Conference Agenda

Metal object detection (MOD) is crucial to prevent wireless power transfer (WPT) systems from the loss of transfer efficiency and fire risk caused by metal object intrusion. The equivalent mathematical model of WPT system with metal object is first investigated. Then, a novel comb pattern sensing coil design is introduced for metal object detection. The induced voltage variation of sensing coil pair is used to detect the presence of metal object. Moreover, by selecting different active channels, the sensing area is adjusted for different size metal objects to avoid blind zone issue and increase detection sensitivity. A finite element method simulation verifies the efficacy and accuracy of the proposed MOD method. Finally, an experiment conducted with 8A transmitter coil current shows that metal objects with different sizes, from 30mm × 30mm to 60mm × 60mm, can be effectively detected.

This paper proposes a novel 2.5-D multi-phase topology optimization method using the Gaussian basis function for permanent magnet motors. The design region in the rotor is sliced into cylindrical layers to each of which two-dimensional topology optimization is performed so that the average torque is maximized while the torque ripple is suppressed as small as possible. The present topology optimization determines the rotor-core and magnet shapes as well as magnetization direction. It is shown that the optimized 2.5-D model whose structure changes in the axial direction has better torque performance than the 2-D model.

Imprecision and uncertainty are often inevitable and unavoidable in design process of Topology Optimization (TO). Therefore, Robust Optimization (RO) has become a significant concern in TO problem. To eliminate the extremely high computational burden and the checkboard pattern problem of existing RO methodology, a novel methodology based on α-min-cut for TO is firstly proposed. The proposed method finds the optimal direction under uncertainty through finding α-min-cut in the transferred determinisitic network. The numerical result of a magnetic actuator with multiple material has validated the proposed method.

This paper investigates a method to reduce iron loss of Internal Permanent Magnet Synchronous Motors (IPMSM) using optimized the voltage waveform of PWM modulation. The iron loss of motor is affected by harmonic component of magnetic flux density due to inverter switching. In this paper, we optimized switching waveform of voltage in order to reduce iron loss. In optimization process, it is necessary to solve Finite Element Analysis (FEA) large number of times. Therefore, we used simplified rotor model in order to reduce calculation load. This enabled to derive optimized voltage waveform in realistic time.

In this paper, a multi-objective topology optimization (MOTO) methodology using Non-dominated Sorting Genetic Algorithm II (NSGAII) and convolutional neural network (CNN) is proposed. The original NSGAII is improved to achieve better global search ability and uniform distribution of Pareto solutions. And CNN is applied as a surrogate model for finite element analysis. The framework of the proposed methodology is elaborated. To validate the proposed methodology, it is applied to the TO of an electromagnetic actuator. Numerical results demonstrate that computational cost of TO can be reduced without deteriorating the optimization quality.

This paper proposes interpolation multimodal optimization algorithm (IMOA) for robust optimal design of electric vehicle traction motor. Target model is interior permanent magnet motor and design objective is minimizing the torque ripple. The proposed algorithm can drastically reduce the number of function calls by interpolating problem regions with existing samples. Additionally, the IMOA can solve the multimodal optimization problem by using intensive searching strategy and diversification strategy. At the end of the algorithm, robustness test is conducted, and robust optimal solution is determined considering the worst-case condition on the uncertainty band. The superior performance of the IMOA is verified at two mathematical test functions.

Recently, the rapid design of electrical machine is possible due to the development of computer performance and the development of finite element analysis tool and the optimization algorithm has made it easier to access the optimal design of electrical machines. Nevertheless, in the problem of considering manufacturing uncertainty, many analytical data are needed and it is difficult to overcome with finite element analysis alone. Practically, manufactured electric machines can not expect high performance for all products due to material deviations and dimensional tolerances. We need to be able to come up with a robust solution with little variability under conditions that do not violate the constraints. Therefore, we must consider the problems that occur while manufacturing to perform an optimization algorithm. However, considering the manufacturing tolerances along with many of the design variables of IPMSM requires a lot of time and cost. Therefore, this paper proposes the application of a multiple-regression method to improve the performance of the conventional particle-swarm optimization algorithm and solve the robust design-optimization problem.

This paper shows that topology optimization can be formulated as a quadratic unconstrained binary optimization (QUBO) problem which can be solved by Digital Annealer with massive parallelization and also quantum computers. The distribution of magnetized cells is determined using Digital Annealer while the magnetization is determined by finite element analysis. The magnetic core structure of a magnetic shielding system is shown to be successfully optimized using the proposed method with reduced number of field computations.

This study proposes an optimal design method for SPMSM based on NSGA-II and analytical method for a specific power. To consider efficiency, the core loss and eddy current losses were calculated through analytical approach to evaluate the motor efficiency. The proposed method was used to design a 6.3-kW SPMSM for a direct drive to test its validity. The optimal model was selected through Pareto-frontier, and the results were compared with the FEM results.

