Evaluation of wake superposition methods for wind-farm flow and power prediction

Abstract In this paper, we conduct a multi-model comparison study to evaluate the prediction accuracy and applicability of different wake superposition methods to provide useful guidance for engineering applications. Based on the considered wind farm LES and SCADA data, we find that local linear and wind product methods demonstrate remarkably good and comparable performance to two momentum-conserving methods regardless of the wake interaction conditions. Therefore, they are recommended for practical engineering applications due to their low computational cost and ease of implementation. In contrast, the global linear and global square methods significantly underestimate wind-farm flow velocity under full-wake conditions, whereas the local square method overestimates power prediction in these conditions. Further evaluations applied to other scenarios (e.g., background inhomogeneous flow) are necessary to validate the universality of above recommendations.