Font Size: a A A

Efficient Surrogate Models For Aerodynamic Analysis Of Airfoils Under Uncertainty

Posted on:2019-09-15Degree:MasterType:Thesis
Country:ChinaCandidate:J F SunFull Text:PDF
GTID:2492306044959969Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Wind turbine blade is the essential and important part of wind turbine generator.Research on aerodynamic performance of wind turbine blade is of great significance to the wind energy efficiency and power output of the wind turbine.The airfoil of blade,acting as part of wind turbine blade,can be a core problem in the research of aerodynamic characteristics of wind turbine.Therefore,wind turbine working stably and reliably plays an important role in the efficiency and quality of power generation.However,during the process of its service,it is inevitable to suffer the influence of various uncertain factors.If the wind turbine is too sensitive to one or several uncertain factors,then the energy conversion efficiency of wind turbine decreases sharply which will lead to huge economic losses to power generation.Therefore,it is of great scientific value and practical significance to carry on the study of flow filed response and aerodynamic characteristics of wind turbine airfoil based on uncertain factors.Taking the wind turbine airfoil NACA 63-215 for research objects,the uncertainty analysis for wind turbine airfoil is carried out based on MATLAB and Fluent joint simulation.During the process of uncertainty analysis,non-intrusive polynomial chaos expansion algorithm which is of high precision and efficiency is introduced.The main research contents of this paper include the following three aspects:(1)The method of aerodynamic performance analysis for wind turbine airfoil is firstly reviewed.Then combining theory of fluid mechanics based on CFD numerical simulation with theory of polynomial chaos expansion,classical fluid mechanics case(lid-driven cavity flow)is introduced to talk about the non-intrusive PCE under the random variables of lid velocity and fluid viscosity.Further more,Monte-Carlo random simulation is utilized to verify the efficiency and accuracy of the non-intrusive PCE algorithm.(2)CFD numerical simulation process of wind turbine airfoil deterministic analysis under different angle of attack is put forward based on MATLAB and Fluent software joint simulation.The aerodynamic characteristics curves(lift coefficient curve,drag coefficient curve,lift-drag ratio curve)are dynamically obtained.And then the optimal angle of attackα*can be succefully obtained on the basis of lift-drag ratio curve.(3)Three important random parameters,namely the inflow velocity,air density and dynamic viscosity is subject to Gaussian distribution in the course of uncertainty analysis.Post-processing analysis of PCE surrogate model of related physical response for wind turbine airfoil flow field and airfoil aerodynamic characteristics is carried out based on the optimal angle of attack α*.And post-processing analysis of PCE surrogate model contains responses statistical moments(mean value,standard deviation),probability distribution(PDF,CDF),sensitivity analysis.The parametric regression analysis between aerodynamic performance of wind turbine airfoil and structure,system excitation parameters is carried out.Utimately,the explicit function relation expression between dependent variable lift drag ratio,independent variables flow velocity,air density and dynamic viscosity is successfully obtained.
Keywords/Search Tags:Polynomial chaos expansion, Non-intrusive, Uncertainty, Wind turbine airfoil, CFD numerical simulation
PDF Full Text Request
Related items