Yuki Shimizu, Motor Drive System Research Group, Electrical and Information Systems, Electrical Engineering and Information Science, Graduate School of Engineering, Osaka Prefecture University
Finite element analysis needs to be performed repeatedly to design the optimal motor shape, and therefore, the computation time becomes longer. In order to solve this problem, “surrogate models” which accurately predict the motor characteristics using machine learning have been attracting attention, and it is possible to perform the optimal design without using finite element analysis. We proposed a learning method for accurately predicting the speed-torque characteristics of IPMSMs for automotive applications. In this lecture, we explain the learning method using JMAG and optimization results.