Model Predictive Torque Control Analysis for Small Agricultural Machinery Traction
摘要
This paper discusses the use of induction motors in traction and then shows direct torque control techniques suitable for electrical traction applications, comparing the classic direct torque control and the established predictive torque control strategies through experiments in a test bench composed of an induction motor attached to a load. The dynamic and steady-state response of torque, flux and speed, as well as the total harmonic distortion of stator currents, the computational burden and the impact of the parametric mismatches in stator resistance and inductance, is analyzed. Results show that using discrete space vector modulation has the best performance but shows high computational burden. MPTC with duty cycle calculation shows signs of sub-optimizations, and the fixed frequency does not directly translate to better THD. The classic PTC has a solid performance and computational cost, being the best choice for direct torque control. The controllers show robustness in the range of parameter mismatch tested for both stator resistance and mutual inductance, showing low impact on the performance, but a robust flux estimator for induction machines is highly recommended independent on the control used.