Realistic visualisations of wind turbine flow to be shown at SC24

The video shows two computational fluid dynamics (CFD) simulations of wind turbine wakes. On the left are two turbines placed directly in line, while the simulation on the right shows the angle of the front turbine "steered" by 24 degrees. This steering redirects the wake from the turbine at the front, and thus its impact on the downstream turbine. This method is designed to increase the overall power output; in this case a 12% increase is achieved. The video highlights both the large scale of the wakes that are created by the turbines and the potential impact of wake steering.

The wake itself is represented by Q-criterion, a quantity that measures vortices in flows. Because the wake is invisible to the eye in normal circumstances, it is represented here with a smoke-like render using volume rendering. The wake has been coloured in orange and blue according to the flow velocity. The colours were chosen to ensure the wake looks sufficiently distinct not to be mistaken for smoke, but realistic enough to blend into a photorealistic render.

The two CFD simulations used to compute the wake from the wind turbines were performed by Andrew Mole (Imperial College) using the high-order compact finite-difference flow solver XCompact3D and were run on ARCHER2. A precursor simulation was run to generate the neutral atmospheric boundary layer. The wind turbine simulations were run with a billion mesh points for 20,000 time steps before collecting the flow field.

The CFD simulations output were first postprocessed using ParaView to compute the Q-criterion. The final video was made using Blender and it took slightly more than a week of work to fine tune all the different elements in it, from the lighting, materials, animation of the camera, ocean simulation, compositing etc. The 590 frames composing the 4k video took 14 hours to render on a workstation with six NVIDIA GeForce RTX 4090 GPUs. This system was funded by the Engineering and Physical Sciences Research Council (Prosperity Partnership Additional Funding November 2023).

Further information

If you want to know more about visualisation work at EPCC, please see this article: How to make an impactful scientific visualisation

For a more hands-on experience of using Blender for scientific visualisation, see this webinar: Using Blender for Scientific Visualisation

Sebastien's work will be displayed in Room B301 for the duration of SC24: https://sc24.conference-program.com/presentation/?id=art112&sess=sess553

ARCHER2: https://www.archer2.ac.uk

SC24: https://sc24.supercomputing.org 

XCompact3D: https://www.incompact3d.com