北京航空航天大学学报 ›› 2021, Vol. 47 ›› Issue (5): 894-903.doi: 10.13700/j.bh.1001-5965.2020.0058

• 论文 • 上一篇    下一篇

改进自适应人工免疫算法求解函数优化问题

孟亚峰1, 王涛2, 李泽西3, 蔡金燕1, 朱赛1, 韩春辉1   

  1. 1. 陆军工程大学 电子与光学工程系, 石家庄 050003;
    2. 中国人民解放军 63769部队, 西安 710000;
    3. 陆军装备部驻西安地区军事代表局, 西安 710000
  • 收稿日期:2020-02-28 发布日期:2021-05-28
  • 通讯作者: 王涛 E-mail:wangtao920110@126.com
  • 作者简介:孟亚峰,男,博士,副教授,硕士生导师。主要研究方向:电子装备可靠性理论、电子装备故障检测与自修复理论;王涛,男,博士,工程师。主要研究方向:电子装备故障检测与自修复理论、可靠性评估理论、非线性规划问题优化;蔡金燕,女,博士,教授,博士生导师。主要研究方向:电子装备可靠性理论、电子装备故障检测与自修复理论、非线性规划问题优化。
  • 基金资助:
    国家自然科学基金(61601495)

Improved adaptive artificial immune algorithm for solving function optimization problems

MENG Yafeng1, WANG Tao2, LI Zexi3, CAI Jinyan1, ZHU Sai1, HAN Chunhui1   

  1. 1. Department of Electronic and Optical Engineering, Army Engineering University, Shijiazhuang 050003, China;
    2. 63769 Unit of PLA, Xi'an 710000, China;
    3. Military Representative Bureau of Army Equipment Department in Xi'an, Xi'an 710000, China
  • Received:2020-02-28 Published:2021-05-28

摘要: 为克服经典人工免疫算法(AIA)在函数优化过程中存在的计算量大、收敛精度不高和收敛速度较慢等不足,引入多个自适应免疫算子,提出了一种改进自适应人工免疫算法(IAAIA)。在经典人工免疫算法中,引入迭代次数对抗体激励度计算算子进行自适应设计,引入种群抗体平均激励度与抗体激励度对免疫选择算子、克隆算子、变异算子与克隆抑制算子进行自适应设计,提升人工免疫算法的收敛速度、收敛精度和稳定性。选择9个典型测试函数作为实验对象,同时选择4种典型人工免疫算法作为对比算法优化实验函数,对比实验结果表明了改进的自适应人工免疫算法在求解函数优化问题的有效性和优越性。

关键词: 人工免疫算法(AIA), 免疫系统, 自适应, 免疫算子, 函数优化

Abstract: In order to overcome the shortcomings of Artificial Immune Algorithm (AIA) used in the function optimization process, such as huge calculation amount, low convergence accuracy and slow convergence speed, multiple adaptive immune operators are introduced, and an Improved Adaptive Artificial Immune Algorithm (IAAIA) is proposed. In the classic AIA, antibody excitation calculation operator is adaptively designed by introducing the number of iterations, and immune selection operator, clone operator, mutation operator and clonal inhibitory operator are adaptively designed by introducing antibody population average excitation and antibody excitation, which can improve the convergence accuracy, convergence speed and stability of AIA. Nine kinds of typical and widely used functions are chosen as experiment function, and four kinds of typical AIAs are selected as comparative algorithms to optimize the experiment functions. The comparative experiment results indicate the effectiveness and superiority of the IAAIA for solving function optimization problems.

Key words: Artificial Immune Algorithm (AIA), immune system, adaptive, immune operator, function optimization

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