Sensitivity of Wake Modelling Setups

Abstract

Engineering wake modelling with single turbine wake models and wake superposition models (SupModel) requires the choice of two crucial settings: the type of wake superposition and the wake expansion rate parameter. Based on offshore wind farm measurements, the appropriate tuning of such wake modelling setups with respect to wind farm efficiency and their sensitivity to this tuning is investigated. For this purpose, universal and inflow-individual wake expansion rate parameters are optimised for different SupModels via minimising the deviation between simulated and measured power and to compare the results to simulations using a literature value. The work reveals a logarithmic, SupModel-independent sensitivity of the farm efficiency to the wake expansion rate, sufficient to allow equally accurate modelling performance for all setups if a matching parameter is chosen. The choice of the SupModel only matters in case of an unadapted wake expansion rate parameter. Beyond that, the results indicate a systematic flaw of wake models in reproducing farm power measurements for both weak and full wake situations with the same model tuning. Inflow-individual parameters obtained by this work mitigate this with a systematic pattern over wind speed and wind direction and can significantly decrease the modelling uncertainty over varying inflow conditions.