25 March 2019 in IEEE Transactions on Magnetics
The computational cost of topology optimization based on the stochastic algorithm is shown to be greatly reduced by deep learning. In the learning phase, the cross-sectional image of an interior permanent magnet motor, represented in RGB, is used to train a convolutional neural network (CNN) to infer the torque properties. In the optimization phase, all the individuals are approximately evaluated by the trained CNN, while finite element analysis for accurate evaluation is performed only for a limited number of individuals. It is numerically shown that the computational cost for the topology optimization can be reduced without the loss of optimization quality.
23 April 2015 in IEEE Transactions on Magnetics
This paper presents a topology optimization method for multimaterial models based on the normalized Gaussian network. In this method, one can determine optimal shapes of machines that are composed of various materials, such as iron, magnetic, and non-magnetic material. The present method is applied to the optimization of the average torque for interior permanent magnet motor to determine the distributions of magnetic core and flux barrier as well as magnets. The optimization results show that average torque can be improved using as small amount of magnet as possible. In addition, the characteristic of the present method is discussed in detail.
1 May 1999 in IEEE Transactions on Magnetics
12 March 2019 in IEEE Transactions on Magnetics
This paper presents the fast topology optimization methods for rotating machines based on deep learning. The cross-sectional image of electric motors and their performances obtained during a multi-objective topology optimization based on the finite-element method and genetic algorithm (GA) is used for training of the convolutional neural network (CNN). Two different approaches are proposed: 1) CNN trained by preliminary optimization with a small population for GA is used for the main optimization with a large population and 2) CNN is used for screening of torque performances in the optimization with respect to the motor efficiency.
17 August 2016 in IEEE Transactions on Magnetics
Native application of finite-element method (FEM) to the analysis of skin and proximity effects in multi-turn coils results in large equation systems, whose solution needs long computational time. This paper proposes a semi-analytical approach to overcome this problem. For the analysis of the proximity effect, the complex permeability of a round conducting wire immersed in uniform time-harmonic magnetic fields is represented in a closed form. Then, the homogenized complex permeability over the cross section of the multi-turn coil is analytically evaluated using the Ollendorff formula. The magnetoquasistatic problem is thus replaced by the magnetostatic one, in which the multi-turn coil is treated as a uniform material with the homogenized complex permeability. The skin effect is taken into consideration by introducing the corresponding impedance in the circuit equation. The proposed method is shown to give the impedance of multi-turn coils, which is in good agreement with that obtained by the conventional FEM as well as experiments.
16 May 2005 in IEEE Transactions on Magnetics
This paper proposes the real-coded clonal selection algorithm (RCSA) for use in electromagnetic design optimization. Some features of the algorithm, such as the number of clones, mutation range, and the fraction of the population selected each generation are discussed. The TEAM Workshop problem 22 is investigated, in order to illustrate the performance of the algorithm in a real electromagnetic problem. The results obtained, a set of optimal solutions representing a broader range of options for the designer, are compared with those achieved by a genetic algorithm, showing the efficiency of the RCSA in practical optimization problems.