Wind farm wake recovery: LES and engineering models compared to wind tunnel data

Abstract We compare LES and an engineering model to wind tunnel measurements of the wake of two model wind farms with staggered and aligned layouts. LES and measurements agree very well with a global deviation of less than 8% on TI and WS. Most of the differences were localized in the very near wake of the actuator disks, where the turbulence intensity is underestimated of 20-50%, and the region in between the wakes of the tubines columns of the aligned layout, where velocity is underestimated of 7%. The engineering model ( TurbOPark ) failed to accurately match the wind farm wake. The default TurbOPark underestimates wake recovery in the near-wake and overestimates it in the far-wake, and the turbines’ added turbulence intensity decays too slowly. Tuning the correlation coefficient between wake expansion and turbulence intensity could not gain a good fit of the experimental data. Finally, we use LES to further investigate the recovery of the two wind farms’ wakes. We propose a criterion to identify the wake recovery region for which we calculated wake recovery budgets. We found that single-turbine wake-recovery traits survive much longer downstream of the aligned wind farm, as more space is required for the wakes of turbine columns to merge and recover as a unique entity. This was identified by the horizontal divergence of the Reynolds stresses <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline" overflow="scroll"> <mml:mrow> <mml:mo>(</mml:mo> <mml:mo>∂</mml:mo> <mml:mover accent="true"> <mml:mrow> <mml:msup> <mml:mi>u</mml:mi> <mml:mo>′</mml:mo> </mml:msup> <mml:msup> <mml:mi>υ</mml:mi> <mml:mo>′</mml:mo> </mml:msup> </mml:mrow> <mml:mo stretchy="true">¯</mml:mo> </mml:mover> <mml:mo>/</mml:mo> <mml:mo>∂</mml:mo> <mml:mi>y</mml:mi> <mml:mo>)</mml:mo> </mml:mrow> </mml:math> contributing to recovery up until ∼ 25 D downstream the last farm row, while only up until ∼ 5 D for the staggered wind farm.