Aerodynamic Damping Identification And Flutter Suppression Of Large-scale Operational Wind Turbines

Wind energy is one of the most rapidly growing renewable sources of energy due to the fact that it has little negative impact on environment incomparison with other alternative sources of energy.To meet the growing demand of world wind energy industry and further reduce the wind generation costs,wind turbines are being scaled up both in size and output power rating.This trend means much higher loads and more flexible blades than current multi-megawatt wind turbines,most probably resulting in aeroelastic instabilities not commonly seen in the current multi-megawatt wind turbines.If not mitigated,it can cause undesirable performance or even lead to early failure of the whole wind turbine system.Hence,refering to the megewatt class wind turbine aeroelastic issues,the researches on estimation of aerodynamic damping for wind turbines and the flutter vibration control of wind turbine blades are performed in this paper.The specific investigations may be concluded as follow:(1)A brief introduction of both world and China wind energy industrial prospect of is provided.Following an overview related to wind turbine aeroelastics.Further,the aerodynamic damping identification techniques for wind turbines and flutter control approaches for wind turbine blades are introduced indetail.(2)The classical Blade Element Momentum theory(BEM)is briefly introduced.Then the Prandtl’s tip loss function and the Glauert correction model are described,respectively.Since those classical aerodynamic loads correction models are not sufficient to account for the 3D effects caused by both the blade tip vortices as well as the centrifugal force and Coriolis force,a new tip lose function proposed by Shen a nd a new stall delay model developed by Bak are introduced.The 13-DOFs aeroelastic wind turbine model and the refined aeroelastic wind turbine model are developed,respectively.The 13-DOFs aeroelastic wind turbine model,which is a modal model,is used for identifying aerodynamic damping of structures.The deformation of blades in the refined aeroelastic wind turbine model is described through the physical model with 3d rotational model and Shen’s correction model involved.The corresponding dynamic characteristics including the eigen frequencies and dynamic response of those two wind turbine models are investigated.Furthermore,the accuracy and validation of those two models are verified based on relevant reports.A wind tunnel experiment is carried out to obtaine the aerodynamic coefficients of a typical blade crosssection in the DTU 10 MW wind turbine blade.(3)The basic concepts of both the statistical linearization method and the continuous wavelet transform based equivalent linearization method are introduced,as well as their limitations.A new discrete wavelet transform based equivalent linearization method is developed.On the based a single degree of freedom Duffing oscillator,the validation and application of both the statistical linearization method and the continuous wavelet transform based equivalent linearization method is verified through numerical simulations under different external excitations.Such single degree of freedom Duffing oscillator with different nonlinearities is tested under stationary excitation,nonstationary excitation and deterministic excitation,respectively.The comparisons between those two linearization approaches are made.Results show the performance of the continuous wavelet transform based equivalent linearizatio n method is much better than the statistical linearization method.The discrete wavelet transform based equivalent linearization method is as accurate as the continuous wavelet transform based equivalent linearization method but with higher efficiency,which provides an effective tool for subsequent researches.(4)A brief introduction of prior aerodynamic damping identification approaches for wind turbines is provided,and the corresponding limitations of those approaches are also described.The continouse wavelet linearization based aerodynamic damping identification method and the discrete wavelet linearization based aerodynamic damping identification method are developed.The idea of both the continouse and discrete wavelet linearization based aerodynamic damping identification methods are introduced in detail.Numerical simulations are performed based on the 13-DOFs aeroelastic wind turbine model with specific values obtained from NREL 5MW wind turbine prototype to verify the validity and accuracy of tha t method.The effect of measurement noise on identification results through those two methods are discussed under three different noise conditions.Results show the discrete wavelet linearization based aerodynamic damping identification method is more accu rate and efficient than the continouse wavelet linearization based aerodynamic damping identification method,since it is capable of reconstructing signals in a efficient way.(5)The mechanism of classical flutter for wind turbine blade is investigated by use of a typical 2D blade crosssection model under quasi-steady aerodynamic loads.Two novel and effective approaches(i.e.the passive eddy current damper for blade and the active double-pitched blade)for improving the flutter-suppressing capability of wind turbines are developed.Numerical simulations are performed based on the refined aeroelastic wind turbine model with specific values obtained from DTU 10 MW wind turbine prototype.The optimal values for two core parameters(i.e.the length of transmission shaft and the damper constant)of this passive eddy current damper in terms of flutter are investigated in detail.Furthermore,the effects of the mean wind velocity,turbulence intensity,structural damping ratio,different damping ratio combination between flapwise mode and torsional mode on flutter characteristics of wind turbine are discussed.It is proved that such passive eddy current damper can significantly enhance the flutter critical rotational speed of the rotor.The construction of this double pitched blade is described in detail,and also the control strategy and the transfer of thrust forces and torque from blade tip part to the inner blade part are introduced.Numerous simulations have been carried out to obtain the optimal length of the blade tip part and the corresponding position with respected to the balde crosssection.Results show this double pitched blade with feedback controller can largly reduce the torsional deformation of wind turbine blades.