Predictive Torque Control of Three Phase Axial Flux Permanent Magnet Synchronous Machines

This paper presents a predictive strategy to control torque and flux of an axial flux permanent magnet (AFPM) machine. Unlike conventional direct torque control (DTC) for permanent magnet machines that only six actives voltage vectors of the inverter are used to control the torque and flux of the machine, in predictive torque control (PTC), zero voltage vectors are used to control too. Thus, the number of voltage vectors to control AFPM increases and leads to faster dynamic torque response and lower ripples of torque and flux. In predictive torque control presented in this paper, responses of torque and flux are computed for all possible switching states of the inverter at every sample time according to the discrete time model of the machine and then the switching state that optimizes ripples of torque and flux will be applied in the next discrete-time interval. Simulation results, which confirm the good performance of the proposed predictive torque control are presented.