计算物理 ›› 2012, Vol. 29 ›› Issue (3): 326-332.

• 论文 • 上一篇    下一篇

EGO方法的训练算法及应用

邓枫1,2, 覃宁1,2, 伍贻兆1   

  1. 1. 南京航空航天大学航空宇航学院, 江苏 南京, 210016;
    2. Department of Mechanical Engineering, University of Sheffield, Sheffield, S1 3JD, UK
  • 收稿日期:2011-07-15 修回日期:2011-11-26 出版日期:2012-05-25 发布日期:2012-05-25
  • 作者简介:邓枫(1982-),男,博士生,主要从事计算流体力学、气动优化研究,E-mail:dengfengl@yahoo.cn

Training Algorithms for EGO Method and Applications

DENG Feng1,2, QIN Ning1,2, WU Yizhao1   

  1. 1. College of Aerospace Engineering, Nanjing University of Aeronautics & Astronautics, Nanjing 210016, China;
    2. Department of Mechanical Engineering, University of Sheffield, Sheffield, S1 3JD, UK
  • Received:2011-07-15 Revised:2011-11-26 Online:2012-05-25 Published:2012-05-25

摘要: 针对高效全局优化(Efficient Global Optimization,简称EGO)方法的训练问题,选择元启发式(Meta-heuristic)算法、随机取样算法以及低频序列算法,并选用三个无约束、两个带约束解析优化算例以及两个气动优化算例,对这三类训练算法进行详细地比较研究,发现在元启发式算法中差分进化算法最具应用潜力,而低频序列算法可以有效降低EGO方法的随机性,其中Faure序列平均性能最优.

关键词: 计算流体力学, 气动外形优化, 克里金模型, 全局优化

Abstract: Three kinds of training algorithms for efficient global optimization(EGO) method are investigated.A kind of training algorithm based on low-discrepancy sequences is proposed to reduce randomness of EGO method.Performance of EGO method depends on a good training algorithm.Since training problems in EGO are non-convex and non-smooth,meta-heuristic algorithms,random algorithm and low-discrepancy sequences are chosen to address five benchmark optimization problems and two aerodynamic shape optimization problems.In these problems,differential evolution algorithm was found the best in meta-heuristic algorithms.Training algorithm based on low-discrepancy sequences can effectively reduce randomness of EGO method and Faure sequence has the best performance.

Key words: computational fluid dynamic, aerodynamic shape optimization, Kriging model, global optimization

中图分类号: