Thesis or Dissertation 電磁界解析を用いた電気機器の最適設計に関する研究

佐藤, 孝洋

Electromagnetic machines and devices, such as motors and inductors, are one of the important components in various electronics products. There is, thus, a great demand for high efficiency the electromagnetic machines and devices. For this reason, design optimization techniques have been attracted attention in order to maximize the performance under some constraints and specifications. Because the devices and machines are sometimes optimized by the try and errors based on the designer's experiences, numerical optimization method which is based on the finite element(FE) analysis are necessary. In this work, two optimization methods are proposed to advance the numerical optimization method of electromagnetic devices. First, parameter optimization based on the nonconforming voxel FEM is proposed, by which FE meshes for 3D shapes can be drastically generated with low computational cost. The inductor for boost circuit is optimized using the proposed method. Next, a new on/off-based topology optimization method is also proposed. Topology optimization,in which the topology of the shapes can be changes, can seek for the optimum shapes with high shape representation ability, and is thus effective to find innovative and initial conceptual designs. Although the conventional on/off-based method often finds complicated shapes which is not suitable for manufacture, a new method is proposed to overcome this problem. The proposed method is applied to the optimization of the synchronous motor. Moreover, as an application of the FE analysis, a new vibration energy harvester is designed. Vibration energy harvester, which transforms the vibration energy into electric energy through magnetic induction, has been receiving attentions as a power source of wireless sensor ICs. In this work, based on the FE analysis, a harvester with both higher output and wider operational bandwidth is designed.
xiv, 197p
Hokkaido University(北海道大学). 博士(情報科学)

Number of accesses :  

Other information