Metaheuristic optimization of predictive torque control for induction motor control

Predictive torque control (PTC) is a high-performance control method of induction motors (IMs), which is still open to research. It provides many advantages over mature control techniques, such as straightforward imple-mentation, the ability to handle nonlinearities, easy inclusion of additional control objectives, and modulator-free structure. However, it has problems with the selection of weighting factors (WFs) involved in the cost function in PTC. In conventional PTC, these WFs are generally selected by the trial-and-error method. Also, a few studies optimize these WFs with a multi-objective optimization algorithm using both torque and flux errors. In this paper, the WF associated with the flux component is optimized by a genetic algorithm over the speed errors only. The optimized PTC is verified by simulation studies considering different operating conditions. Finally, good control performance has been achieved.