航空学报 > 2015, Vol. 36 Issue (12): 3774-3784   doi: 10.7527/S1000-6893.2015.0118

超声速翼型气动优化设计

曹长强1, 蔡晋生1, 段焰辉2   

  1. 1. 西北工业大学航空学院, 西安 710072;
    2. 中国空气动力研究与发展中心计算空气动力学研究所, 绵阳 621000
  • 收稿日期:2015-02-06 修回日期:2015-05-04 出版日期:2015-12-15 发布日期:2015-05-05
  • 通讯作者: 蔡晋生,Tel.:029-88495381,E-mail:caijsh@nwpu.edu.cn E-mail:caijsh@nwpu.edu.cn
  • 作者简介:曹长强,男,博士研究生。主要研究方向:气动优化设计,降阶模型,E-mail:caoch0501@163.com;蔡晋生,男,博士,教授,博士生导师。主要研究方向:飞行器气动布局设计,计算流体力学,复杂流动控制。Tel:029-88495381,E-mail:caijsh@nwpu.edu.cn;段焰辉,男,博士。主要研究方向:计算流体力学,优化设计,E-mail:duanyanhui@foxmail.com

Aerodynamic design optimization of supersonic airfoils

CAO Changqiang1, CAI Jinsheng1, DUAN Yanhui2   

  1. 1. School of Aeronautics, Northwestern Polytechnical University, Xi'an 710072, China;
    2. Computational Aerodynamics Institute, China Aerodynamics Research and Development Centre, Mianyang 621000, China
  • Received:2015-02-06 Revised:2015-05-04 Online:2015-12-15 Published:2015-05-05

摘要:

首先分析了几何外形和相对厚度对超声速翼型气动特性的影响。基于遗传算法(GA)和气动力快速工程算法,对于相对厚度为3.5%的多边形翼型进行优化设计,多边形翼型的优化外形趋于四边形,最大厚度点后移到翼型弦线的60%左右,随着迎角或者马赫数增大下翼面会变薄,上翼面变厚,最大厚度点相应稍有后移。对于相对厚度为4%的双圆弧翼型,采用两步优化设计方法,第1步优化结合基于B样条的类别形状函数变换(CST)参数化方法与小波分解方法,实现几何外形的局部控制与光顺处理,并且采用本征正交分解(POD)代理模型降低优化过程中流场计算的工作量;第2步优化采用基于Navier-Stokes方程的最速下降法(SDA),修正第1步优化中代理模型和小波光顺引入的误差;优化设计得到的翼型近似为四边形,其相对厚度最大点后移到翼型弦线的60%~65%处,升阻比可以提高7%。

关键词: 超声速翼型, 优化设计, 代理模型, 参数化, 小波分解

Abstract:

Firstly, the aerodynamic performance of supersonic airfoils with different geometric shapes and relative thickness is compared and analyzed. Secondly, design optimization is implemented for the polygonal airfoils of 3.5% relative thickness by combining genetic algorithm (GA) and rapid aerodynamic engineering algorithm. The design optimization process makes the polygonal airfoil approach to be a quadrilateral in shape with the maximum thickness location moving to about 60% of the chord. As the angle of attack or Mach number increases, the lower surface will become thinner, the upper surface will become thicker and the maximum thickness location will move backward slightly. Furthermore, for the biconvex airfoil of 4% relative thickness, a two-step design optimization method is used. In the first step, class and shape transformation (CST) based on B-spline basis function is used combined with wavelet decomposition to enhance the local control and fairing abilities. A surrogate model based on proper orthogonal decomposition (POD) is chosen to reduce the computational workload. The second step optimization employs the steepest decent algorithm (SDA) based on the Navier-Stokes equations to correct the errors caused by POD method and wavelet decomposition in the first step optimization. The optimized airfoil approximates to be a quadrilateral in shape, with the maximum thickness location moving to 60%-65% of the chord and lift-to-drag ratio increasing by 7%.

Key words: supersonic airfoils, design optimization, surrogate model, parameterization, wavelet decomposition

中图分类号: