风力机翼型失速流动数据同化
Research on data assimilation of wind turbine airfoils in stall
孟令庭 1杨俊伟 2杨华1
作者信息
- 1. 扬州大学电气与能源动力工程学院,扬州 225127
- 2. 扬州大学广陵学院,扬州 225000
- 折叠
摘要
为验证数据同化优化湍流模型参数方法在不同厚度翼型上的效果以及同一优化参数在不同翼型上的适用性,对应用在风力机上的 3 种不同厚度的翼型(NACA63415、S809、DU97W300)进行失速流动下的数据同化研究.基于现有的风洞试验数据,使用集合卡尔曼滤波方法优化S-A(Spalart-Allramas)湍流模型的参数,确定各模型参数对数值模拟的影响程度,并对比同一翼型优化前后以及在不同翼型上计算的表面压力分布结果.结果表明,模型参数优化后,计算误差减小,模拟的气流分离点和压力分布曲线都更加贴近实验值.此外,翼型S809 的优化参数同样可以减小另两个翼型的计算误差,能够适应其他翼型的气动性能计算,但是误差略大于其自身优化计算结果.同化后,3 种翼型的模型参数中变化较大的是Cb1、Cυ1和σ,其中Cb1变化幅度最大,且在各翼型上的变化趋势都表现为减小.因此可以推测Cb1对于模型参数的适用性至关重要,至少对于本文研究的 3 种翼型而言是如此.
Abstract
In the present paper,we utilize the ensemble Kalman filter(EnKF)method to optimize parameters of the turbulence model Spalart-Allramas(S-A)to simulate flow fields around three wind turbine airfoils(NACA63415,S809,and DU97W300)at the same stall degree.The optimized parameters of the medium-thickness airfoil S809 are further applied to the other two airfoils to investigate their performance on airfoils with various thicknesses.For each airfoil,comparisons to experimental data regarding the pressure coefficient distribution and separation location show that using the optimized parameters can reduce numerical errors significantly.Even though applying the optimized parameters of S809 to the other two can also considerably reduce the numerical errors,the errors are slightly larger than those obtained by the optimized parameters of their own.In addition,the model parameters with significant changes after assimilation are Cb1,Cυ1,and σ,of which Cb1 undergoes the most prominent change and decreases posterior of assimilation for all three airfoils.It is speculated that Cb1 is crucial for the model parameters'universality,at least for the present three airfoils,of the model parameters.
关键词
数据同化/集合卡尔曼滤波/失速/S-A湍流模型/压力分布Key words
data assimilation/ensemble Kalman filter/stall/S-A turbulence model/pressure distribution引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据