Control of a BDFIG Based on Current and Sliding Mode Predictive Approaches

Abstract

The Brushless Doubly Fed Induction Generator (BDFIG) is a great attraction nowadays thanks to its high reliability and high low voltage ride-through capability. In variable speed generation systems, the most used control techniques are the vector control, direct torque/power control. Vector control is highly sensitive to parameters variations, while the two others offer high amount of ripples reducing power quality. This paper therefore proposes a Sliding Mode Power Predictive Control (SMPPC) and an Indirect Model Power Predictive Control (I-MPPC) for a simple and efficient power control of the BDFIG. The SMPPC takes current components in the αβ frame as sliding surfaces derived from the I-MPPC based on current predictions to track the desired active and reactive powers in the BDFIG. A cost function defined as currents’ quadratic errors is used to choose the appropriate voltage vector for control winding supply, and the Particle Swarm Optimization algorithm is used to determine the optimal gain for the SMPPC. Simulations on MATLAB/Simulink are carried out to show the effectiveness of the control schemes as well as their robustness to some parameters variations. More so, power compensation is used to improve the capability of the control strategy to provide sinusoidal and balanced currents under unbalanced grid conditions. The proposed methods compared to direct power control show good performances in terms of total harmonic distortion reduction, low ripples and robustness, strengthening the idea of using the BDFIG as an alternative to new generation of wind energy conversion systems.