Non‐linear multivariable permanent magnet synchronous machine control: A robust non‐linear generalized predictive controller approach

Abstract Permanent magnet synchronous motors (PMSM) have become prevalent in industry and play an essential role in managing industrial processes, automation systems, and renewable energy sources due to their superior efficiency, torque, and power density. However, because it operates like a non‐linear system with quick dynamics, variable parameters during operation, and unknown disturbances, PMSM presents challenges for machine control. Non‐linear controls are required to account for the non‐linearities of the permanent magnet synchronous machine. Recently, predictive control techniques for non‐linear multi‐variable systems have gained popularity. In this work, a novel approach to robust non‐linear generalized predictive control (RNGPC) has been developed for PMSM, with the aim of tracking the reference speed while maintaining minimum reactive power, robustness to external disturbances, and parameter uncertainties. A new finite horizon cost function is integrated, with an integral action introduced in the control law. The main advantage of this technique is that it does not require the measurement and observation of external disturbance as well as parametric uncertainties. The control strategy method has been tested in the MATLAB/Simulink environment with various operating conditions. The results showed good robustness against parameter changes and ensured fast convergence.