复杂网络连通鲁棒性优化  

Optimization of Connectivity Robustness in Complex Networks

作　　者：聂君凤 Nie Jun-feng(Sichuan Vocational College of Health and Rehabilitation,Zigong 643000,Sichuan Province,China)

机构地区：[1]四川卫生康复职业学院,四川自贡643000

出　　处：《科学与信息化》2025年第7期78-80,共3页Technology and Information

摘　　要：近年来,复杂网络的鲁棒性优化问题引起人们广泛关注。一个代理模型不可能适用于评估所有类型复杂网络的连通性鲁棒性。本文将D-S理论应用于代理模型选择及其组合,并在优化过程中把代理模型组合得到的几个评估值求平均作为该复杂网络的连通性鲁棒性。此方法在SF、ER、SW三种合成网络上的实验结果表明,此方法超过了现有的方法,在计算效率上有明显的提高。In recent years,the optimization of robustness in complex networks has attracted widespread attention.A single proxy model cannot be applied to assess the connectivity robustness of all types of complex networks.This paper applies the D-S theory to the selection and combination of proxy models,and in the optimization process,the average of several evaluation values obtained from the combination of proxy models is used as the connectivity robustness of the complex network.Experimental results on three types of synthetic networks—Scale-Free(SF),Erdős-Rényi(ER),and Small-World(SW)—demonstrate that this method outperforms existing methods and shows significant improvement in computational efficiency.

关 键 词：代理模型 进化算法 复杂网络 连通性鲁棒性 D-S理论 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]