Prediction Model of Torque Control Parameters Considering Temperature Dependency of IPMSM

Abstract

This study proposes a computationally efficient torque estimation model for the interior permanent magnet synchronous motor (IPMSM), considering the temperature-dependent magnetic properties of the permanent magnet (PM). It is well known that the magnetic flux of the IPMSM is caused by both the PM and the current source in the stator winding. Hence, the magnetic torque depends on PM temperature as PM flux varies with temperature variation. Using finite element analysis (FEAs), the proposed model accounts for nonlinear characteristics, such as $d, q$ cross-coupling and saturation effects based on flux linkage mapping. Moreover, the static FEA is conducted to achieve time-efficient computation, and the least number of required simulations is considered. An IPMSM designed for the target high-speed train traction motor is employed to validate the proposed method. Validation is conducted by comparing the results obtained from the proposed prediction model with the results from time-transient (TT) FEA at different PM temperatures. The results show a good agreement between the proposed method and the FEA results while significant reduction simulation time.