Evaluation of Model Predictive Control for IPMSM using High-Fidelity Electro-thermal Model of Inverter for Electric Vehicle Applications

This paper presents a high-fidelity electro-thermal model of a half-bridge that consists of IGBTs and anti-parallel diodes. The model calculates and estimates the half-bridge voltages, currents, switching and conduction losses considering the operating temperature and current conditions. Moreover, this model is suitable for varying switching frequency operation. The electro-thermal model can be used as an evaluation tool to analyze the performance of control strategies for traction inverter from efficiency, temperature and component stress point of view. In this paper performance of Direct Torque Model Predictive Control (DTMPC) of an Interior Permanent Magnet Synchronous Motor (IPMSM) is evaluated in comparison with Indirect Field Oriented Control (IFOC) with sinusoidal pulse width modulation (PWM). The inverter model and the MPC are both implemented in C-Mex for rapid execution. The MPC algorithm implemented in this research tracks torque reference while using Maximum-Torque-Per-Ampere (MTPA) strategy and minimizing switching losses. Both control systems are able to follow the speed reference. The MPC shows a decrease in losses compared to IFOC when tested with low-speed and high-speed parts of the WLTC profile.