电子学报 ›› 2020, Vol. 48 ›› Issue (7): 1361-1368.DOI: 10.3969/j.issn.0372-2112.2020.07.015

• 学术论文 • 上一篇    下一篇

求解高维混合指标优化问题的交互式进化计算

郭广颂1, 陈良骥2, 文振华3, 张勇4   

  1. 1. 郑州航空工业管理学院智能工程学院, 河南郑州 450046;
    2. 天津工业大学机械工程学院, 天津 300387;
    3. 郑州航空工业管理学院航空工程学院, 河南郑州 450046;
    4. 中国矿业大学信息与控制工程学院, 江苏徐州 221116
  • 收稿日期:2019-06-03 修回日期:2019-12-16 出版日期:2020-07-25
    • 通讯作者:
    • 陈良骥
    • 作者简介:
    • 郭广颂 男,1978年9月生,吉林集安人,副教授、硕导,主要研究方向为进化优化与智能控制.E-mail:guogs78@126.com
    • 基金资助:
    • 国家自然科学基金 (No.51975539,No.61876185); 航空科学基金 (No.2018ZD55008); 河南省教育厅科技重点研究项目 (No.19A460030)

Sloving Multidimensional Optimization Problems with Hybird Indices by Interactive Evolutionary Computation

GUO Guang-song1, CHEN Liang-ji2, WEN Zhen-hua3, ZHANG Yong4   

  1. 1. School of Intelligent Engineering, Zhengzhou University of Aeronautics, Zhengzhou, Henan 450046, China;
    2. School of Mechanical Engineering, Tianjin Polytechnic University, Tianjin 300387, China;
    3. School of Aeronautical Engineering, Zhengzhou University of Aeronautics, Zhengzhou, Henan 450046, China;
    4. School of Information and Control Engineering, China University of Mining and Technology, Xuzhou, Jiangsu 221116, China
  • Received:2019-06-03 Revised:2019-12-16 Online:2020-07-25 Published:2020-07-25
    • Corresponding author:
    • CHEN Liang-ji
    • Supported by:
    • National Natural Science Foundation of China (No.51975539, No.61876185); Aeronautical Science Foundation of China, ASFC (No.2018ZD55008); Science and Technology Key Research Program of Education Department of Henan Province (No.19A460030)

摘要: 同时存在区间显式指标和模糊隐式指标的高维混合指标优化问题是一类难以求解的不确定多目标优化问题.针对该问题,首先,分别对高维显式指标和隐式指标的主要参数按确定性多目标优化,根据获得的相关权值,将高维显式指标和高维隐式指标分别降维成一维等效区间适应值和一维等效模糊适应值,二者合成个体等效指标体;然后,依据等效指标体的占优情况,通过确定自适应参考点和偏好区域面积选择个体;最后,在大规模种群NSGA-II范式下,采用隐式指标估计策略和种群聚类方法实现交互式进化优化算法.将本文算法应用于2种混合性能指标优化问题,验证所提算法的有效性和泛化性.

关键词: 进化优化, 混合性能指标, 遗传算法, 交互

Abstract: The multidimensional hybrid indices optimization problem is a kind of uncertainty multi-objective optimization problems that is difficult to solve. First, we can get relevant weights by optimizing the main parameters of explicit and implicit indices. According to these weights, multidimensional explicit indices can be reduced to an equivalent-interval fitness, and multidimensional implicit indices can be reduced to an equivalent-fuzzy fitness. Equivalent-interval fitness and equivalent-fuzzy fitness can be synthesized to an equivalent-index body. Then, we select advantage individual on the basis of equivalent-index bodies dominant situation according to adaptive reference point and preference area size. Finally, we adopt an implicit-indices estimation strategy with cluster method to realize interactive evolutionary algorithm within the framework of NSGA-II. The proposed algorithm is applied to two optimization problems with hybrid indices, and the results validate its efficiency and generalization.

Key words: evolutionary optimization, hybrid indices, genetic algorithms, interaction

中图分类号: