Development of High-speed and High-accuracy Design Optimization Method with Precise Behavior of Space-time Electromagnetic Field in Electrical Machines

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K04357
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 21010:Power engineering-related
• Research Institution: Hosei University
• Principal Investigator: Okamoto Yoshifumi 法政大学, 理工学部, 教授 (40415112)
• Co-Investigator(Kenkyū-buntansha): 若尾 真治 早稲田大学, 理工学術院, 教授 (70257210)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 電磁界数値解析 / トポロジー最適化 / 磁気特性測定 / 深層学習 / 感度解析 / 随伴変数法 / 永久磁石同期電動機

Outline of Final Research Achievements

Because the half of energy in Japan is consumed by the motor, the research target was designated to the first topic “Development of innovative design optimization method using the numerical electromagnetic field analysis coupled with the topology optimization” and the second one “The elapsed time reduction for the measurement of the magnetic property which is required on the design optimization method”. In the first topic, the sensitivity calculation method, with the consideration of strongly coupled analysis between electromagnetic field and three-phase electric circuit, the rotor rotation, the magnetic nonlinearity of magnetic core, and steady state, was proposed. Furthermore, it was successfully applied to the topology optimization. In the second topic, the elapsed time reduction of being sinusoidal wave form derived from B-coil was achieved.

Free Research Field

電磁気工学

Academic Significance and Societal Importance of the Research Achievements

三相交流回路で駆動する永久磁石同期電動機の設計最適化において，回転子の回転と磁界・回路双方の動特性を考慮して得られた感度を使用した手法は世界初の技術である．また，実際の時空間現象を緻密に考慮できるため，設計現場における高精度設計に役立つ技術となり得る．また，モータ鉄芯の磁気特性測定では，波形制御に相応の反復回数・時間が必要であったが，Deep Neural Networkの導入により，大幅に高速化することができた．設計最適化への導入だけではなく，新たな材料開発のためにも大いに役立つ技術であると考える．