The development of wind farms requires the optimal placement of wind turbines in relation to each other in order to maximize overall production. If the turbines are too close to each other, wake effects prevent maximum power output and can lead to mechanical fatigue issues caused by turbulence. Conversely, it they are too far apart, wind resources are not fully exploited.
Currently, offshore wake-related power loss can be as high as 40%. Consequently, optimizing the placement of wind turbines has become a major challenge, which can be addressed by precisely simulating wind flows through the farm. This is achieved by modeling the wake effects of the turbines in representative weather conditions, particularly from the point of view of the atmospheric boundary layera.
Such a model, created by IFPEN, was implemented in Meso-NH(1), the fine mesh weather model designed by the CNRM (French National Center for Meteorological Research, Météo France) and the Laboratoire d’aérologie (French Aerology Laboratory). In it, wind turbines are represented by the aerodynamic forces they generate (drag, lift), determined by simplified models. Hence, turbine wakes are defined thanks to the interaction of the aerodynamic forces with the wind, itself influenced by meteorological phenomena.
Validation studies were conducted using wind tunnel tests (model representing five wind turbines on a hill). The calculation/experience comparison was conducted by studying the vertical profiles of wind speed and turbulence intensity, downstream of each of the wind turbines(1). It has thus been demonstrated that consideration of meteorological conditions offers a more accurate prediction of the wake generated by a wind farm.
This new tool will be employed to study the influence of atmospheric conditions specific to the offshore environment, in order to optimize the layout and productivity of wind farms and to study their potential environmental impact on local meteorology.
a - The part of the atmosphere directly influenced by the presence of the earth’s surface (continental or oceanic).
(1) P-A. Joulin et al., EWEA PhD Seminar, 2018.