CAJ | 학술논문

很多现实的优化问题都是动态多目标问题，这类问题不仅具有多个目标，并且也受环境的影响不断变化。本文基于引导个体的预测策略提出一种新的求解动态多目标优化问题的策略。通过记录每次环境变化初始时和种群自主进化一小段时间后种群中心点位置的前后变化，预测最优解的所在方向。同时根据在该方向上均匀分布的若干检测个体，选出一串非支配的个体作为当前环境下的引导个体。为了避免陷入局部最优，在选出的引导个体周围一个小的区域半径内随机产生若干伴随引导个体。实验结果表明，新策略具有更快的响应环境变化的能力。
흔다현실적우화문제도시동태다목표문제，저류문제불부구유다개목표，병차야수배경적영향불단변화。본문기우인도개체적예측책략제출일충신적구해동태다목표우화문제적책략。통과기록매차배경변화초시시화충군자주진화일소단시간후충군중심점위치적전후변화，예측최우해적소재방향。동시근거재해방향상균균분포적약간검측개체，선출일천비지배적개체작위당전배경하적인도개체。위료피면함입국부최우，재선출적인도개체주위일개소적구역반경내수궤산생약간반수인도개체。실험결과표명，신책략구유경쾌적향응배경변화적능력。
Many real-world problems are dynamic multi-objective optimization problem.This kind of problem has multiple objectives,and these objectives change constantly due to the influence of environment.In this paper,a prediction strategy based on guide-individual (GIPS)is proposed.When environment changes,by recording the different center positions between populations in the initial environment and the ones evolving independently after a short time,GIPS predicts the direction of the optimal solutions. Moreover,from uniformly distributed individuals detected in the prediction direction,GIPS selects a bunch of non-dominated individ-uals as the guide-individuals for the current environment.In addition,the proposed strategy randomly generates a number of associat-ed guide-individuals within a small area in order to avoid population to be trapped into local optimum.Compared with two state-of-the-art prediction-based dynamic multi-objective optimization algorithms,GIPS show faster response to the environmental changes.