Reactive power and voltage control in grid-connected wind farms: an online optimization based fast model predictive control approach

This paper presents the application of an online optimization based fast model predictive control scheme to grid-connected wind farms for reactive power and voltage control. A linear prediction model of the network was used to predict the behavior of the system for a certain prediction horizon, while a modified quadratic programming problem was used for the optimization process. The proposed controller was tested in a 5-bus test system hosting three subwind farms of total 36 MW active power production capacity connected in series to the external network. The controller performed its control action by changing the reactive power output of the sub-wind farms and voltage set-points of an online load tap changer transformer to respect the safety limit imposed on the bus voltages and desired reactive power exchange.