Search

Purpose The purpose of this paper is to solve efficiently the topology optimization (TO) in time domain problem with magnetic nonlinearity requiring a large-scale finite element mesh. As an actual application model, the proposed method is applied to induction heating apparatus. Design/methodology/approach To achieve TO with efficient computation time, a multistage topology is proposed. This method can derive the optimum structure by repeatedly reducing the design domain and regenerating the finite element mesh. Findings It was clarified that the structure derived from proposed method can be similar to the structure derived from the conventional method, and that the computation time can be made more efficient by parameter tuning of the frequency and volume constraint value. In addition, as a time domain induction heating apparatus problem of an actual application model, an optimum topology considering magnetic nonlinearity was derived from the proposed method. Originality/value Whereas the entire design domain must be filled with small triangles in the conventional TO method, the proposed method requires finer mesh division of only the stepwise-reduced design domain. Therefore, the mesh scale is reduced, and there is a possibility that the computation time for TO can be shortened.

Because a synchronous reluctance motor (SynRM) does not require permanent magnets, its industrial applications focus on lowering manufacturing costs. However, the SynRM has disadvantages. It experiences a lower average torque and oscillations derived from torque ripple compared to a permanent magnet-based motor. Therefore, the improvement of torque characteristics is investigated using topology optimization (TO) method which allows a greater variety of rotor structure than the shape optimization technique and does not require the concrete initial structure. Although the level set method is well known among TO methods, the smoothed-heaviside-function-based TO implemented by method of moving asymptotes is more suitable due to the feasibility of implementing the physical structure and the convergence speed. In this paper, the reasonable structure of rotor core in SynRM is investigated using TO method to improve the torque characteristic.

The axle counter as a train detector is one of the important railway signaling devices for accurate and safe train control systems. This paper deals with a design optimization of the axle counter by large-scale magnetic field computations. To efficiently reduce the computational burden in the optimization, we accurately estimate the signaling level by applying infinite edge elements to an extensive air domain, which results in a practical design approach within an acceptable CPU-time.

This study derives a high performance flux barrier for an interior permanent magnet (IPM) motor using binary-based topology optimization (TO). TO has a higher degree of optimization than size or shape optimization. An IPM motor has many design parameters including the current phase angle, core shape of rotor and stator, and magnet position and shape. The flux barrier in an IPM motor plays a particularly important role, influencing several characteristics. We apply TO, using a multistep genetic algorithm, to the rotor core in order to determine a flux barrier with the highest torque performance.

This paper derives an effective shape of the flux barrier in an interior permanent magnet (IPM) motor. IPM motors generally have many design parameters such as the current phase angle, shape of the rotor and stator's iron core, and shape and position of the magnet. The flux barrier plays an important role in controlling torque characteristics. We apply topology optimization (TO) to the rotor core using a multistep genetic algorithm to determine an effective flux barrier. Furthermore, we extend the TO to combinatorial optimization for considering its effect on the current phase angle. Thus, a reasonable flux barrier with an optimal phase angle is determined.

This paper deals with a large-scale eddy-current analysis of the constructions in railway system from the view point of electromagnetic noise assessment. To reduce computational costs with the calculation accuracy kept, we develop some numerical methods and demonstrate their effectiveness using a practical large-scale model.

Against a background of environmental problems and energy issues, it is expected that PV systems will be introduced rapidly and connected with power grids on a large scale in the future. For this reason, concerns about how PV power generation will affect supply and demand adjustment in electric power in the future arise, and the technique of correctly grasping PV power generation becomes increasingly important. PV power generation depends on solar irradiance, module temperature, and solar spectral irradiance. Solar spectral irradiance is the distribution of the light intensity for every wavelength. As the spectral sensitivity of solar cells depends on the kind of solar cell, it becomes important for an exact grasp of PV power generation. But the preparation of solar spectral irradiance data is not easy because observational instruments for solar spectral irradiance are expensive. Against this background, in this paper, we propose a new method based on statistical pattern recognition for estimating the spectrum center, which is a representative index of solar spectral irradiance. Some numerical examples obtained by the proposed method are also presented. © 2013 Wiley Periodicals, Inc. Electr Eng Jpn, 184(1): 10–18, 2013; Published online in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/eej.22393

The axle counter which detects the position of the train is important railway signal device. This paper deals with a multi-objective optimization of the axle counter by 3D magnetic field analysis. In the optimization, we newly propose the introduction of the conductor plate and carry out its design optimization. It is revealed that the shape optimization of conductor plate enables us to well improve the detection performance of the axle counter from the multi-objective point of view.

