The role of atmospheric stability/turbulence on wakes at the Egmond aan Zee offshore wind farm

Wake Conference 2015

Barthelmie, R.J.^1 ; Churchfield, M.J.^2 ; Moriarty, P.J.^2 ; Lundquist, J.K.^2,3 ; Oxley, G.S.^4 ; Hahn, S.^4 ; Pryor, S.C.^1

Cornell University, Ithaca

NY

14853, United States^1

National Renewable Energy Laboratory, Golden

CO

80401, United States^2

University of Colorado, Boulder

CO

80309, United States^3

Vestas Wind Systems A/S, Aarhus

8200, Denmark^4

关键词: Atmospheric stability;    Directional dependence;    Monin-Obukhov length;    Observation Period;    Out-of-plane bending;    Standard deviation;    Temperature differences;    Turbulence intensity;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/625/1/012002/pdf
DOI  :  10.1088/1742-6596/625/1/012002

来源: IOP

【 摘 要 】

The aim of the paper is to present results from the NREL SOWFA project that compares simulations from models of different fidelity to meteorological and turbine data from the Egmond aan Zee wind farm. Initial results illustrate that wake behavior and impacts are strongly impacted by turbulence intensity [1]. This includes both power losses from wakes and loading illustrated by the out of plane bending moment. Here we focus on understanding the relationship between turbulence and atmospheric stability and whether power losses due to wakes can effectively be characterized by measures of turbulence alone or whether atmospheric stability as a whole plays a fundamental role in wake behavior. The study defines atmospheric stability using the Monin-Obukhov length estimated based on the temperature difference between 116 and 70 m. The data subset selected using this method for the calculation of the Monin-Obukhov length indicate little diurnal or directional dependence of the stability classes but a dominance of stable classes in the spring/unstable classes in fall and of near-neutral classes at high wind speeds (Figure 2). The analysis is complicated by the need to define turbulence intensity. We can select the ratio of the standard deviation of wind speed to mean wind speed in each observation period using data from the meteorological mast, in which case a substantial amount of data must be excluded due to the presence of the wind farm. An alternative is to use data from the wind turbines which could provide a larger data set for analysis. These approaches are examined and compared to illustrate their robustness. Finally, power losses from wakes are categorized according to stability and/or turbulence in order to understand their relative importance in determining the behavior of wind turbine wakes.