文献类型：期刊文献

中文题名：基于网格拥挤度的自适应参考点多目标优化算法

英文题名：An Adaptive Multi-Objective Optimization Algorithm with Reference Point Based on Grid Congestion Degree

作者：王学武[1];高永亮[1];顾幸生[1]

机构：[1]华东理工大学能源化工过程智能制造教育部重点实验室,上海200237

年份：2025

卷号：51

期号：4

起止页码：522

中文期刊名：华东理工大学学报(自然科学版)

外文期刊名：Journal of East China University of Science and Technology

收录：;北大核心:【北大核心2023】；

基金：国家自然科学基金(62076095)。

语种：中文

中文关键词：进化算法;IGD-NS指标;多目标优化;网格拥挤度;进化计算

外文关键词：evolutionary algorithm;IGD-NS indicator;multi-objective optimization;grid congestion degree;evolutionary computation

摘要：在多目标优化中,对于搜索到的种群要兼顾收敛性和分布性。基于指标的参考点自适应多目标优化算法(AR-MOEA)算法强调IGD-NS指标的最优,算法收敛过程加快,容易陷入局部最优,导致种群不能覆盖到完整的Pareto前沿。本文提出了一种基于网格拥挤度的自适应参考点多目标优化算法(AR-MOEA-GC),该算法区分了种群中贡献个体与非贡献个体的适应度计算方法,保证种群的分布性和收敛性;同时,为了加快种群在算法后期的收敛速度,融入了参考点调整策略,辅助种群向真实Pareto进化。将改进的算法与6个先进的多目标进化算法在3类测试函数上测试,结果表明AR-MOEA-GC在三维的多目标优化问题上有着一定的竞争力。
In multi-objective optimization,balancing both convergence and diversity in the searching population is paramount.However,the AR-MOEA algorithm,which prioritizes optimizing the IGD-NS indicator,and accelerates the convergence process,is prone to trapping in local optima,leading to incomplete coverage of the entire Pareto front by the population.To address this issue,this paper introduces the Adaptive Reference Point Multi-Objective Evolutionary Algorithm based on Grid Crowding(AR-MOEA-GC).This algorithm differentiates the fitness calculation methods for contributing and non-contributing individuals within the population,thereby ensuring both diversity and convergence.Furthermore,to expedite the convergence speed of the population during the later stages of the algorithm,a reference point adjustment strategy is integrated to guide the population to evolve towards the true Pareto front.The enhanced algorithm was rigorously tested against six advanced multi-objective evolutionary algorithms using three types of test functions.The results demonstrate that AR-MOEA-GC exhibits competitive performance in solving three-dimensional multi-objective optimization problems.