Topology optimization (TO) is an attractive design tools for electrical machinery. In particular, the level-set-based TO has been investigated enthusiastically. In this paper, a level-set-based topology optimization method supported by the method of moving asymptotes is proposed to realize fast convergence. The performance of the proposed method is verified in comparison with the conventional level-set-based method, which is composed of both gradient-based-correction and regularization.

This paper investigates the parallelization of the time-periodic finite-element method in nonlinear magnetic field analyses of rotating machines. The developed method, which can obtain the steady state solutions directly, provides large granularity even in the small-scale problems compared with the ordinary parallel FEM based on the domain decomposition approach. Furthermore, we apply the parallel TPFEM to analyses of induction motors which have different time periodicities in stator and rotor regions due to the slip. Numerical results verify the effectiveness of the developed method.

With a background of environmental problems and energy issues, it is expected that PV systems will be introduced rapidly and connected with power grids on a large scale in the future. For this reason, the concern to which PV power generation will affect supply and demand adjustment in electric power in the future arises and the technique of correctly grasping the PV power generation becomes increasingly important. The PV power generation depends on solar irradiance, temperature of a module and solar spectral irradiance. Solar spectral irradiance is distribution of the strength of the light for every wavelength. As the spectrum sensitivity of solar cell depends on kind of solar cell, it becomes important for exact grasp of PV power generation. Especially the preparation of solar spectral irradiance is, however, not easy because the observational instrument of solar spectral irradiance is expensive. With this background, in this paper, we propose a new method based on statistical pattern recognition for estimating the spectrum center which is representative index of solar spectral irradiance. Some numerical examples obtained by the proposed method are also presented.

This paper reports a large-scale eddy-current analysis of conductive frame of large-capacity inverter. The eddy-current loss obtained by the electromagnetic field analysis is used as a heat source for the thermal analysis. Then, the suitable configuration of the frame for the suppression of temperature increase is investigated and designed. The hybrid finite-element and boundary-element (FE-BE) method is very suitable for the analysis of a complicated-shaped frame because it does not require mesh division for a free space and can easily treat eddy current. For the reduction of large computational costs, the fast multipole method (FMM) is introduced. Furthermore, in order to improve the convergence characteristic of iterative methods for system matrix, we develop the application of the IDR(s) method to the minor iterative preconditioning technique (MIP) and evaluate the performance. Finally, some numerical results that demonstrate the effectiveness of the developed method are presented.

This paper investigates a starting procedure of rotational sensorless permanent magnet synchronous motor (PMSM) in the rotating condition, which is suitable for some applications with higher maximum rotor frequency and longer current sampling time. Instantaneous estimation of initial rotor position and frequency, right after gates start, is essential for restarting the operation of inverters. Existing estimation methods are not available for the aforementioned applications because of their higher maximum rotor frequency and longer current sampling time. To cope with this problem, we propose a novel initial rotor frequency estimation method. The proposed estimation method is verified by experiments using a 2-kW interior PMSM and a numerical simulation.

This paper proposes a novel robust optimization approach by combination of sensitivity analysis and sigma level estimation in multi-objective problems. In order to obtain robust Pareto-optima within acceptable computational costs, we actively reuse the already calculated data for the sensitivity analysis in multipoint search algorithm. The sigma level estimation is also utilized to select the significant solutions from a large number of Pareto-optima based on both the robustness and the performance. Some numerical results that verify the effectiveness of the proposed methods are presented.

This paper shows large-scale eddy-current analysis by a hybrid finite-element and boundary-element (FE-BE) method. We introduce the fast multipole method (FMM) into the two kinds of hybrid FE-BE formulations developed by ourselves. To achieve more reduction of CPU time, we propose a novel preconditioning technique suitable for the hybrid FE-BE methods with the FMM. Furthermore, we improve the FMM calculation process in the hybrid method using magnetic field intensity. Some numerical results that demonstrate the effectiveness of these approaches are also presented

This paper shows a large-scale analysis of surface charge in eddy-current problems, which is useful for designing an electric machine. We introduce the fast multipole method with the pseudoparticle method into the large-scale surface charge analysis and further propose an approach to dealing with the symmetry and periodicity of the analysis model by the concept of the method of images. Some numerical results, which will show the validity of the proposed method that calculation time is reduced while maintaining high precision, are presented.

