IBMSMA: An Indicator-based Multi-swarm Slime Mould Algorithm for Multi-objective Truss Optimization Problems  

作　　者：Shihong Yin Qifang Luo Yongquan Zhou 

机构地区：[1]College of Artificial Intelligence,Guangxi University for Nationalities,Nanning,530006,China [2]Guangxi Key Laboratories of Hybrid Computation and IC Design Analysis,Nanning,530006,China [3]Xiangsihu College of Guangxi University for Nationalities,Nanning,530225,China

出　　处：《Journal of Bionic Engineering》2023年第3期1333-1360,共28页仿生工程学报（英文版）

基　　金：supported by the National Science Foundation of China under Grant No.U21A20464,62066005;Innovation Project of Guangxi University for Nationalities Graduate Education under Grant gxun-chxs2021058.

摘　　要：This work proposes an improved multi-objective slime mould algorithm, called IBMSMA, for solving the multi-objective truss optimization problem. In IBMSMA, the chaotic grouping mechanism and dynamic regrouping strategy are employed to improve population diversity;the shift density estimation is used to assess the superiority of search agents and to provide selection pressure for population evolution;and the Pareto external archive is utilized to maintain the convergence and distribution of the non-dominated solution set. To evaluate the performance of IBMSMA, it is applied to eight multi-objective truss optimization problems. The results obtained by IBMSMA are compared with other 14 well-known optimization algorithms on hypervolume, inverted generational distance and spacing-to-extent indicators. The Wilcoxon statistical test and Friedman ranking are used for statistical analysis. The results of this study reveal that IBMSMA can find the Pareto front with better convergence and diversity in less time than state-of-the-art algorithms, demonstrating its capability in tackling large-scale engineering design problems.

关 键 词：Slime mould algorithm Shift-based density estimation Multi-swarm strategy Multi-objective optimization Truss optimization 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]