• 北大核心期刊(《中文核心期刊要目总览》2017版)
  • 中国科技核心期刊(中国科技论文统计源期刊)
  • JST 日本科学技术振兴机构数据库(日)收录期刊

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

求解约束函数优化的改进型果蝇视觉进化神经网络

刘健 张著洪

刘健, 张著洪. 求解约束函数优化的改进型果蝇视觉进化神经网络[J]. 微电子学与计算机, 2022, 39(4): 41-48. doi: 10.19304/J.ISSN1000-7180.2021.1090
引用本文: 刘健, 张著洪. 求解约束函数优化的改进型果蝇视觉进化神经网络[J]. 微电子学与计算机, 2022, 39(4): 41-48. doi: 10.19304/J.ISSN1000-7180.2021.1090
LIU Jian, ZHANG ZhuHong. Fly visual evolutionary neural network solving constrained function optimization[J]. Microelectronics & Computer, 2022, 39(4): 41-48. doi: 10.19304/J.ISSN1000-7180.2021.1090
Citation: LIU Jian, ZHANG ZhuHong. Fly visual evolutionary neural network solving constrained function optimization[J]. Microelectronics & Computer, 2022, 39(4): 41-48. doi: 10.19304/J.ISSN1000-7180.2021.1090

求解约束函数优化的改进型果蝇视觉进化神经网络

doi: 10.19304/J.ISSN1000-7180.2021.1090
基金项目: 

国家自然科学基金资助项目 62063002

国家自然科学基金资助项目 61563009

详细信息
    作者简介:

    刘健  男,(1995-),硕士研究生.研究方向为智能优化算法

    通讯作者:

    张著洪(通讯作者)   男,(1966-),博士,教授,博士生导师.研究方向为数据科学与计算智能、深度学习等. E-mail: zhzhang@gzu.edu.cn

  • 中图分类号: TP18

Fly visual evolutionary neural network solving constrained function optimization

  • 摘要:

    具有广泛工程应用背景的强非线性约束优化是最优化领域极为困难的科技问题,如何寻找快速有效的优化算法求解其全局最优化解,仍然是该问题研究的关键.为此,针对强非线性约束函数优化求解难的问题,融合果蝇视觉系统的信息处理机制与种群进化思想,提出一种基于状态矩阵转移的改进型果蝇视觉进化神经网络.模型设计中,将候选解视为状态,构建以状态作为元素的状态矩阵,进而将状态矩阵中各元素对应的目标值形成的灰度图视为输入;依据果蝇视觉系统的分层视觉信息处理特性,构建能有效处理约束条件的改进型果蝇视觉前馈神经网络,进而将其输出作为状态转移的全局学习率;依据鲸鱼捕食的行为特性建立转移状态的更新策略.由此,获得仅含两个可调参数且计算复杂度仅由输入灰度图分辨率确定的视觉进化神经网络.比较性的数值实验表明,此神经网络的寻优质量具有明显优势,对工程优化问题的解决具有重要参考价值.

     

  • 图 1  IFVNN的结构设计示意图

    Figure 1.  Structure design drawing of IFVNN

    图 2  IFVENN的算法流程图

    Figure 2.  IFVENN algorithm flow chart

    图 3  各算法关于g09、g16、g24的箱线图比较;A、B、C、D、E依次指代WOA、CGWO、ODPSO、NDE及IFVENN.

    Figure 3.  Comparison of the box plots of each algorithm for g09, g16, and g24; A, B, C, D, and E refer to WOA, CGWO, ODPSO, NDE and IFVENN in turn.

    图 4  算法求解CSIDP的搜索曲线比较

    Figure 4.  Comparison of search curves of various algorithms to solve CSIDP

    表  1  算法独立运行25次后获得的统计结果比较

    Table  1.   Comparison of statistical results obtained after the algorithm runs 25 times independently

