Presentations

Abstract

With the increasing number of closely spaced wind farms, it becomes vital to understand the impact they have on each other's yield. Accurate estimation of not only internal wake losses but also those due to neighbouring farms is crucial for operators, given their direct impact on the expected yield. While uncalibrated analytical wake models can estimate these losses, calibration using SCADA incorporates turbine-, farm-, and site-specific factors, improving model accuracy. Traditional calibration methods, performed solely on intra-farm losses, prove insufficient, especially for closely spaced wind farms. Calibration should therefore account for neighbouring wind farms, and, if possible, the entire concession zone. This research aims to calibrate the tuning parameters of analytical wake models using SCADA data from a wind farm cluster with a capacity exceeding 2GW at rated power. Though previously validated on an individual wind farm, its application to an offshore cluster aims to analyze intra- and inter-farm wake effects in the Belgian-Dutch offshore zone. Results show an overestimation of the predicted wake losses when the tuning parameter is calibrated on a single wind farm with inter-farm losses compared to calibration applied on multiple wind farms within the offshore concession zone. This implies that losses in larger wind farms can potentially be underestimated when considering calibrated tuning parameters from smaller wind farms. Additionally, results for the entire concession zone are validated for wind coming from north-west, showing that the model is accurately able to account for cluster wake effects. Meanwhile, the model is unable to account for large heterogeneous inflow from the south-east, due to the homogeneous inflow assumption. The observed trend of inflow heterogeneity within the offshore concession zone suggests that the discrepancy cannot be attributed solely to blockage effects.