Font Size: a A A

Research On Uncertainty Quantification Methods For Aeroelastic Stability Of Wind Turbine Blade

Posted on:2023-08-11Degree:DoctorType:Dissertation
Country:ChinaCandidate:J X YuFull Text:PDF
GTID:1522306938493524Subject:Power Engineering and Engineering Thermophysics
Abstract/Summary:PDF Full Text Request
Vigorously developing wind power is an important way to achieve dual-carbon strategic objectives.With the large-scale wind turbines,wind turbine blades are getting longer,more flexible,aeroelastic problem is more and more prominent.Studying the aeroelastic stability boundary of long flexible blades is the main task of the design and operation of large wind turbines.Aeroelastic stability of wind turbine blades is affected by aerodynamic loads and structural characteristics.Blade geometry determines its aerodynamic performance;blade material properties and ply parameters determine the mechanical properties of the structure.Due to imperfect design methods,the geometric shape and ply parameters of blades may be uncertain to some extent.As a result,the aerodynamic and aeroelastic characteristics of blades also have certain uncertainties,and there are potential risks affecting the safe and stable operation of wind turbines.In view of the above problems,it is of great significance to study the aeroelastic stability analysis method of wind turbine blades considering the influence of uncertain factors.Taking the wind turbine blade airfoils and the blade of the 10 MW reference wind turbine of the Danish Technical University(DTU)as the research objects,the uncertainty in the aeroelastic stability of the wind turbine blade is quantified based on the uncertainty quantification method.The main research work and achievements are as follows:(1)Uncertainty quantification method and its coupling method with numerical simulation are studied.The accuracy and efficiency of sampling methods(Monte Carlo,Stratified sampling,Latin hypercube)and polynomial chaos approaches(non-intrusive polynomial chaos and sparse grid-based polynomial chaos)are verified and compared by typical math test functions in 3,6 and 10 dimensions,respectively.On this basis,the above method is combined with the Computational Fluid Dynamics(CFD)method.Based on the open source solver OpenFOAM,a quantitative numerical calculation method for the uncertainty of aerodynamic parameters is established.The typical CFD example-random cavity flow is used to verify and confirm the method.The random variation of parameters in the flow field is analyzed when the cavity boundary velocity and fluid viscosity are random variables subject to independent Gaussian distribution(the dimension of random variables is 2~5).The results show that compared with the deterministic numerical calculation method,CFD simulation method considering uncertainty quantification analyzes output from multiple perspectives and can give the statistical information of the target quantities.By comparing the efficiency and accuracy of different methods,polynomial chaos method and sparse grid-based polynomial chaos method have higher accuracy.In addition,in high-dimensional problems,sparse grid-based polynomial chaos method has high efficiency.(2)Using the validated uncertainty quantification and numerical simulation coupling method,the aerodynamic characteristics of airfoils with high-dimensional random geometric parameters are studied.Using class function/shape function transformation(CST)geometric parameterization method,aiming at the random change of geometric shape design parameters,the geometric random parameterization model of airfoil is established with the geometric shape parameters of airfoil as uncertain variables(the dimension of random variables is 12 dimensions).The influence of geometric shape uncertainty on the aerodynamic performance of S809 airfoil and FFA-W3-241 airfoil is analyzed at Reynolds numbers of 1 million,6 million and 12 million,and the sensitivity of airfoil aerodynamic performance to airfoil geometric parameters is revealed.The results show that the geometric uncertainty has a more significant effect on the aerodynamic performance after flow separation.When the flow is an attached flow,the influence on the pressure coefficient of the upper surface is greater than that of the lower surface.After the flow separation occurs,the pressure coefficient near the leading edge of the upper surface is affected more.The aerodynamic forces of S809 airfoil are sensitive to the maximum camber parameter and the trailing edge parameter of the suction surface under the attached flow,and are sensitive to the leading edge and the maximum camber parameters after stall.With the increase of Reynolds number,the influence of the parameters controlling the maximum camber of the pressure surface on the aerodynamic force is weakened.FFA-W3-241 has no prominent sensitive parameters under the attached flow,and the sensitive parameters after stall are the same as those of S809.Aiming at the airfoil oscillation problem,a two-dimensional simplified model of blade aeroelastic stability,based on the uncertainty numerical calculation method,the independent variable parameters of the sine and cosine motion law of the airfoil:the average angle of attack,the pitch amplitude,the plunge amplitude,the pitch reduced frequency and the plunge reduced frequency are the uncertainty quantities obeying the Gaussian distribution(the dimension of the random variable is 5 dimensions),and the FFA-W3-241 airfoil stall flutter boundary at 1.6 million Reynolds number is found.It is found that the sensitivity of the flutter boundary of pitch motion to uncertain variables is:average angle of attack>pitch reduced frequency>pitch amplitude.The sensitivity of the flutter boundary of the plunge motion to the uncertain variables is as follows:plunge reduced frequency>plunge amplitude>average angle of attack.In the coupled motion,the most important factor affecting the flutter boundary is the pitch reduced frequency,followed by the plunge reduced frequency.The sensitivities of the plunge amplitude and average angle of attack to flutter are not high.(4)Aiming at the aeroelastic stability problem of wind turbine blades,the blade of DTU 10MW rotor is taken as the research object.It is assumed that the ply angles of 11 parts of blade sections are uncertain variables obeying independent Gaussian distribution(the dimension of random variable is 11).The influence of the ply angle of each region on the structural performance and stability of the blade is analyzed in depth,and the region with the greatest influence on the stability of the blade and the recommended value range of the ply angle in this region are given.It is found that the effect of ply angle uncertainty on the edgewise-torsion coupling is greater than that on the flapwise-torsion coupling.The influence on the stiffness coefficient from the root to 20%blade span is greater than that of the other parts.The edgewise-torsion coupling coefficient,torsional stiffness and edgewise-torsion coupling stiffness are more sensitive to the uncertainty of ply angle at the Caps and Web A.The stability analysis is carried out with the ply angle of the Caps and Web A as the uncertain variables,and it is found that the critical blade tip speed obeys the skewed distribution when it tends to be unstable,and the mean value is 130.9m/s,which is lower than the critical blade tip speed(133.2m/s)under deterministic conditions.The relationship between critical blade tip velocity and ply angle of Caps is parabolic.The critical blade tip speed will be increased and the stability of the unit will be improved when the ply angle of the Caps is near 0°and the ply angle of the Web A is-5°~19°.
Keywords/Search Tags:wind turbine blades, aeroelastic stability, uncertainty quantification method, flutter boundary, wind turbine airfoil
PDF Full Text Request
Related items