    Problem/ Optimal value Features WOA CGWO ODPSO NDE IFVENN
    g01/-15 Mean -14.413 11 -14.851 58 -14.990 28 -15 -15
    Std 1.77E-01 1.33E-01 3.58E-03 0.00E+00 0.00E+00
    g02/-0.803619 Mean -0.781 43 -0.781 64 -0.768 75 -0.790 52 -0.803 53
    Std 9.98E-03 1.33E-02 9.69E-03 2.51E-02 8.99E-05
    g03/-1 Mean -0.998 19 -0.999 73 -0.981 17 -1 -1
    Std 7.80E-04 1.29E-04 8.93E-03 0.00E+00 0.00E+00
    g04/-30665.53 Mean -30 653.924 -30 664.285 -30 664.675 -30 665.538 -30 665.538
    Std 4.32E+00 2.26E-01 9.28E-06 1.26E-07 0.00E+00
    g05/5126.4981 Mean 5 198.264 0 5 137.560 4 5 182.024 0 5 127.028 9 5 126.506 1
    Std 4.98E+01 1.44E+01 5.56E+01 8.93E-01 1.79E-03
    g06/-6 961.813 Mean -6 961.470 2 -6 959.091 6 -6 961.713 8 -6 961.813 8 -6 961.813 8
    Std 1.72E-01 1.41E+00 3.71E-02 0.00E+00 0.00E+00
    g07/24.306 209 Mean 26.854 04 27.286 34 24.405 89 25.541 38 24.389 3
    Std 1.82E+00 1.84E+00 3.35E-01 8.91E-01 2.80E-02
    g08/-0.095 825 Mean -0.095 83 -0.095 83 -0.095 83 -0.095 83 -0.095 83
    Std 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E-00
    g09/680.630 05 Mean 680.757 0 680.639 3 682.013 2 681.164 4 680.634 1
    Std 1.02E-01 6.92E-02 1.12E+00 1.71E+00 1.52E-03
    g10/7 049.330 7 Mean 7 592.310 0 7 114.987 6 7 216.025 9 7 171.707 6 7 099.347 9
    Std 3.64E+02 4.15E+01 1.57E+02 6.61E+01 1.23E+01
    g11/0.75 Mean 0.75 0.75 0.75 0.75 0.75
    Std 0.00E+00 0.00E+00 0.00E+00 0.00E+00 0.00E+00
    g12/-1 Mean -1.000 0 -1.000 0 -1.000 0 -1.000 0 -1.000 0
    Std 00E+00 0.00E+00 0.00E-00 0.00E+00 0.00E+00
    g13/0.0539415 Mean 0.675 2 0.099 5 0.304 4 0.245 2 0.278 2
    Std 1.28E-01 1.08E-01 1.96E-01 1.16E-01 1.10E-01
    g14/-47.76488 Mean -46.750 1 -47.410 2 -45.935 7 -47.583 7 -47.049 3
    Std 5.21E-01 2.06E-01 5.71E-01 2.35E-02 7.41E-02
    g15/961.71502 Mean 961.716 1 961.716 6 961.812 2 961.719 0 961.719 2
    Std 1.09E-03 1.22E-03 8.05E-02 6.62E-03 1.44E-02
    g16/-1.905155 Mean -1.902 1 -1.902 1 -1.876 6 -1.905 1 -1.905 15
    Std 6.63E-04 9.41E-04 2.79E-02 2.63E-05 2.73E-09
    g17/8 853.539 6 Mean 8 853.949 6 8 853.702 0 8 880.316 0 8 853.727 6 8 853.539 8
    Std 2.42E-01 1.17E-01 2.02E+01 2.85E-01 0.00E+00
    g18/-0.866 025 Mean -0.669 1 -0.863 56 -0.828 67 -0.854 05 -0.865 5
    Std 9.41E-02 3.78E-03 2.25E-02 7.40E-03 1.70E-03
    g19/32.655 59 Mean 40.554 7 35.572 13 35.690 7 41.185 79 40.965 4
    Std 2.49E+00 7.27E-01 1.18E+00 1.55E+00 3.12E+00
    g21/193.724 Mean 357.367 8 297.339 56 260.020 6 360.556 62 277.392 3
    Std 4.70E+01 4.78E+01 4.37E+01 8.49E+01 3.75E+02
    g23/-400.055 Mean -182.809 7 267.184 00 -318.564 36 -48.517 27 -315.447 7
    Std 1.33E+01 3.75E+02 5.77E+00 7.39E+01 7.68E+01
    g24/-5.508 01 Mean -5.508 01 -5.508 00 -5.504 4 -5.508 01 -5.508 01
    Std 5.10E-06 8.36E-07 2.17E-03 0.00E+00 0.00E+00
    w/t/l 15/7/0 12/7/3 14/7/1 8/13/1 /
    下载: 导出CSV

    表  2  算法的统计结果比较

    Table  2.   Comparison of statistical results of various algorithms

    算法 最好值 均值 最差值 标准差
    WOA 22.918 7 23.058 2 23.409 8 1.24E-01
    CGWO 22.902 0 22.979 0 23.206 9 7.40E-02
    ODPSO 22.913 8 23.029 9 23.209 1 8.09E-02
    NDE 22.862 9 22.908 1 23.104 3 6.12E-02
    IFVENN 22.863 2 22.879 2 22.897 6 8.04E-03
    下载: 导出CSV
  • [1] SARGHINI F, DE VIVO A. Application of constrained optimization techniques in optimal shape design of a freezer to dosing line splitter for ice cream production[J]. Food Engineering Reviews, 2021, 13(1): 262-273. DOI: 10.1007/s12393-020-09258-5.
    [2] 陈宝林. 最优化理论与算法[M]. 北京: 清华大学出版社, 1989.

