›› 2015, Vol. 28 ›› Issue (6): 51-.

• 论文 • 上一篇    下一篇

基于GA的BP神经网络在多目标优化中的应用

苏宇逍   

  1. (中国空空导弹研究院 总体部,河南 洛阳 471009)
  • 出版日期:2015-06-15 发布日期:2015-06-20
  • 作者简介:苏宇逍(1980—),男,工程师。研究方向:导弹总体性能测试。E-mail:snail_s@aliyun.com

Application of BP Neural Network Based on Genetic Algorithm in Multi-objective Optimization

SU Yuxiao   

  1. (General Department,China Airborne Missile Academy,Luoyang 471009,China)
  • Online:2015-06-15 Published:2015-06-20

摘要:

针对传统多目标优化的求解方法通常存在目标权值主观性大,优化目标仅为各目标加权和以及在求解过程中各目标优化的不可操作性等问题,文中提出了一种新颖的多目标优化算法,其将改进后的遗传算法与BP神经网络融合,提出了基于遗传算法的BP神经网络融合算法。该算法将遗传算法与BP神经网络算法相结合,充分发挥遗传算法的全局搜索能力优势和BP算法的局部搜索能力特点,使得多目标优化问题得以求解,加快收敛速度,从而提高了收敛精度。

关键词: 遗传算法, BP神经网络, 多目标优化, 算法融合

Abstract:

For the existing problems in traditional more target optimization method such as target weighted value subjectivity big,optimization target only for each target weighted and in solution process optimization of not operation in the each target,this paper has proposed a novelty more target optimization algorithm,and it improved genetic algorithm and BP neural network based on genetic algorithm of BP neural network algorithm. The algorithm combining genetic algorithm with BP neural network algorithm,gives full advantage of global search capability of genetic algorithm and BP algorithm in local search feature,making multi-objective optimization problem can be solved,converging speed and improving accuracy.

Key words: genetic algorithm;BP neural network algorithm;MOP;algorithm convergence

中图分类号: 

  • TP18