Low-voltage ride-through handling in wind farm with doubly fed induction generators based on variable-step model predictive control

A severe voltage sag on the wind-farm grid side with doubly fed induction generators (DFIGs) can induce a peak inrush current in the rotors and damage converters, resulting in wind-turbine disconnection from the grid. To prevent this from happening, a variable-step model predictive control (VS-MPC) strategy is proposed for improving the wind turbines' ability to operate without disconnecting themselves from the grid when a fault occurs. First, the predictive-control state-space model of a doubly fed wind farm is established according to its working principle. Second, model predictive control (MPC) is applied on the rotor side of the DFIG to realize the rapid tracking of the rotor current to the reference value during low-voltage ride-through (LVRT) of the DFIG. Finally, a variable-step size algorithm is introduced into the MPC controller to change the step size continuously during the LVRT period. This increases the control accuracy, realizing rapid attenuation of each transient component, whereby the wind-farm LVRT capability is enhanced. The proposed control strategy was simulated and verified using MATLAB/Simulink. The simulation results indicated that VS-MPC can effectively handle LVRT, allowing the recovery of a wind farm that uses DFIGs and improving the performance of wind-farm.