Shape optimization of switched reluctance motors considering complicate shape features and electric parameters

Abstract

Considering numbers of the complicate shape features and electric parameters of switched reluctance motors, the development of shape optimization methods which can minimize objective function (e.g., motor weight) while satisfying constraints (e.g., average torque, torque fluctuation, maximum current) was a major issue in research field of motors. Our goal is to propose the shape optimization framework which can consider both complicate shape features and electric parameters through the gradient-based optimization, thereby developing high performance and low cost switched reluctance motors. In this study, we proposed new design strategies (e.g., a circular hole, pole-notch representation using spline nodes, and conforming coils) for the complicate shape design. Also, in order to verify the new shape optimization framework, we performed the shape optimization in both geometric and electric domains (e.g., the rotor/stator of SRMs and the Voltage on-off angles) simultaneously. In order to enhance the product competitiveness, the optimization formulation was set as minimizing the motor weight while satisfying the adequate average torque and torque ripple levels. In order to prevent severe thermal loss, the maximum current flowing the winding coils was also limited under the proper value. In this research, the nonlinear magnetic analysis was performed by using ANSYS Maxwell which is a finite element analysis software, and the 3-phase torque was analytically derived by using a virtual work principle. As results, the optimal shape of the switched reluctance motors having low weight was determined through the proposed shape optimization while satisfying the desired average torque, torque ripple, maximum current, and other geometric constraints.