Cost functions in finite control set model predictive control of permanent magnet DC machine

Abstract

This paper presents a finite control set model predictive control strategy for the speed control of brushed permanent magnet DC machine. Main objective is to introduce and compare basic cost functions regarding speed control. Full order Kalman filter is used for the observation of the non-measured load torque and reduction of speed measurement noise. Influence of the proposed cost functions on the control quality is demonstrated by simulations. The performance of the best cost function is shown through experimental verification with a 250 W machine and laboratory prototype of H-bridge controlled by RT-LAB control platform.