Intelligent MPC for DFIG wind turbines

Abstract

This paper presents and evaluates advanced control strategies to enhance power tracking and robustness in doubly fed induction generator systems operating under realistic and perturbed wind conditions. In addition to the conventional field-oriented control, we develop a model predictive control approach that determines the optimal rotor voltage vectors by minimizing a quadratic cost function, as well as a fuzzy-weighted model predictive control in which the cost weight is adjusted online based on the tracking error and its derivative. The dynamic models used accurately represent the key behaviors of the doubly fed induction generator and its rotor-side and grid-side converters. MATLAB/Simulink simulations are carried out using two wind scenarios: a smooth sinusoidal profile and a filtered stochastic profile, while a robustness test introduces variations in rotor parameters during operation. The results demonstrate that the fuzzy-weighted model predictive control achieves faster convergence, lower steady-state error, and improved robustness, all while maintaining reasonable converter effort and acceptable power quality.