Wind turbine interference in a wind farm layout optimization mixed integer linear programming model

This work develops a wind intensity interference coefficient which captures the interference caused by an upwind turbine on a downwind turbine in the same wind flow. This coefficient includes the use of a Weibull distribution to handle variability in the wind velocity, and also accounts for the geometric relationship between the turbines and the boundaries of the wind sector. This interference coefficient then forms part of a mixed integer linear program (MILP) which is used to optimise the locations of wind turbines within a wind farm site. The MILP approach is an exact method that can guarantee that the turbine locations determined by the model are optimal with respect to an a priori chosen set of possible turbine locations. Tests on an example data set (based on a demonstration case in the WindFarmer software) show, for the particular wind resource used, that interference results in a loss of power of approximately 4.2% when compared to the power production which would be predicted without accounting for interference.