FacebookTwitterLinkedInImprimer

Wind energy accounts for an increasing share of the energy mix thank to the construction of wind farms designed to reduce investment and operating costs. However, within a farm, downstream wind turbines are exposed to the wakes of upstream rotors, leading to a reduction in their production and an increase in their fatigue stresses. Moreover, wind turbines operate in a complex environment, known as the atmospheric boundary layer (ABL), which interacts with these wakes. This zone of the atmosphere is subjected to large-scale effects (pressure gradients, Coriolis effectsa) as well as local effects, such as thermal exchange with the ground and the surrounding topography.

For numerical simulation (as illustrated in Figure 1), a real challenge exists, associated with the multi-scale nature of the problem: it is necessary to analyze and model the impact of kilometric phenomena (atmospheric turbulence) on wakes, the origin of which is on a metric scale (flow around a blade).

To address this problem, a research partnership was set up with the CNRM (French National Meteorology Research Center) hinged around the Meso-NH simulation tool. The CNRM integrated precise ABL modeling into the tool, including the various phenomena of importance for wind energy technology, such as turbulence, thermal stratification and topography[1]. For its part, IFPEN implemented representative wind turbine models in the tool, based on actuator line and disc approaches [2] [3].
As a result of these changes, researchers were able to use Meso-NH to study wake behavior [4][5][6] and develop analytical models that have since been incorporated in FarmShadowTM, the wind farm design tool developed by IFPEN.
  

Figure 1 Coupe horizontale du champ de vent, simulé par waLBerla


The resulting improvement compared to existing simulation tools is that FarmShadowTM makes it possible, in a few seconds of calculation time, to estimate the production of wind farm. Combined with IFPEN’s optimization methods, it also makes it possible to maximize a farm’s overall production by adjusting wind turbine placement. 

Moreover, new analytical models of unsteady wakes interacting with the ABL will be employed in the DeepLines WindTM aero-hydro-servo-elastic computational code, used to design a wind turbine within a farm. 

Regarding high-fidelity simulations, one sticking point still to be overcome is that of calculation time: conducted by LESd, simulations arelimited from a practical point of view to a few configurations including only a few turbines. An alternative approach, based on Lattice Boltzmann methods (LBM), may make it possible to overcome this challenge and is currently the focus of research (figures 1 and 2), in partnership with Erlangen University [7]. By exploiting the capacities of computer graphics cards, early results show a 400-fold reduction in calculation time. Work is in progress to implement in this solver the physical models necessary for the simulation of CLA.

Figure 2



a- Coriolis force: an inertial force that acts perpendicular to the direction of movement of objects in motion in a medium itself rotating uniformly
b- waLBerla: Widely applicable Lattice Boltzmann from Erlangen: massively parallel framework for multi-physical applications
c- FAU-Erlangen: Friedrich-Alexander-Universität Erlangen-Nürnberg
   

References:

  1. Lac, C., Chaboureau, J. P., Masson, V., Pinty, J. P., Tulet, P., Escobar, J., ... & Wautelet, P. (2018). Overview of the Meso-NH model version 5.4 and its applications. Geoscientific Model Development, 11(5), 1929-1969.
    >> DOI: 10.5194/gmd-11-1929-2018
       
  2. Joulin, P. A. (2019). Modélisation à fine échelle des interactions entre parcs éoliens et météorologie locale (Doctoral dissertation).
    >> https://www.theses.fr/2019INPT0135
      
  3. Joulin, P. A., Mayol, M. L., Masson, V., Blondel, F., Rodier, Q., Cathelain, M., & Lac, C. (2020). The actuator line method in the meteorological LES model meso-NH to analyze the horns rev 1 wind farm photo case. Frontiers in Earth Science, 7, 350.
    >> DOI: 10.3389/feart.2019.00350
      
  4.  Blondel, F., Cathelain, M., An alternative form of the super-Gaussian wind turbine wake model, Wind Energ. Sci., 2020 
    >> DOI: 10.5194/wes-5-1225-2020
       
  5. Blondel, F., Cathelain, M., Joulin, P.A., Bozonnet, P., An adaptation of the super-Gaussian wake model for yawed wind turbines, J. Phys.: Conf. Ser. 1618 062031, 2020
    >> DOI: 10.1088/1742-6596/1618/6/062031
       
  6.  Jézéquel, E., Blondel, F., and Masson, V.: Breakdown of the velocity and turbulence in the wake of a wind turbine – Part 2: Analytical modeling, Wind Energ. Sci. Discuss. [preprint], in review, 2022
    >> DOI: 10.5194/wes-2022-47
         
  7. Schottenhamml, H., Anciaux-Sedrakian A., Blondel F., Borras-Nadal, A., Joulin, P.A., Rüde, U., Evaluation of a lattice Boltzmann-based wind-turbine actuator line model against a Navier-Stokes approach, 2022
    >> DOI: 10.1088/1742-6596/2265/2/022027
       

Scientific contacts: pierre-antoine.joulin@ifpen.fr ; frederic.blondel@ifpen.fr

>> ISSUE  49 OF SCIENCE@IFPEN