A data-driven, cooperative wind farm control to maximize the total power production

Abstract

This study investigates the feasibility of using a data-driven optimization approach to determine the coordinated control actions of wind turbines that maximize the total wind farm power production. Conventionally, for a given wind condition, an individual wind turbine maximizes its own power production without taking into consideration the conditions of other wind turbines. Under this greedy control strategy, the wake formed by the upstream wind turbine, resulting in reduced wind speed and increased turbulence intensity inside the wake, would affect and lower the power productions of the downstream wind turbines. To increase the overall wind farm power production, cooperative wind turbine control approaches have been proposed to coordinate the control actions that mitigate the wake interference among the wind turbines and would thus increase the total wind farm power production. This study explores the use of a data-driven approach to identify the optimum coordinated control actions of the wind turbines using limited amount of data. Specifically, we study the feasibility of the Bayesian Ascent (BA) algorithm, a probabilistic optimization algorithm based on non-parametric Gaussian Process regression technique, for the wind farm power maximization problem. The BA algorithm is employed to maximize an analytical wind farm power function that is constructed based on wind farm configurations and wind conditions. The results show that the BA algorithm can achieve a monotonic increase in the total wind farm power production using a small number of function evaluations and has the potentials to be used for real-time wind farm control.