To estimate the characteristics of electric machines that have conductors of complicated shape, it is effective to analyze the surface charge related to eddy current phenomena. However, the eddy current field is generally treated as a quasi-static field in which the displacement current is neglected and the electric field is not defined in the nonconductive region in the calculation process. Therefore, when we use only the A–ϕ finite element method (FEM) as a field analysis method, it is difficult to calculate the surface charge directly at the interface between conductive and nonconductive regions. In this paper, with this background, we propose a novel method of analysis of the surface charge based on both the A–ϕ FEM with edge elements and the integral equation. This approach enables us to precisely calculate the surface charge. Some numerical results that demonstrate the validity of the proposed method are also presented, such as the surface charge analysis of linear induction motors for evaluating the relationships among the surface charge, eddy current, and conductor shape. © 2003 Wiley Periodicals, Inc. Electr Eng Jpn, 145(1): 59–66, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/eej.10173

This paper investigates the permanent magnet overhang effect in brushless dc motor. The characteristics are analyzed according to the variation of symmetric overhang length and asymmetric overhang ratio. A three-dimensional equivalent magnetic circuit network (EMCN) method is used for the accurate and efficient analysis. From the results, it is quantitatively verified that the increase of overhang length enhances the linkage fluxes of the stator and the asymmetric overhang ratio has a large effect on the axial magnetic force.

In this paper, we propose a boundary structure for improving performances of Single-sided Linear Induction Motors (SLIMs). In such machines, solid-plate conductors and back irons are usually used for secondary stators throughout operating regions. However, to take account of deformation due to thermal expansions, secondary stators are constructed from some sections of limited length. Therefore, a propulsive force characteristic declines when the primary armature crosses the boundary between secondary sections. In this study, by adopting a three-dimensional finite element method, detailed phenomena at the boundary between secondary sections have been clarified from the viewpoint of propulsive force and secondary current distributions. Besides the validity of calculated results is verified by the measured results. Further, to suppress the aggravation of characteristic caused by the secondary structural transition, we connect the secondary sections electrically. Finally, the propulsive force characteristic has been improved in spite of the existence of the boundary.

In recent years the numerical design optimization of magnetic devices has been the focus of interest of many engineers. In designing the practical devices such as induction motors, we have to estimate many kinds of objective physical quantities in the optimization process: e. g. magnetic flux, eddy current, torque, etc. This makes it difficult to obtain the effective solution that achieves the high device performance boosts within an acceptable CPU time. Thus, some contrivances to the practical design optimization are indispensable according to the device characteristics.In this paper, with this background, we propose an effective approach to the design optimization of rotating machinery by using the numerical analysis combined with the analytical design methodology and physical considerations. To verify the proposed approach, some numerical results, i.e., the shape optimization of ventilation holes in an induction motor, are also presented. In the numerical example, our goal is to widen the ventilation hole area without making the influence of the rotor-yoke saturation more serious.

In this article, we investigate a means of extricating the position information involved in terminal voltages and currents to estimate the rotor position of a PM synchronous motor. One way to extract the position information is to construct the state observer based on the motor model including electrical and mechanical equations. Then, the stability of the observer is an important issue in providing accurate position information for the motor drive system. In order to stabilize the system, the gains of the observer have to be optimized. However, these schemes have some difficulties in determining the optimum gains of the observer under any operating condition because the electrical equations are nonlinear in such motor models. We present a new approach for constructing the observer to extract precise information from the rotor position, and its speed from the technical quantities. First, to overcome the aforementioned problem, we propose the control inputs that can eliminate the nonlinear term in the electrical equations and realize the linearization of the motor model. This permits use of linear control strategies to determine the characteristics of the system. Second, we perform the rotor-speed estimation by using the reduced-order observer based on the linearized model. Third, the rotor position is calculated by using the estimated rotor speed and then corrected by the d-axis current error. Thus, we realize the stable drive of the motor without position sensors by measuring only two line currents. Finally, the validity of the proposed control scheme is confirmed by experiments.

In information lines, copper wire cables have been replaced by optical fiber cables that cannot supply power to the optical network unit (ONU). Thus, in the case of disasters, communication function paralyses may take place. Therefore, the reliability of power supply to the ONU has become important in recent years. With this background, we have developed a PV system for the ONU. In the process of development, we have proposed a novel estimation method for PV systems, that enables us to easily grasp the relationship between the instruments in the system from an overall view by regarding the instruments as the black boxes and paying attentions only to the energy flows. The proposed method is useful in determining the suitable parameter arrangement of the PV system. Utilizing the method, we developed 100W-order PV systems and carried out the experimental operation for one year. The experimental results showed enough system performances to meet not only the design requirement but also the spatial constraint.

