Simulations aux grandes échelles de la turbulence du sillage d’une éolienne située dans écoulement de couche limite atmosphérique

Abstract

The wake created by the presence of a wind turbine corresponds to a slowed and highly turbulent flow. If a wind turbine is in the wake of an upstream wind turbine, it suffers both a loss of production and an increase in structural constraints, which can lead to premature deterioration of the equipment. Studying the propagation of the wake of a wind turbine located in an atmospheric flow will allow a better prediction of changes in atmospheric flow by the presence of wind turbines. And therefore, an optimization of the layout of the wind farms and an increase in both the production and the life of the machines. This work aims to study numerically the evolution of the wake of a wind turbine located in atmospheric boundary layer flow. The problem is dealt with in two parts, the first dealing with the modelling of the atmospheric boundary layer flow, with particular attention to the modelling of the turbulence of the atmospheric flow. The second part deals with the study of the wake of an actuator disc subjected to a very turbulent atmospheric flow. Numerical modelling of atmospheric flow turbulence requires the use of a stochastic turbulence generation technique suitable for the application of atmospheric flow, with careful attention to modelling large atmospheric turbulence scales. The second part deals with the study of the wake of an actuator disc subjected to the atmospheric flow modelled in the first part. A comparison between the wake produced and a wake resulting from a simulation without prescribed turbulence allows conclusions to be drawn. One of the main results is that wind wake evolution for large eddy simulation with prescribed turbulence is more realistic than the resulting wake without prescribed turbulence which requires a very important mesh resolution to solve all flow scales.

Date	20 Dec 2021
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Marlène Sanjosé (Supervisor) & Louis Dufresne (Co-supervisor)