Distributed coordinated active and reactive power control of wind farms based on model predictive control

This paper proposes a distributed coordinated active and reactive power control scheme for wind farms based on the model predictive control (MPC) along with the consensus-based distributed information synchronization and estimation, which can optimally dispatch the active power of wind turbines (WTs) and regulate the voltages within the wind farm. For the active power control, the pitch angle and generator torque of WTs are optimally controlled to alleviate fatigue loads of WTs while tracking the power reference of the wind farm required by system operators. For the reactive power/voltage control, the reactive power outputs of WTs are controlled to mitigate the voltage deviations and simultaneously optimize reactive power sharing. Considering the high R/X ratio of the wind farm collector systems, the impact of active power variations on voltages is taken into account to improve the voltage regulation. The proposed scheme is center-free and only requires a sparse communication network. Each WT only exchanges information with its immediate neighbors and the local optimal control problems are solved in parallel, implying good scalability and flexibility for large-scale wind farms. The predictive model of a WT is derived and then the MPC problem is formulated. A wind farm with ten WTs was used to verify the proposed control scheme.