Presentations

Abstract

Traditional Energy Production Estimates (EPEs) for windfarms rely heavily on engineering wake models calibrated with heuristics from previous-generation operational farms. With the expansion of offshore wind energy, especially in mega-clusters, the gap between calibration and modelled scenarios has grown too wide for historical approaches. Notably, extreme disparities exist between high-fidelity and most engineering models (even when these models agree on smaller scale problems). RWE have transitioned from heuristics to models based on physically justified assumptions, including the in-house engineering model (VV3) and higher order Computational Fluid Dynamics (CFD). This allows realistic modelling of future mega-cluster scenarios with thousands of turbines. VV3 has been developed with a rigorous adherence to physically grounded assumptions; using only simplifications that have been well justified and validated outside of the wind industry and at multiple scales (e.g. Prandtl 1904 mixing length theory). The individual axisymmetric viscous wake is modelled using the Ainslie 1988 Eddy Viscosity (EV) model. Three sources of EV are captured: wake and ambient (as in Ainslie), and third, EV induced by neighbouring wakes. The ground effect is captured using an image wake. The wake initialisation and superposition methods are derived in tandem to ensure conservation of momentum. Local inviscid blockage is modelled as a Rankin Half Body (RHB) similar to the work of Gribben & Hawkes 2019 with the RHB sized to ensure conservation of mass at the rotor. The Global Blockage Effect (GBE) is captured by an infinite integral approximation of reflections of the local blockage above the ABL height. Ground effects for local and global blockage are captured using image RHBs. In this paper, the underlying physical assumptions of VV3 will be derived; validation against both CFD and measurements presented; and implications on the EPE of future buildout demonstrated and discussed. Wake impacts are shown in both measured and modelled data over c. 30km, and are shown to have an impact over 100s of km. Within mega-clusters, impact on AEP of order of 10 percentage points (compared to historic models) have been observed. References Prandtl, L. (1925). "7. Bericht über Untersuchungen zur ausgebildeten Turbulenz". Z. Angew. Math. Mech. 5 (1): 136-139. Ainslie, J. F. (1988) "Calculating the Flowfield in the Wake of Wind Turbines" J. Wind. Eng. Ind. Aerodyn., 27 (1-3): 213-224 Gribben, B. J. & Hawkes, G. S. (2019) "A Potential Flow Model for Wind Turbine Induction and Wind Farm Blockage". Frazer-Nash Consultancy, Technical Paper.