This paper proposes a deep learning-based sizing process of SPMSM. The proposed process employs a deep neural network (DNN) to consider the effect of axial leakage flux using 3-D finite element analysis (FEA) based motor parameters. The 3-D FEA-based motor parameters can be derived using size-related variables, 2-D FEA based motor parameters, and trained DNN. Then, the characteristics of the motor according to the size considering the axial leakage flux can be calculated. However, the process for acquiring the data for DNN is inefficient because of the time-consuming FEA. Therefore, we propose a method for data augmentation using electromagnetic domain knowledge. Using the relationship between axial leakage flux, motor parameters, and axial length, the motor parameters according to axial length can be generated even with the motor parameters of unit axial length. Using the proposed process, it is possible to accurately predict the motor characteristics according to the size that satisfies the required specifications. To verify to proposed sizing process, motors of various size will be verified through simulations and experiments.

In this paper subdivided novel kriging-assisted multi objective optimization (SKAMOO) is proposed for the optimal design of interior permanent magnet synchronous motor (IPMSM) for electric vehicles. SKAMOO remarkably reduces computation cost using kriging model. However, the kriging model is possible to different to real model. To complement this problem, sample is added to around the place where there are the most change in the kriging model, increasing the accuracy of surrogate model. Sample is added considering all objective functions, not kriging changes to one objective function. Solution is directly added in the objective area using kriging model. Kriging grids are subdivided to improve algorithm performance. The small kriging grid takes a lot of time, so instead of dividing it small from the beginning, the process is applied to the latter part of the algorithm. The SKAMOO is demonstrated superiority compared to the NSGA-II and the MOPSO. Finally, the proposed algorithm is applied to the optimal design of IPMSM.

Recently, magnetic levitation systems have been applied and studied in various industrial fields. In particular, in-track-type magnetic levitation conveyor systems are being actively studied since it can effectively minimize electromagnetic effects in processes that require a high-clean environment. In this type of system, diverse and multiple sensors are structurally required such that the control performance of an integrated system is primarily governed by the slowest measuring sensor. In this paper, multi-sensor fusion compensator is proposed in order to integrate the outputs obtained from various sensors into one output with the single fastest time rate. Since the state of the system is estimated at the fast time rate, the optimal controller also guarantees fast performance and stability. The computation of electromagnetic fields and control performance of the considered superconducting hybrid system were analyzed by computer simulation, based on finite element method.

This paper presents an automatic design of permanent magnet motors using Monte-Carlo tree search. Using this method, different motor structures, with different number of poles and magnet configuration, for example, can be simultaneously considered. At the leaf of the tree, topology optimization is performed to find optimal geometry. It is shown that the proposed method is promising to realize the automatic motor design.

Thrust ripple caused by the end effect detent force is an inherent problem of permanent magnet linear synchronous motors (PMLSMs). In this paper, a topology optimization method that is used to minimize the thrust ripple for PMLSMs is proposed. This method designs the topology of two mover ends by ON/OFF approach with evolutionary algorithms. Two different PMLSMs are adopted as numerical examples to verify the effectiveness of this method. The finite element analysis simulations show that the proposed technique has an excellent effect on the reduction of detent force and thrust ripple in PMLSMs. Moreover, the present method will not affect the output power of the PMLSMs.

This paper proposes an electrode shape optimization strategy for a bipolar space-charge system. The continuum sensitivity formula for a bipolar space-charge system is used to determine the extremum of objective functions with a state variable constraint. The spacecharge distribution is computed using Poisson’s equation and the charge transport equation with charge injection boundary condition. The geometric change of the design parameter is determined by the velocity field calculated from the continuum sensitivity analysis and is expressed by the level set method. The application model is tested to demonstrate the feasibility and utility of the proposed method.

This paper presents topology optimization of a permanent magnet synchronous motor considering its control system. The rotor shape is determined to minimize the iron loss generated in the stator and rotor cores. In the optimization, the driving currents are generated by the control system which is modeled by d- and q-axis fluxes considering the spatial and time harmonics. It is shown that the iron loss of the optimized model is reduced by 25.2% compared with that of the conventional model under the assumed speed control.

This paper presents a multi-material topology optimization method based on the ON/OFF method. This method can be used to determine the optimal topology of electric machines that are composed of various materials, such as iron, magnetic, and non-magnetic material. The present method is applied to the optimization of an interior permanent magnet motor to redesign the rotor. The optimization results show that the average torque and torque ripple are improved with lesser consumption of the magnet material. This method can provide guidance for the preliminary shape design of magnet and flux barrier in the permanent magnet motors.