Tetsuji Matsuo

Optimization of Inductors Using Evolutionary Algorithms and Its Experimental Validation

This paper presents parameter and topology optimization of inductor shapes using evolutionary algorithms. The goal of the optimization is to reduce the size of inductors satisfying the specifications on inductance values under weak and strong bias-current conditions. The inductance values are computed from the finite-element (FE) method taking magnetic saturation into account. The result of the parameter optimization, which leads to significant reduction in the volume, is realized for test, and the dependence of inductance on bias currents is experimentally measured, which is shown to agree well with the computed values. Moreover, novel methods are introduced for topology optimization to obtain inductor shapes with homogeneous ferrite cores suitable for mass production.

Anisotropic Vector Hysteresis Model Using an Isotropic Vector Play Model

A simple anisotropic vector hysteresis model is developed to represent 2D weakly anisotropic vector hysteretic property. An isotropic vector play model is identified from azimuthally averaged vector property of anisotropic material. An anisotropy matrix is multiplied to the isotropic vector play model to represent anisotropy. Simulations for alternating and rotational flux conditions show that the anisotropy matrix improves the representation of vector hysteretic properties for a non-oriented silicon steel sheet.

Trial Test of Fully HTS Induction/Synchronous Machine for Next Generation Electric Vehicle

Our group has tried to introduce High Temperature Superconducting Technology to electric vehicle (EV) drive motor, and developed High Temperature Superconducting Induction-Synchronous Machine (HTS-ISM). In this paper, 20 kW class HTS-ISM were designed and manufactured with DI-BSCCO wires. First, we developed and tested partial superconducting HTS-ISM (HTS rotor windings with copper stator) for clarification of the performance. We could observe the efficiency at 92.3% even for the light load. Furthermore, we successfully fabricated and conducted no-load test for fully superconducting HTS-ISM. It was verified that stable synchronous rotation at 1200 rpm was succeeded, which is the world-first success of achieving the rotation at over 1000 rpm for fully superconducting motor. Our results are one of the big steps for realizing the HTS-EV.

Isotropic vector hysteresis represented by superposition of stop hysteron models

The present paper first shows that the scalar stop hysteron model has the property of equal vertical chords for back-and-forth input variations of the same amplitude. This property leads to an identification method of the scalar model. Secondly, identification methods are developed for the 3-D and 2-D isotropic vector models that are constructed by the superposition of scalar stop hysteron models. Numerical simulations show that these methods identify the scalar and vector models satisfactorily.

Two Types of Isotropic Vector Play Models and Their Rotational Hysteresis Losses

This paper presents a comparison of two types of isotropic vector play models and their generalized models. The first vector model is represented by a superposition of scalar play models. The other type is given by a geometrical vectorization of play hysteron. The rotational hysteresis losses of both models are discussed. A method is proposed to adjust the simulated rotational hysteresis loss to a measured loss.

Eddy-current analysis using vector hysteresis models with play and stop hysterons

Vector hysteresis models are applied to an eddy-current analysis. The vector hysteresis models are composed by play hysterons and stop hysterons. The eddy-current analysis shows that both play and stop hysteron models can effectively describe isotropic vector hysteretic behavior. The stop hysteron model is more efficient in analyses using the magnetic vector potential than the play model because the stop model can give the magnetic field from the magnetic flux density without an iteration process.

Stop model with input-dependent shape function and its identification methods

We propose an input-dependent shape function for the stop model to remove its property of equal vertical chords regardless of dc bias. We compare several identification methods for the stop model and show that the input-dependent shape function improves the ability to represent the stop model effectively. A least-squares identification method using both symmetric and asymmetric B-H loops achieves the most precise representation of hysteretic characteristics for a silicon steel sheet. However, if only symmetric loops are available for identification, a newly proposed identification method gives more accurate representation than other methods including the least-squares method.

Application of stop and play models to the representation of magnetic characteristics of silicon steel sheet

The stop and play models are applied to the representation of the hysteretic characteristics of a silicon steel sheet. Both models are used to provide hysteretic functions from B to H. The play model identified from symmetric B-H loops describes the hysteretic characteristics much more accurately than does the stop model. Combining the two models achieves a more accurate representation than either of the two models alone.

Dynamic and Anisotropic Vector Hysteresis Model Based on Isotropic Vector Play Model for Nonoriented Silicon Steel Sheet

A phenomenological vector hysteresis model is developed to represent dynamic and weakly anisotropic vector hysteretic properties of nonoriented silicon steel sheets. The anisotropic vector hysteresis is represented by the multiplication of an anisotropy matrix and an isotropic vector play model. An eddy-current field vector given by classical eddy-current theory and an anomaly factor is added to the vector hysteresis model. The anomaly factor is determined to model the decrease in anomalous eddy-current loss under conditions of large rotational magnetic flux. The alternating and rotational properties simulated by the proposed model reconstruct the measured magnetic properties of nonoriented silicon steel sheets.

Space-time finite integration method for electromagnetic field computation

A finite integration method on a four-dimensional space-time grid is studied for the computation of electromagnetic wave propagation, where a non-uniform time-step distribution is naturally introduced. A dual grid based on the Hodge duality and the Lorentz metric is proposed to provide a simple constitutive equation for electromagnetic variables. An explicit time-marching scheme for a non-uniform space-time grid achieves a more efficient electromagnetic field computation than the conventional FDTD method.