Explicit model predictive control of permanent-magnet synchronous motors fed by matrix converter

Abstract

The paper presents the initial research results on applications of explicit model predictive control to electric drive systems using matrix converter fed permanent-magnet synchronous motors. The current process model or the motor’s discrete state model, which is characterized by a bilinear characteristic, can be considered as a linear time-invariant model during the prediction horizon which is not so long. Based on that assumption, it is easy to design explicit model predictive controller with the support of a MATLAB-based toolbox named as Multi-Parametric Toolbox. Besides, a component containing integrator is added to the predictive controller’s output in order to remove the static error caused by the inaccurate model. The effectiveness of the designed predictive control law is validated by digital simulations. It is shown that this method can be a high-quality solution which is suitable for common experimental systems.