Search

The design and the experimental validation of a continuous-time model predictive control (CTMPC) for a permanent magnet synchronous motor (PMSM) drive with disturbance decoupling is discussed. The CTMPC approach uses Taylor series expansion to derive a closed-form solution to the problem of model predictive control even though the system behaviour is described by a non-linear model. This type of controller requires an exact knowledge of the system model to guarantee an accurate prediction of the system behaviour, while the PMSM is usually subjected to model uncertainties and external disturbances such as the load torque. Moreover, in the proposed approach, the predicted speed tracking error is directly used to determine the required voltage command without the need for a cascaded control scheme. As a result, the load torque is seen as unmatched disturbance which makes exact disturbance decoupling more challenging. To overcome such a problem, a non-linear disturbance observer is designed and combined with the CTMPC method to enhance the prediction accuracy under parameter variation and unknown load torque. The feasibility of the proposed approach is experimentally investigated, and good transient and steadystate performances are obtained. [ABSTRACT FROM AUTHOR]