This paper describes the hybrid finite element (FE) and boundary element (BE) method for the eddy current problem including moving conductors. Electromagnetic phenomena due to the three-dimensional movement of conductors can be computed easily and naturally by the presented formulation. Its application to a simple model which has only one conductor is also given. >

An analytical approach for calculating magnetic fields and potentials of massive current coils is presented. The proposed approach enables accurate, fast,and systematic evaluation of magnetic field intensity, vector potential, and scaler potential due to massive currents by means of simple and similar algorithms. >

As usual, brushless dc motor (BLDCM) has the overhang structure of which permanent magnet (PM) length is longer than the stator’s stack length. Specially, in case of asymmetric overhang, we have to pay attention to the choice of suitable overhang ratio of the lower to the upper overhang length because of the effect of the ratio on the magnetic forces. In this article, the magnetic force characteristics of BLDCM with the asymmetric overhang and the cap of the rotor are analyzed. A three-dimensional equivalent magnetic circuit network (EMCN) method is used for the efficient analysis of the magnetic field. The magnetic forces and torque are calculated by Maxwell’s stress tensor. The variation of magnetic force characteristics is also analyzed according to the overhang ratio. From the results, when the overhang ratio is 0.38, the axial magnetic force characteristic could be improved remarkably. The validity of the analysis results is also clarified by comparison between calculated results and measured ones.

The authors consider a boundary-element method (BEM) using edge elements for 3-D eddy current problems. The eddy current region is formulated only by means of the electric field intensity E, while the free space is formulated only by the magnetic field intensity H. In each region, both unknown variables are expressed independently of each other. By physical consideration and the appropriate boundary-element discretization, these physical quantities can be connected on the interface with ease. >

This paper investigates the acceleration of convergence to a steady state in a transient analysis of rotating machines using the time-periodic explicit error correction (TP-EEC) method. The efficient method to deal with the movement of the rotor and different time-periodicity in the fixed and moving parts of the mesh is developed. Numerical results verify the effectiveness of the developed method.

This article describes a design optimization of an axially laminated rotor of the synchronous reluctance motor, using the optimization method combined with the magnetic field analysis. The magnetic flux distribution in the synchronous reluctance motor is distorted due to the complicated rotor configuration and the influence of magnetic saturation, which results in a large number of local minimum solutions. Therefore, taking account of the magnetic saturation by the finite element method, the authors propose a novel optimization approach by coupling the stochastic search method with the deterministic one. In the optimization process a suitable objective function for the stability of machine performances is also proposed. Finally some numerical results that demonstrate the validity of the proposed approach are presented.

For the speed sensor-less control in the ultra lower speed range, the mechanical simulator method is applied to cope with the problems due to the extremely lower induced voltage. In the method, however, the mechanical parameterspsila error, such as the error of the inertia, affects the starting performance. To reveal the conditions for the stable starting, we carry out the phase plane analysis under the condition that the actual inertia J changes from the set inertia Js. Consequently, we find out Js ges J is the condition for the stable acceleration. The derived condition is useful as a design index of the mechanical simulator. Additionally, to cope with the large variation of the inertia, we carry out the phase plane analysis of the IM simulator method which compensates the estimation error of the rotor frequency. The analysis reveals that the method enables the stable acceleration while the estimation error of speed is remained in the case of inertia variation. Furthermore, we propose the modified IM simulator method which eliminates the deviation. By the phase plane analysis and experimental tests, we confirm the effectiveness of modified method.

This article describes the shape optimization of detective coils for initial rotor position estimation in the drive system of permanent magnet synchronous motors. The characteristics of magnetic circuits in rotating machinery are very complicated. The magnetic flux distribution in the air gap is nonsinusoidal due to the influence of slots, magnets, and magnetic saturation, etc. By utilizing the slot ripple influence and taking the effects of both magnetic saturation and permanent magnets into consideration, an optimization of the detective coil by means of a genetic algorithm combined with a hybrid finite element and boundary element method is reported. This produces a global optimal solution in a shorter time.

The authors present a hybrid finite-element (FE) and boundary-element (BE) method using edge elements for an open-boundary 3-D electromagnetic problem. In this method, the modified magnetic vector potential A* and the magnetic field intensity H are adopted as the physical quantities. The appropriate BE discretization described makes it easy to couple the BE region with the FE region through the transformation of unknown variables. Several practical applications of this method to 3-D eddy current problems are presented. The calculated values agree well with the experimental results. >

This paper describes a novel approach to the design optimization of ventilation holes in induction motors. By investigating the relationship between the rotor-yoke configuration and the magnetic flux in the air-gap, the authors determine the desirable arrangement of ventilation holes. Some numerical results for a practical machine, which demonstrate the validity of the proposed approach, are also presented.

