Posters

Abstract

Clearly, wind turbine energy yield is dependent on numerous factors, many of which are considered in detail within an energy yield assessment. However, turbine performance variation due to local wind conditions is often given little consideration, despite it often being the cause of significant energy yield variation from site to site and, especially in complex terrain, between turbines on the same site. As a result, turbine performance can be a significant source of uncertainty. Failure to adequately consider turbine performance throughout the project development cycle can result in non-optimal layouts and reduced net energy yield. This presentation demonstrates how turbine performance is readily predictable by combining standard site measurements and CFD flow modelling to feed a turbine performance model. This leads to more confident development decision making, from early-stage projects right through to financial close. Turbine performance is influenced by wind speed, shear, turbulence intensity, veer and in-flow angle. Consideration of all these parameters, along with high-quality modelling, at an early stage enables areas with high probability of reduced turbine performance to be clearly identified and avoided when installing measurements and turbines. Designing wind farms with increased awareness of the likely performance impact maximises the quality of measurement datasets and hence provides the lowest energy yield uncertainty, as well as optimising the turbine layout for increased energy yield. In this presentation, a complex terrain wind farm site is used to clearly show how potential turbine locations that experience onerous flow conditions can be identified and avoided at an early stage of project development, even prior to carrying out any site measurements. Avoiding these locations removes the risk of both reduced yield and increased turbine loading. Furthermore, as the wind farm development progresses, the same CFD models as used at the early stage can be updated to incorporate the now available site measurement data, increasing their value in uncertainty reduction. The flow modelling results combined with both the site measurements and a higher-fidelity approach to turbine performance modelling enables more accurate turbine performance prediction, leading to higher investor confidence and lower cost of capital. The author will present a well-validated turbine performance model, which leverages over 100 turbine and measurement pairs to provide a clear data driven approach. This approach is derived from a measurement-based non-dimensionalised relationship between wind conditions and turbine performance. As well as clear benefits for commonly experienced flow conditions, the method has been extended to account for any flow separation predicted by the CFD model. The same complex terrain site as considered earlier in the presentation is used to illustrate how this approach provides further additional value to the wind farm project through reduced EYA uncertainty and hence, increased p90 values. The approach applied for this case study will be described in the presentation and the audience will gain clear insights into the benefits of considering turbine performance together with CFD modelling from the early stages of development.