北京航空航天大学学报 ›› 2006, Vol. 32 ›› Issue (06): 675-679.

• 论文 • 上一篇    下一篇

基于Hopfield网络的飞机设计

周盛强, 向锦武   

  1. 北京航空航天大学 航空科学与工程学院, 北京 100083
  • 收稿日期:2005-06-20 出版日期:2006-06-30 发布日期:2010-09-20
  • 作者简介:周盛强(1975-),男,江西安福人,博士生, zsq@ase.buaa.edu.cn.
  • 基金资助:

    新世纪优秀人才支持计划资助项目(NCET2042010)

Hopfield network based approach to aircraft design

Zhou Shengqiang, Xiang Jinwu   

  1. School of Aeronautic Science and Technology, Beijing University of Aeronautics and Astronautics, Beijing 100083, China
  • Received:2005-06-20 Online:2006-06-30 Published:2010-09-20

摘要: Hopfield神经网络与增广拉格朗日乘子法相结合来求解非线性约束优化.神经网络作为求解乘子法的子问题的动力学方法,仅需计算一阶导数.引入逐渐衰减的高斯噪声信号构造随机神经网络.同时针对随机网络受初始温度制约,跳出局部极小值能力有限的问题,网络运行采用结合模拟退火的欧拉法.用该方法对某喷气教练机进行总体优化设计,结果表明,算法的数值稳定性较好,求解精度高.并基于拉氏乘子提供的约束敏度信息,做了设计要求权衡.最后研究了某型干线旅客机的机翼气动/结构综合设计问题.

Abstract: Combining augmented lagrange multiplier(ALM) method to Hopfield neural network(HNN), was proposed to solve nonlinear constrained optimization. HNNis taken as a dynamic approach for minimization subproblem in ALM method, only needing the first derivative of the Lagrange function. The random neural network was extended by adding Gaussian noise gradually reducing with the temperature, whose ablity escapeing from the attraction of the localminimum points was limitedby the initial temperature. Combined with the simulated annealing, an improved algorithm for Euler method was presented for numerical implementation of the network. The approach was applied to jet trainer aircraft preliminary optimization. The results show that the computing process is stable and the optimal result is enough precise. The trade-off of design requirements was studied through the Lagrange multipliers. An aerodynamic/structural design optimization for the wing of a trunkeliner was studied.

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发