Simulation Validation and Measurement Masts

Validating our models is an important part of our work and our integrity as engineers. Just as with our validations for wind around buildings and mixing processes, we do validate our simulations for wind turbines. Enter Bolund.

Bolund is an small island near Roskilde in Denmark that was subject to an experiment in atmospheric-sized wind measurement. A number of measurement masts were erected throughout the island sporting various measurement equipment. The island itself was LIDAR-scanned to provide a detailed model and published together with measurement results to provide a complete test case for validating atmospheric wind simulations.

The geometry of the island is especially interesting since the sharply dropping cliff in windward direction provides similar conditions than the ones wind turbines on ridges are subject to (even if those have some additional specialties, that we'll cover in the future publication). That makes it especially well suited to validate it against our atmospheric LES wind models and inlet gusts. Have a look at the video:


Download video: Full HD - 1080p | HD Ready - 720p

A direct comparison of the time-averaged simulation data with the measurement data of the wind masts shows an excellent fit! However other than the measurement masts, the simulation can cover a large area, so that the best suited spots can be selected. The simulation can also explore what-if scenarios, for example including wakes of planned wind turbines and large buildings that might influence the performance of a wind park.

The following plots compare simulated and experimental turbulence statistics at measurement mast positions. The turbulence statistic basically shows that the simulation does not only do an excellent job in predicting the wind speed, it also does provide a very good prediction of the changes in wind speed due to gusts. Since maximum gust wind speeds can easily be twice as high as the average wind speed, this is an important factor to know when assessing wind loads on structures (from turbines over sound protection walls to high-rise buildings). The prediction of these velocity changes is especially complex, since gusts vary with height: they are relatively small at ground level, increasing in variation towards the average height of the built environment and then slowly fade into the much more uniform flow at greater heights.

While measurement masts do play an important role to determine the overall wind conditions at a planned site, additional CFD simulation gives better insight into finding the best spots for individual wind turbines including their wake interaction, thus providing crucial information to maximize the performance of a wind park.