Presentations

Abstract

Blockage effects have recently come to the forefront of research in the offshore wind industry, in particular with Ørsted's announcement attributing adjustment in production forecasts for offshore wind farms to both blockage and farm-to-farm wake effects. The majority of the work on blockage has focused on characterising the impact on available wind resource and wind farm yield, although the potential for blockage effects on power curves (BEPC) has been identified [1]. To be suitable for yield calculations and power performance verification, measured power curves are typically required to be representative of a single wind turbine operating in isolation. However, Power Performance Tests (PPTs) are seldom conducted on turbines in isolation, and thus measured power curves are vulnerable to aerodynamic influence from neighbouring turbines (i.e. blockage). Initial work has been reported in the literature to develop a candidate numerical scheme to correct for BEPC [2]. The technical viability of this approach has recently been demonstrated when applied to correct high-fidelity Reynolds-Averaged Navier-Stokes Computational Fluid Dynamics (RANS CFD) model results [3]. This paper reports work conducted by Frazer-Nash and UL Solution for the Carbon Trust Offshore Wind Accelerator which complements and augments previous studies. Here we incorporate evidence from rapid engineering models and importantly SCADA and nacelle-mounted lidar measurements from the Rødsand 2 wind farm, alongside CFD analyses. The emerging evidence on BEPC from this study is expected to help the industry refine its approaches to accounting for blockage in yield calculations and PPTs. Both our rapid engineering model and CFD model results indicate that BEPC does exist, consistent with previous studies. However, at typical wind speed measurement locations 2 - 4D upstream of a turbine under test, these results show that the bias on power curves due to blockage is relatively small compared to the bias due to the induction field of a single turbine. Evidence of trends consistent with BEPC could not be detected in the Rødsand 2 measurement data. We expect this is because BEPC is small in relation to other sources of residual scatter after correction for density, turbulence intensity, and wind shear. Hence, we conclude that, although a correction in measured datasets for BEPC could be implemented, the benefit of doing so would be small and the effort significant. A consistent correction for the induction field of a single turbine would appear to provide more significant reductions in bias on measured power curves. References: [1] James Bleeg et al., Wind Farm Blockage and the Consequences of Neglecting Its Impact on Energy Production, energies, 2018. [2] James Bleeg, Correcting for the impact of blockage on measured power curves, 2019. [3] Alessandro Sebastiani et al., A method to correct for the effect of blockage and wakes on power performance measurements, Wind Energy Science, 2023.