Please wait a minute...
浙江大学学报(工学版)
JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE)  2017, Vol. 51 Issue (7): 1405-1411    DOI: 10.3785/j.issn.1008-973X.2017.07.019
Aeronautics and Astronautics Technology     
Aerodynamic optimization design of airfoil configurations based on cascade feedforward neural network
ZHANG Xuan-wu1,2, ZHENG Yao1,2, YANG Bo-wei1,2, ZHANG Ji-fa1,2
1. Center for Engineering and Scientific Computation, Zhejiang University, Hangzhou 310027, China;
2. School of Aeronautics and Astronautics, Zhejiang University, Hangzhou 310027, China
Download:   PDF(1114KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

The cascade feedforward neural network was applied as a surrogate model in order to solve the problem that aerodynamic optimization based on the genetic algorithm consumed larger amounts of computing resource and time. The computing resource and time can be reduced. The class-shape function transformation (CST) method was used to parameterize an airfoil. The samples of airfoils were randomly generated in a constrained space, and the samples were employed to train the cascade feedforward network. The trained cascade feedforward network with required accuracy was served as the surrogate model to replace the fluid dynamics solver for computing fluid field around an airfoil. The lift-drag ratio that was correspondingly calculated by the cascade feedforward network and fluid dynamics solver was employed as the objective function by using the single objective genetic algorithm. The CST parameters of the sharp function of the airfoil were selected as the genes of an individual in order to optimize the original airfoil. The numerical experiments showed that the cascade feedforward network provided the required accuracy with a significant reduction of computing time under a specific optimization objective.

Received: 30 May 2016      Published: 08 July 2017
CLC:  V212  
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Cite this article:

ZHANG Xuan-wu, ZHENG Yao, YANG Bo-wei, ZHANG Ji-fa. Aerodynamic optimization design of airfoil configurations based on cascade feedforward neural network. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2017, 51(7): 1405-1411.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2017.07.019     OR     http://www.zjujournals.com/eng/Y2017/V51/I7/1405


基于级联前向网络的翼型优化设计

针对应用遗传算法进行气动优化需要巨大计算量和计算时间的问题，采用将级联前向神经网络作为流场计算的代理模型的方法，能够减少计算量和计算时间.采用类别形状函数 （CST）参数化方法，对翼型进行参数化，在限定的范围内随机生成翼型样本，应用样本对级联前向神经网络进行训练，用训练后精度达到要求的级联前向网络作为翼型流场数值计算的代理模型.采用单目标的遗传算法，将级联前向网络和流场数值计算的升阻比作为目标函数，将翼型的CST参数作为单位个体的所有基因，对标准翼型进行优化.数值试验表明，用级联前向网络计算出的升阻比可以达到进行气动优化所需要的精度要求，对于给定的优化目标可以节约大量的计算时间.

[1] BELEGUNDU A, CONSTANS E, SALAGAME R, et al. Multi-objective optimization of laminated ceramic composites using genetic algorithms [C]//5th Symposium on Multidisciplinary Analysis and Optimization. Portugal: [s.n.], 1994: 312-315.
[2] LEE J, HAJELA P. Parallel genetic algorithm implementation in multidisciplinary rotor blade design [J]. Journal of Aircraft, 1995, 33(5): 962-969.
[3] 李静,高正红,黄江涛,等.基于分布式粒子群算法的翼型优化设计[J].空气动力学学报,2011, 29(04):464-469. LI Jing, GAO Zheng-hong, HUANG Jiang-tao, et al. Airfoil optimization based on distributed particle swarm algorithm [J]. Acta Aerodynamica Sinica, 2011,29(04): 464-469.
[4] 周盛强,向锦武.基于Hopfield网络的飞机设计[J].北京航空航天大学学报,2006, 32(6): 675-679. ZHOU Sheng-qiang, XIANG Jin-wu. Hopfield network based approach to aircraft design [J]. Journal of Beijing University of Aeronautics and Astronautics, 2006,32(6): 675-679.
[5] 朱莉,高正红.基于神经网络的翼型优化设计方法研究[J].航空计算技术,2007,37(3): 33-36. ZHU Li, GAO Zheng-hong. Aero dynamic optimization design of airfoil based on neural networks [J]. Aeronautical Computing Technique, 2007, 37(3): 33-36.
[6] 杨华,姚卫星.基于径向基函数的机翼二维气动代理模型设计[J].计算力学学报,2008,25(6): 797-802. YANG Hua, YAO Wei-xing. 2D surrogate model of wing lift distribution based on radial basis function [J]. Chinese Journal of Computational Mechanics, 2008,25(6): 797-802.
[7] 李静,高正红,黄江涛,等.基于CST参数化方法气动优化设计研究[J].空气动力学学报,2012, 30(04): 443-449. LI Jing, GAO Zheng-hong, HUANG Jiang-tao, et al. Aerodynamic optimization system based on CST technique [J]. Acta Aerodynamica Sinica, 2012, 30(04): 443-449.
[8] HAGAN M T, DEMUTH H B, BEALE M. Neural network design [M]. Beijing: China Machine Press, 2002: 357.
[9] LEVEN S. The roots of backpropagation: from ordered derivatives to neural networks and political forecasting [M]. New York: Wiley, 1996, 9(3): 543-544.
[10] SIMPSON T W, BOOKER A J, GHOSH D, et al. Approximation methods in multidisciplinary analysis and optimization: a panel discussion [J]. Structural and Multidisciplinary Optimization, 2004, 27(5):302-313.
[11] KULFAN B M, BUSSOLETTI J E. Fundamental parametric geometry representations for aircraft component shapes [C]//11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference. [S. l.]: AIAA, 2006: 3-12.
[12] 王小川. MATLAB神经网络43个案例分析[M].北京：北京航空航天大学出版社, 2013: 34-39.
[13] MITCHELL T M, CARBONELL J G, MICHALSKI R S. Machine learning [M].[S. l.]: McGraw, 1986: 417-433.
[14] 尹强, 高正红. 外形参数化方法对气动优化过程的影响[J]. 科学技术与工程, 2012, 12(14): 3394-3398. YI Qiang, GAO Zheng-hong. Effect of shape parameterization on aerodynamic shape optimization [J]. Science Technology and Engineering, 2012, 12(14):3394-3398.
[15] RAE 2822 transonic airfoil: study #4. 2002-05-03. http://www.grc.nasa.gov/WWW/wind/valid/raetaf/raetaf04/raetaf04.html.
[16] NACA 0012 transonic airfoil. 1. Inviscid transonic flow. 2003-08-12. http://xoptimum.narod.ru/eng/results/compressible/naca0012/naca0012.htm.
[17] 孙俊峰,刘刚,江雄,等.基于Kriging模型的旋翼翼型优化设计研究[J].空气动力学学报,2013, 31(04): 437-441. SUN Jun-feng, LIU Gang, JIANG Xiong, et al. Research of rotor airfoil design optimization based on the Kriging model [J]. Acta Aerodynamica Sinica, 2013, 31(04): 437-441.
[1] CHEN Ran, LIAN Guang-yao, QIN Zi-long, LI Bao-chen, ZHANG Xi-shan. Fault injection approach for line replaceable module based on hierarchical model[J]. JOURNAL OF ZHEJIANG UNIVERSITY (ENGINEERING SCIENCE), 2017, 51(7): 1390-1396.