Original scientific paper
https://doi.org/10.1080/00051144.2019.1614345
Modified predictive torque control method of induction machines for torque ripple reduction
Dubravko Krušelj
; Končar – Electronics and Informatics, Inc., Zagreb, Croatia
Damir Sumina
; University of Zagreb Faculty of Electrical Engineering and Computing, Zagreb, Croatia
Abstract
Direct torque control, model predictive control and field oriented control are control methods mostly used in high performance induction machine drives. In the direct torque control method, control variables are estimated from the stator variables, and the only parameter required is the stator resistance. The predictive torque control with horizon one has recently attracted much research attention but it requires the use of the induction machine speed, and both the stator and the rotor parameters, usually requires adjustment of the weighting factors, and has high computational burden. This paper proposes a modified predictive torque control method of induction machines. The estimated and predicted values are calculated from the stator variables, and the method uses the cost function without the weighting factor. When the two-level three-phase voltage source inverter is analyzed, it is shown that the predicted values should be calculated for three voltage vectors. The modified predictive torque control results in a better steady state performance regarding torque ripple in comparison with the conventional direct torque control and the predictive torque control methods. Simulation and experimental results for the main propulsion drive of the low-floor tram are presented in order to validate the effectiveness of the proposed method.
Keywords
Induction machine; traction; direct torque control; model predictive control; predictive torque control; torque ripple; stator flux ripple
Hrčak ID:
239787
URI
Publication date:
20.5.2019.
Visits: 964 *