摘要
为了更加准确地判断环境是否发生变化并快速追踪动态多目标规划问题(dynamicmulti-objectiveoptimization problem,DMOP)当前时刻的Pareto前沿,提出了一种基于数据流的Kolmogorov-Smirnov(K-S)变化检测的动态多目标规划(DSK-SDMOP)算法。该算法以NSGA-Ⅱ为基础,通过数据流建立2个时刻的检验窗口,再利用K-S检验基于数据流的Pareto最优前沿是否发生变化,检测2个窗口的数据是否服从同一分布来判断环境是否发生变化,并就环境变化的剧烈程度实行相应的应答机制,以提高对环境的适应程度。利用基于数据流的K-S检测方法,对环境变化不会过于敏感,而且不用提前假设对应目标值的分布,易于操作。通过5个动态多目标规划标准测试函数对该算法进行测试,并和现有的2种算法进行对比分析,结果表明该算法处理动态多目标规划问题具有良好的性能。
In order to more accurately determine whether the environment has changed and quickly track the Pareto front of the dynamic multi-objective programming problem at the current moment,this paper proposes a dynamic multi-objective programming algorithm based on Kolmogorov-Smirnov(K-S)change detection of data stream(DSK-SDMOP).Based on NSGA-Ⅱ,the algorithm establishes two test windows through the data stream,and then uses K-S test to detect whether the data of the two windows obey the same distribution to determine whether the environment changes,and implements the corresponding response mechanism according to the intensity of environmental changes.The proposed algorithm is tested by five standard test functions of dynamic multi-objective programming,and compared with two existing algorithms,the results show that the proposed algorithm has good performance in dealing with dynamic multi-objective programming problems.
作者
张涛
周晨
杜锋
陈芳
刘瑞林
ZHANG Tao;ZHOU Chen;DU Feng;CHEN Fang;LIU Ruilin(School of Information and Mathematics,Yangtze University,Jingzhou 434023,Hubei;School of Mathematics and Physics,Jingchu University of Technology,Jingmen 448000,Huibei)
出处
《长江大学学报（自然科学版）》
2024年第1期109-116,共8页
Journal of Yangtze University(Natural Science Edition)
基金
国家自然科学基金项目“动态多目标规划问题深度学习算法与应用研究”(62373066)。
