Consideration of atmospheric stability in wind energy modelling

Abstract

Improvements in wind flow modelling accuracy can impact positively on wind farm business cases and therefore contribute towards meeting national and global energy decarbonisation targets. ‘Atmospheric stability’ is a meteorological phenomenon which is often disregarded in conventional wind flow modelling practice but which can have significant impact on wind farm energy predictions. The concept relates to the reaction of the near-surface atmosphere to diurnal and seasonal variation in the heating and cooling influences of the Earth’s surface. This thesis details a demonstration of incorporating atmospheric stability effects into Computational Fluid Dynamics (CFD) wind flow modelling methodology to improve wind farm energy yield assessment and layout design. Measurements of virtual potential temperature differential between 10m and 60 - 100m above ground level (a proxy for atmospheric stability) were made at ten onshore and two offshore sites. A comparison was then made between these measured data and commercially available ‘Vortex’ Weather Research & Forecasting (WRF) mesoscale data and the latter was shown to be a viable, cheaper and more readily accessible input data source. A new technique for parametrising stability data was developed as part of a streamlined CFD-based wind flow modelling approach. CFD simulations of key wind flow parameters (wind shear, turbulence intensity and wind speed ratio) were made using both a ‘neutral’ assumption (a widely used industry-standard approach) and a more sophisticated ‘diabatic’ assumption, which incorporated site-specific inputs. These predictions were compared against on-site measurements, demonstrating an overall improvement in modelling accuracy when the diabatic assumption was implemented. Energy yields for onshore wind farms with fixed layouts calculated using neutral and diabatic flow modelling were shown to differ from one another by 1.4% on average (0.1 to 4.8%). Further, onshore wind farm layouts optimised based on the diabatic flow modelling method showed an average increase in predicted energy yield of 0.5% (0.1 to 1.1%) compared to those optimised based on neutral flow modelling. For a mature energy technology such as onshore wind, energy yield uplifts in this range can significantly impact project viability.