南京师范大学学报（工程技术版）2025，Vol.25Issue(2)：1-13,13.DOI:10.3969/j.issn.1672-1292.2025.02.001
基于增量Kriging模型辅助的双指标采样昂贵高维优化算法
Incremental Kriging Model-based Expensive High-dimensional Optimization Algorithm with Dual Index Sampling
摘要
Abstract
For expensive many-objective optimization problems,the performance index selection mechanism plays a key role in evaluating the convergence and diversity of candidate solutions.However,these mechanisms face challenges in dealing with expensive problems due to the restricted practical function evaluation.At the same time,relying on a single index may introduce bias,which makes it difficult to balance the convergence and diversity of the population.In order to solve these problems,this paper proposes an expensive many objective optimization algorithm for dual-index sampling assisted by incremental Kriging model.By introducing an incremental Kriging model to approximate the computationally expensive objective function,the proposed algorithm effectively reduces the computational cost and time cost.At the same time,a random ranking selection mechanism based on the best dual-index selection is adopted as an effective model management strategy.This strategy adopts the Iε+(x,y)and ISDE(x,y)indexes to evaluate the quality of candidate solutions simultaneously,which further improves the search efficiency and finally achieves a balance between convergence and diversity.In order to verify the effectiveness of the proposed algorithm,this paper tests it on DTLZ and WFG multi-objective optimization test problems and actual engineering optimization problems,and compares the results with five excellent similar algorithms proposed in recent years.The experimental results show that the proposed algorithm is significantly effective in solving expensive many-objective optimization problems.关键词
昂贵高维多目标优化/代理辅助进化算法/增量Kriging模型/模型管理/性能指标/填充准则Key words
expensive many-objective optimization/agent-aided evolutionary algorithm/incremental Kriging model/model management/performance index/the fill criterion分类
计算机与自动化引用本文复制引用
李二超,唐静..基于增量Kriging模型辅助的双指标采样昂贵高维优化算法[J].南京师范大学学报（工程技术版）,2025,25(2):1-13,13.基金项目
国家自然科学基金项目(62063019)、甘肃省自然科学基金重点项目(24JRRA173)、甘肃省优秀博士生项目(24JRRA205). （62063019）
