Wind turbines simulated over a complex terrain

This is my first try to solve a CFD case for two wind turbines over a complex terrain. OpenFOAM is used to solve the full 3D Navier-Stokes equations using the k – ε turbulence closure scheme appropriately parameterized for atmospheric flows. The rotor disks are approximated as a momentum sink, to represent the axial force across the disk, associated with a constant thrust coefficient over the rotor area. Enjoy!

Import the terrain.stl

Meshing the geometry

Set up two actuator discs

Velocity X (zero at the ground)

Kinetic k [1.4, 3.6]

  

Dissipation rate ε [0.1, 1]

Below the results of velocity-X and kinetic-k as they have been presented in an iso-surface near the hub height of the wind turbine rotors, inside the atmosperic boundary layer.

  

 Velocity X - near the hub height

(atmospheric boundary layer)

Kinetic k - near the hub height

(atmospheric boundary layer)



Wind energy development in Chios

Wind energy development in Chios and potential impact on microclimate and agriculture

CFD simulation of z-velocity (6-8 m/s)

Wind turbine single and multiple wake effects

  The layout of wind turbines with respect to the prevailing wind direction is a major subject of the current research efforts aiming to minimize the wake interactions or “wake effects” of wind turbines that can be installed in a specific area and finally to maximize the efficiency of the wind farm. The term “wake effects” is used to describe the generation and development of the wake downstream of a wind turbine as well as the wake interactions within a wind farm.

  The kernel part of a method to optimize the layout of the turbines of a wind farm is a wake model. Wake modeling of wind turbines has attracted a lot of attention by the research community during the past decade especially for very large offshore wind farm installations where wake effects are responsible for quite a significant percentage of power output losses. Although a number of analytical and semi-empirical simple models have been developed for wake modeling, they are not capable of providing satisfactory predictions for different turbulence conditions. The substantial progress in computer resources has permitted cost efficient calculations using more sophisticated methods such as the solution of Navier-Stokes equations.