Independent renewable energy consultancy Natural Power has published a paper on time-dependant computational fluid dynamics (CFD) analysis which has been accepted by the peer-reviewed Journal of Wind Engineering and Industrial Aerodynamics. The journal reports on all aspects of wind engineering and the activities of the International Association for Wind Engineering.
The paper was written by three industry experts – Claude Abiven, Senior Technical Manager of Natural Power France, Oisin Brady, Director of Natural Power France and Dr. Jose Palma from the University of Porto. The title of the paper is ‘High-Frequency Field Measurements and Time Dependant Computational Modelling for Wind Turbine Siting’.
The work took advantage of high frequency wind data made available by the North Harris Trust on a highly complex site in the Outer Hebrides, off the west coast of Scotland. Spectral analyses were carried out on this dataset and showed the existence of complex wind flow phenomena that changed rapidly over very short time scales, including turbulence and direction veer. CFD modelling was carried out using the VENTOS® CFD wind flow model in time-dependent mode, a feature not readily available in many CFD software packages available on the market. The CFD modelling showed the complex and transient flows to be the result of complex terrain, and allowed the flows to be accurately mapped across the site, away from the measurement location.
Transient wind flow phenomena can significantly affect wind turbine performance and can also lead to substantial uncertainty in wind resource assessment, therefore it is important the flow field is fully understood over both space and time. The techniques pioneered here, and described in the published paper, can provide project engineers and developers with a validated understanding that allows the impacts to be addressed and mitigated. Spectral analyses of the simulations showed that the model was able to accurately reproduce the preferred frequencies recorded at the mast. These preferred frequencies were then linked to a physical phenomenon using Empirical Orthogonal Functions, a technique virtually un-used within the wind power community. The methodology described in the paper enables the characterisation of complex flow structures such as zones of recirculation based on a single mast measurement.
[mappress]
Source: Natural Power, March 02, 2011