The Offshore Wind Accelerator (OWA) identifies wind turbine wake interaction modelling as one of the top five technologies to improve offshore wind farm efficiency. Recognising this, and the fact that general-purpose Computational Fluid Dynamics (CFD) are a limiting factor on standard hardware, the Simulated Wake Effects Platform for Turbines (SWEPT2) project investigates the use of GPU-based CFD, a faster and more scalable alternative.

The SWEPT2 project, led by DNV-GL has a consortium of SSE, ORE Catapult, STFC Hartree Centre, The Centre for Modelling and Simulation (CFMS), Zenotech and the University of Surrey, University of Strathclyde, University of Bristol and Imperial College London.

Wind Farm Modelling with Advanced CFD

The SWEPT2 project provides the green energy industry with an advanced toolset to design larger turbines and arrays, improving the prediction of wind farm failure, reducing costs and allowing energy primes to cut carbon emissions and design better farm layouts to deliver cheaper, greener electricity.

SWEPT2 investigates the use of machine learning methods to infer the wake interaction patterns and consequent power production from sites. CFMS led the use of Artificial Intelligence (AI) technology to deliver insights from the large volumes of data produced. These datasets included a range of wind directions, terrain maps and full CFD fields.

The SWEPT2 project developed new tools to improve the utility and accuracy of CFD-based wake interaction modelling. The project developed a new CFD capability for wind turbine modelling, with validation and benchmarking completed utilising High Performance Computing (HPC).

Talk to us today