Abstract
Today, a clear trend in electrification process has emerged in all areas to cope with carbon emissions. For this purpose, the widespread use of electric cars and wind energy conversion systems has increased the attention and importance of electric machines. To overcome limitations in mature control techniques, model predictive control (MPC) strategies have been proposed. Of these strategies, predictive torque control (PTC) has been well accepted in the control of electric machines. However, it suffers from the selection of weighting factors in the cost function. In this paper, the weighting factor associated with the flux error term is optimised by the non-dominated sorting genetic algorithm (NSGA-II) algorithm through torque and flux errors. The NSGA-II algorithm generates a set of optimal solutions called Pareto front solutions, and a possible solution must be selected from among the Pareto front solutions for use in the PTC strategy. Unlike the current literature, three decision-making methods are applied to the Pareto front solutions and the weighting factors selected by each method are tested under different operating conditions in terms of torque ripples, flux ripples, cur-rent harmonics and average switching frequencies. Finally, a decision-making method is recommended.
Weighting Factors Optimization of Model Predictive Torque Control of Induction Motor Using NSGA-II With TOPSIS Decision Making
