计算机应用研究2018，Vol.35Issue(2)：441-447,7.DOI:10.3969/j.issn.1001-3695.2018.02.026
环链种群结构的多目标教与学优化算法
Ring-chain population structure of multi-objective teaching-learning-based optimization
摘要
Abstract
According to the teaching-learning-based optimization algorithm using greedy evolutionary mechanism,it is easy to cause poor diversity in the population.This paper proposed an efficient improved TLBO based on ring-chain topology population structure and improving self-learning strategy for multi-objective optimization (MOTLBO-RCP).The proposed algorithm used a ring-chain topology divided the populations into multiple neighborhoods represented a small group and there was an overlap between adjacent groups.In the process of teaching and learning evolution,a specified teacher guided each individual group to evolve independently and the adjacent exchanged information with each other.In order to balance the exploration and exploitation,it proposed an improved learning mechanism.The MOTLBO-RCP was compared with other five typical multi-objective optimization algorithms on a benchmark test set including 12 multi-objective optimization functions.Experimental resuits demonstrate that the proposed algorithm is superior or competitive to the other peer algorithms in convergence,diversity and stability.关键词
教与学优化算法/多目标优化问题/环链种群结构/学习机制Key words
teaching-learning-based optimization (TLBO)/multi-objective optimization problem (MOP)/ring-chain population structure/self-learning mechanism分类
信息技术与安全科学引用本文复制引用
林震,陈辉金,帅剑平..环链种群结构的多目标教与学优化算法[J].计算机应用研究,2018,35(2):441-447,7.基金项目
国家自然科学基金资助项目(61261017) （61261017）
桂林电子科技大学教育教学改革项(JGB201431,JGB201530,ZJW43030) （JGB201431,JGB201530,ZJW43030）
