EU-Norsewind: assessment of viability of open source CFD code for the wind industry

In 2008 the EC programme NORSEWInD kicked off with a mission to deliver high quality offshore wind speed data for the wind industry. The aim of the project is to deliver offshore wind speed data by measuring wind speed data from remote sensing anemometers on off shore platforms. To reduce the overall cost of the project it was hoped that open source Computational Fluid Dynamics codes could be used. This work reports on the comparison between the results from the commercial code, FLUENT, and the open source code, OpenFOAM, for use in modelling flows over the mounting platforms. Part of the overall NORSEWInD project is the use of LiDAR remote sensing (RS) systems mounted on offshore platforms to measure wind velocity profiles at a number of locations offshore. The data acquired from the offshore RS measurements will be fed into a large and novel wind speed dataset suitable for use by the wind industry. One significant problem identified early on at the NORSEWInD project planning stage was the problem of platform interference effects on the RS data. It was obvious that the airflow data measured above the mounting platforms would be distorted by the presence of the platform However; the extent to which the flow field above the various mounting platforms would be distorted was unknown. Therefore, part of the fundamental research incorporated into the NORSEWInD project was a computational and experimental investigation into the possible scale and extent of the interference. Offshore platforms have extremely complicated shapes and require a very large number of cells in the computational domain in order to generate a sufficiently high resolution of the mesh to model the flows surrounding them. The high number of cells leads to very large computational models requiring many hours of compute time to solve even on a high speed processor. One way of reducing the compute time is by employing parallel processing on a number of computational nodes. However; increasing the number of computational nodes may involve the purchase of extra licenses if using a standard commercial code. The cost of the extra licences can become the limit on the final number of nodes employed. Whilst there are significant benefits to be found when using a commercial code which has a user friendly interface and has undergone significant verification testing the financial advantages of using an open source CFD code, which may not have undergone such rigorous verification and may be more difficult to use, are attractive. This paper reports on the Computational Fluid Dynamics (CFD) modelling of the wind flows over the platforms on which the RS systems are mounted and more specifically on a comparison between the results of simulations created by the commercial code FLUENT and the open source code OpenFOAM. An assessment of the ease with which the open source code can be used is also included