Quantifying variability of Large Eddy Simulations of very large wind farms

Wake Conference 2015

Andersen, S.J.^1 ; Witha, B.^2 ; Breton, S.-P.^3 ; Sørensen, J.N.^1,3 ; Mikkelsen, R.F.^1 ; Ivanell, S.^3

DTU Wind Energy, Technical University of Denmark, Lyngby

2800, Denmark^1

ForWind-Center for Wind Energy Research, Carl von Ossietzky University Oldenburg, Ammerländer Heerstr. 136, Oldenburg

26131, Germany^2

Uppsala University, Dep. of Earth Sciences, Campus Gotland, Visby, Sweden^3

关键词: Actuator disc;    Atmospheric flows;    Blind test;    Flow model;    Large-scales;    Long simulation time;    Numerical flow models;    Wind farm simulations;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/625/1/012027/pdf
DOI  :  10.1088/1742-6596/625/1/012027

来源: IOP

【 摘 要 】

Large Eddy Simulations are inherently dynamic as the largest scales are resolved and the smallest scales are modeled temporally. This raises challenges for simulations including very large scales such as atmospheric flows, which require very long simulation times. Simple averages fail at capturing these dynamics and potentially yield misleading interpretations concerning the capabilities of different models when tested in blind tests or in benchmarking exercises such as Wakebench, where results from different flow models are compared. This article will present results from very large wind farm simulations using Actuator Disc (AD) and Line (AL) models for two different turbine spacings with turbulent inflow. The results of each numerical flow model include a certain variability, and it will be examined if different models result in comparable probability distributions.