Posters

Abstract

Offshore wind continues to grow and play an important part in the energy transition. Whilst Europe is currently leading the way with the majority share of global installed offshore wind capacity, development is gaining pace in other geographies and new wind climates, particularly in Asia-Pacific. Offshore wind turbine technology is also advancing, with models extending further into the atmospheric boundary layer and tip heights surpassing 250m. These developments call for some consideration to expanding the suite of modelling approaches to better account for varying atmospheric conditions. Accurate energy yield assessments are the cornerstone for feasible and bankable wind farm projects. Traditionally, Financial-Grade energy yield assessments have relied on engineering wake models validated with operational wind farm production datasets. Previous offshore energy validation exercises performed by the authors have demonstrated that an ensemble of engineering wake models can reliably predict real operational performance across the wind turbines, farm sizes and geographies covered by the validation dataset. However, engineering wake models, despite their well-understood limitations, can now be complemented by new models designed to better capture the effect of atmospheric physics on wake effects. Weather-based wake models of varying resolutions have emerged, offering mature and validated studies. These models are becoming affordable within commercial energy yield assessments due to advances in modern computing. Can both types of wake models co-exist in today's wind industry? Should they address different questions, or work together to produce more viable results for bankable energy yield assessments? To explore these questions, the authors conducted a comparative study, examining the performance of these models against a subset of the aforementioned validation dataset. The study focuses on a well-understood wind farm cluster comprising of 9 offshore wind farms in the North Sea, using publicly available operational datasets. Wake effects of the chosen wind farm cluster have been estimated using commercially available weather-based simulation providers. The resulting wake impacts were then compared to traditional engineering wake models, which have been validated and ‘tuned’ against the author’s full validation dataset (comprising operational offshore wind farm data from over 15 GW of capacity across 7 countries). Results reveal general alignment in modelled wake efficiencies on a relative basis, albeit with some differences in estimated magnitude. At the 9 sites investigated, the weather-based models paired with wind regimes derived from measurements achieved similar, and sometimes superior, alignment to the ’real world’ production data output when compared to the estimates from traditional engineering wake models with the same input wind data. This evidence supports further integration and adoption of weather-based models into energy yield assessment methodologies as they reach commercial maturity, along with a re-evaluation of wake modelling uncertainties. This study demonstrates the value in leveraging comprehensive validation datasets to investigate emerging weather-based wake models, improving confidence and understanding of their performance in a commercial context. Ultimately, reduced uncertainty in pre-construction energy yield estimates (of which wakes are a significant contributor for offshore wind farm projects) leads to improved cost of finance, resulting in more viable offshore wind farms that hopefully reach construction.