Wind Turbine and Wind Farm Wakes
Title: Wind Turbine and Wind Farm Wakes
SNIC Project: SNIC 2013/26-16
Project Type: SNAC Large
Principal Investigator: Stefan Ivanell <stefan.ivanell@geo.uu.se>
Affiliation: Uppsala University Campus Gotalnd
Duration: 2014-01-01 – 2014-07-01
Classification: 20306 20304 20301
Homepage: http://picard.hgo.se/~ivanell/?q=node/6
Keywords:

Abstract

Our research group at Uppsala University Campus Gotland has been performing research on wind turbine wakes for about 10 years. The interests of our group span all the way from a fundamental understanding of the development of the wakes behind wind turbines and a study of the influence of turbulence on the breakdown of the wake, to the interaction of wind turbines on each other and the optimization of full wind farms in terms of production and reduction of loads. We are also investigating the interaction of wind farms on each other which is soon to become an important issue with the rapid development of offshore wind farms in Europe. We are also part of two annexes from the International Energy Agency where work is made on acquiring a better understanding of the near wake, as well as of wake interactions in full farm configurations. In the Computational Fluid Dynamics (CFD) simulations we perform with the code EllipSys3D, we are using different models to represent the wind turbine rotors, namely the actuator disc and actuator line models. Using these models, the presence of the rotor is modeled with body forces. This implies that the development of the boundary layer on the blades does not need to be computed, which allows to save precious computational time that can be used instead for modeling the wake behind the turbines. Despite these simplifications, large computational resources are needed to perform these CFD simulations. To our knowledge, our group has been the first one to submit a paper where a full scale simulation of the offshore wind turbine park Lillgrund has been performed with the Large-Eddy Simulation (LES) technique coupled with the actuator disc representation of the wind turbine rotors. We would like to push this study further in the near future, modeling the turbine rotors as actuator lines in this park, which will necessitate substantially more computational resources. As the project description related to this application shows, there are many other projects we would like to pursue using HPC resources. The increase in size of our group also implies growing needs for computational resources. The activity report included in this application shows that the experience we have using such resources has resulted in many publications. We acknowledge SNIC for the work that we could perform up to now, and we hope that the present demand will result in a positive outcome, allowing us to bring our research forward.