The authors describe a hybrid finite-element (FE) and boundary-element (BE) method for electromagnetic problems. To attain good accuracy, the magnetic field intensity H is used as much as possible as the physical quantity in the hybrid method. A novel boundary element discretization in the H formulation, which makes it easy to couple the BE region with the FE region by introducing the edge boundary element, is proposed. A practical application of this method to the 3-D eddy current problem is presented. The calculated values agree well with the experimental results. >

This paper describes a large-scale micromagnetic simulation by using the fast multipole method (FMM) specialized for uniform brick elements. The fast Fourier transform (FFT) is widely used to reduce computational costs of the demagnetizing field calculation. However, the FFT still requires operation counts of O(N log N), where N is the number of elements, which results in the huge computational costs in large-scale problems. To overcome the difficulties, we develop an O(N) approach based on the FMM. In a micromagnetic simulation, an analyzed region is usually subdivided into uniform elements. By making the best use of the periodic structure of uniformly distributed elements, the computational costs of the FMM can be reduced drastically. A large-scale micromagnetic simulation of a single-pole-type head demonstrates the effectiveness of the specialized FMM from the viewpoints of calculation time and memory requirements, compared with the FFT.

This article proposes an effective computational approach to design optimization of an outer-rotor type permanent magnet synchronous motor. As usual, because of the complicated rotor configuration and the complex magnetic saturation effects, it is difficult to design the lightweight permanent magnet synchronous motor structure that makes good use of reluctance torque within an acceptable CPU time. In this article, we adopt the finite element method as a magnetic field analysis method and the genetic algorithms as a search method. Furthermore, the response surface methodology, which enables us to evaluate the objective physical quantities in a much shorter time, is introduced into the above methods in the proposed approach. This optimization approach results in an overall increase in the optimization speed, that is, substantial CPU time reduction in comparison with the case of a conventional one. Some numerical results that demonstrate the validity of the proposed approach are also presented.

We propose an optimization approach based on the estimation of permeability distribution, which enables us to make an effective objective function of optimization methods for the efficient design of magnetic devices. By the proposed approach, we can improve magnetic device characteristics by evaluating the deviation in the permeability distribution of the device as an objective function. In this approach, we separate the iron areas and calculate the permeability of each block to consider the permeability distribution. To verify the proposed approach, some numerical results, i.e., the shape optimization of ventilation holes in an induction motor, which demonstrate the validity of the proposed approach, are also presented. The optimization concept based on the estimation of permeability distribution is effective and can be readily applied to the practical design optimization of various magnetic devices.

In a conventional field oriented control of an induction motor the rotor flux is assumed to be constant so that the torque is controlled by controlling the q-axis current only. In field-weakening region the rotor flux is decreased to achieve a very high rotor speed, as a consequence the available torque is reduced. On the other hand, to produce a maximum torque the rotor flux should be generated as much as possible. In this paper, a novel, simple field-weakening strategy, which provides a maximum torque capability for a sinusoidal PWM VSI-fed induction motor drive, is presented. The purpose of the proposed strategy is to add the absolute q-axis current to d-axis current reference so that proper flux and torque for any given torque-producing current reference are produced automatically. The maximum torque when it is required can be obtained over the entire field-weakening region without using machine parameters or special control strategy. A saturated sinusoidal PWM is used which limits the amplitude of the inverter voltage references. As an advantage, the effective output voltage as seen by the motor is boosted. Simulation and experimental results verify the performance of the proposed method.

This paper describes a large-scale magnetic field analysis by means of the hybrid finite element-boundary element (FE-BE) method. The hybrid FE-BE method is well-suited for solving open electromagnetic field problems that comprise movement, nonlinear media, and eddy current. In general, however, large memory and computational costs are required due to the dense blocks in the system matrix generated by the BE part of the hybrid formulation. In order to overcome the above difficulties, we introduce the fast multipole method (FMM) to the hybrid FE-BE formulation developed by ourselves. Furthermore, we propose a novel preconditioning technique suitable for the hybrid FE-BE method with the FMM. Some numerical results that demonstrate the effectiveness of the proposed approach are also presented. ©2007 Wiley Periodicals, Inc. Electr Eng Jpn, 162(1): 73–80, 2008; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/eej.20508