    CHEN B L. Optimization theory and algorithm[M]. Beijing: Tsinghua University Press, 1989.
    [3] KARADENIZ A, ELK Y. Whale optimization algorithm for numerical constrained optimization[J]. Academic Platform Journal of Engineering and Science, 2020, 8(8): 547-554. DOI: 10.21541/apjes.551526.
    [4] KOHLI M, ARORA S. Chaotic grey wolf optimization algorithm for constrained optimization problems[J]. Journal of Computational Design and Engineering, 2018, 5(4): 458-472. DOI: 10.1016/j.jcde.2017.02.005.
    [5] 梁静, 葛士磊, 瞿博阳, 等. 求解电力系统经济调度问题的改进粒子群优化算法[J]. 控制与决策, 2020, 35(8): 1813-1822. DOI: 10.13195/j.kzyjc.2018.1490.

    LIANG J, GE S L, QU B Y, et al. Improved particle swarm optimization algorithm for solving power system economic dispatch problem[J]. Control and Decision, 2020, 35(8): 1813-1822. DOI: 10.13195/j.kzjc.2018.1490.
    [6] MOHAMED A W. A novel differential evolution algorithm for solving constrained engineering optimization problems[J]. Journal of Intelligent Manufacturing, 2018, 29(3): 659-692. DOI: 10.1007%2Fs10845-017-1294-6.
    [7] ZHANG Z H, YUE S G, ZHANG G P. Fly visual system inspired artificial neural network for collision detection[J]. Neurocomputing, 2015, 153: 221-234. DOI: 10.1016/j.neucom.2014.11.033.
    [8] ZHANG Z H, LI L, LU J X. Gradient-based fly immune visual recurrent neural network solving large-scale global optimization[J]. Neurocomputing, 2021, 454: 238-253. DOI: 10.1016/j.neucom.2021.05.002.
    [9] ZHANG Z H, XIAO T Y, QIN X C. Fly visual evolutionary neural network solving large-scale global optimization[J]. International Journal of Intelligent Systems, 2021, 36(11): 6680-6712. DOI: 10.1002/int.22564.
    [10] MIRJALILI S, LEWIS A. The whale optimization algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67. DOI: 10.1016/j.advengsoft.2016.01.008.
    [11] 王涛, CHELLALI R. 非线性权重和收敛因子的鲸鱼算法[J]. 微电子学与计算机, 2019, 36(1): 11-15. DOI: 10.19304/j.cnki.issn1000-7180.2019.01.003.

    WANG T, CHELLALI R. Whale optimization algorithm with nonlinear weight and convergence factor[J]. Microelectronics & Computer, 2019, 36(1): 11-15. DOI: 10.19304/j.cnki.issn1000-7180.2019.01.003.
    [12] BRAXUN D, SZCZECINSKI N S, KULJ G, et al. Artificial compound eye and synthetic neural system for motion recognition[C]// 7th International Conference on Biomimetic and Biohybrid Systems. Paris, France: Springer International Publishing, 2018: 52-63.
    [13] J. J. Liang, Thomas Philip Runarsson, Efren Mezura-Montes, Maurice Clerc, P. N. Suganthan, Carlos A. Coello Coello, K. Deb, Problem Definitions and Evaluation Criteria for the CEC 2006 Special Session on Constrained Real-Parameter Optimization. Technical Report, Nanyang Technological University, Singapore, March 2006, http://www5.zzu.edu.cn/cilab/fblw/jsbg.htm
    [14] GU L, YANG R J, THO C H, et al. Optimisation and robustness for crashworthiness of side impact[J]. International Journal of Vehicle Design, 2001, 26(4): 348-360. DOI: 10.1504/IJVD.2001.005210.
  • 加载中
图(4) / 表(2)
计量
  • 文章访问数:  16
  • HTML全文浏览量:  17
  • PDF下载量:  0
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-09-14
  • 修回日期:  2021-10-19
  • 网络出版日期:  2022-05-12

目录

    /

    返回文章
    返回