Presentations

Abstract

Assessing wake losses in offshore wind farms increasingly requires consideration of mesoscale flow features due to their dependence on the Atmospheric Boundary Layer (ABL) structure, including atmospheric stability and internal boundary layers; as well as the increasing scale of wind farm clusters being developed. While legacy wake models (e.g., Jensen, TurbOPark) and steady-state CFD are widely used and can be adapted to account for ABL features, these adaptations rely on extensive SCADA data for tuning and validation. Their applicability to regions with unique mesoscale phenomena, such as the US and APAC, remains limited. Mesoscale and large eddy simulation (LES) weather models inherently account for ABL characteristics and can integrate wind farm and turbine parameterizations. These models solve atmospheric flow equations directly, representing key phenomena like blockage and wake interactions. However, their validity and integration into yield assessment workflows are less established compared to legacy models, with limited third-party validation studies available. This study evaluates the validity and applicability of two commercial mesoscale weather models for wind farm wake assessments, in conjunction with legacy engineering wake models. First, model performance is validated against publicly available in-situ wind measurements (e.g., lidar and mast data), satellite-based wind datasets (SAR), and publicly available wind farm power time series. Validation focuses on near- and far-wake wind speeds across different atmospheric stability classes, identifying model biases and proposing tuning where relevant. Second, the models’ applicability in Energy Yield Assessments (EYA) is examined through a reference wind farm cluster, with recommendations provided based on cost-benefit analyses and uncertainty reduction at various EYA stages. Results demonstrate the validity of the evaluated mesoscale models—a WRF model with wind farm parameterization and a full mesoscale-coupled LES code—for wind speed predictions under different stability conditions. Relative wind speed differences (e.g., between lidar locations or pre-/post-wind farm buildout scenarios) are accurately captured, with stability-specific biases identified and tuning discussed. A workflow for integrating mesoscale models into EYA processes is proposed, highlighting their ability to reduce uncertainty and complement legacy models. This study makes a novel contribution by providing detailed validation of two commercial mesoscale models in the same analysis, using high-quality in-situ measurements and publicly available wind farm data. It also explores the integration of mesoscale models into legacy workflows, proposing a framework for offshore wake modelling that emphasizes transparency and reproducibility. Implications for Rotor Nacelle Assembly site suitability and support structure design are discussed, offering insights for advancing wake modelling practices. Delegates will gain an independent validation of mesoscale wake models alongside legacy models, methods for characterizing model bias using atmospheric stability, and best practices for validation and interpretation. Practical guidance on combining mesoscale model outputs with legacy models across different EYA phases—spanning early-stage assessment to final investment decision—is also provided.