarily designed for paramagnetic materials, and are poorly suited to the highly grain-oriented steels used in modern transformers. Therefore, we propose two expressions for approximating the magnetization curves of grain-oriented silicon-iron steels. Both methods give close agreement with measurements and can be extrapolated to in order to describe the major loop.
The magnetic properties of a power transformer core are generally held to be quite similar to those of the core steel itself. Due mainly to it being rare and practically difficult to acquire a transformer for testing and verification, testing of large units is usually only performed concerning no-load losses. However, other parts of the magnetic hysteresis loop are more sensitive to variations in material and geometry and could be used for more detailed diagnostics. This paper shows that measurements of magnetic hysteresis can be performed with good results on large power transformers. Methods to compensate for capacitive currents and to calculate the effective magnetic length of the core are shown and the results are compared to standard material measurements. The results show good agreement with Epstein frame measurements on annealed samples.
In many transformer applications, it is necessary to have a core magnetization model that takes into account both magnetic and electrical effects. This becomes particularly important in three-phase transformers, where the zero-sequence impedance is generally high, and therefore affects the magnetization very strongly. In this paper, we demonstrate a time-step topological simulation method that uses a lumped-element approach to accurately model both the electrical and magnetic circuits. The simulation method is independent of the used hysteresis model. In this paper, a hysteresis model based on the first-order reversal-curve has been used.
This book contains all information regarding magnetism and magnetic materials that an electrical engineer needs to know to be able to understand and design magnetic devices. The handbook comprises chapters comprising basic electromagnetism, basic quantum mechanics, ferromagnetism, magnetic materials, magnetic material characterization, modeling of magnetic materials, and magnetic design. A comprehensive description of the physical origin of magnetism of materials is given chapter two and a thorough review of the physics behind ferromagnetism is given in chapter three. All chapters are written in a textbook fashion such that they can easily be assimilated separately. The book gathers in an understandable the multidisciplinary topic of magnetism and magnetic materials in way that it can serve as a comprehensive introduction to engineers that considers use of magnetic materials in their designs. The book covers all major modeling techniques of magnetic materials including the well-known Presiach, Jiles-Atherton and lag models. General magnetic design approaches including major and new design tools also are presented. The book also serves as a guideline regarding the choice of feasible materials in specific applications regarding both soft and hard magnetic materials with an inventory of alternatives to electrical steel. Relevant performance criteria then are given such that appropriate materials can be selected. The final chapter offers a list of current electrical steel and magnetic material suppliers.
Statistical mechanics is employed to derive magnetic and magnetostrictive bulk properties of soft magnetic materials dominated by domain wall motion. The material is divided into an ensemble of volumes with uniform magnetization, each one much smaller than a typical domain but much larger than one atom. The energy for such a volume element is assumed to consist of an internal energy: the interaction with the external magnetic field and a mean field interaction with the surrounding medium. The volume fractions of discrete mesoscopic magnetizations in the medium are derived from Boltzmann statistics. The bulk magnetization is the weighted average of the mesoscopic values. Magnetostriction is treated in a similar way. Calculations are compared with two-dimensional measurements on grain-oriented silicon #x2013;iron. #xa9; 1999 American Institute of Physics.
A model for medium frequency power transformers is suggested, The model is able to treat arbitrary topologies by use of Cauer circuits for core and windings. An implementation example is given for a 1-phase transformer with turn ratio 1:1. The model can be extended to include winding insulation materials by use of lumped capacitive and resistive circuit elements.