Reduced-order modelling of wind turbine airfoil unsteady aerodynamic loading

Abstract

In this paper, three different reduced-order models (auto-regressive and moving average, Volterra series and a surrogate-based recurrence framework model) are presented for the prediction of the unsteady dynamic loading of wind turbine airfoils. A wind turbine blade section can experience unsteady aerodynamic loads when subjected to an unsteady aerodynamic environment. The generations of three reduced-order models for the evaluation of unsteady aerodynamic loads are investigated, with the different models used as a representation of linear or non-linear loading. The validity of the presented reduced-order models is assessed mainly by comparing the model output with unsteady time-accurate computational fluid dynamics (CFD) simulations. The results reveal an encouraging agreement between the computational fluid dynamics simulations and the model predictions under different conditions. All three reduced-order models would therefore be useful in engineering conditioning for aeroelastic analysis and wind turbine design optimization.