基于反向学习的微种群教与学优化算法及其应用  

Opposition-based Learning Teaching-learning-based Optimization Algorithm with a Micro Population and Its Application

作　　者：范勤勤 柳缔西子 王筱薇 韩新[3] 王维莉 FAN Qinqin;LIU Dixizi;WANG Xiaowei;HAN Xin;WANG Weili(Logistics Research Center,Shanghai Maritime University,Shanghai 201306,China;School of Electronic Information and Electrical Engineering,Shanghai Jiao Tong University,Shanghai 200237,China;Shanghai Institute of Disaster Prevention and Relief,Tongji University,Shanghai 200092,China)

机构地区：[1]上海海事大学物流研究中心,上海201306 [2]上海交通大学电子信息与电气工程学院,上海200237 [3]同济大学上海防灾救灾研究所,上海200092

出　　处：《郑州大学学报（工学版）》2020年第4期59-67,共9页Journal of Zhengzhou University（Engineering Science）

基　　金：国家重点研发计划资助项目(2016YFC0800200);国家自然科学基金资助项目(61603244);中国博士后科学基金资助项目(2018M642017)。

摘　　要：为提高教与学优化算法的收敛速率且保证其可靠性,提出了一种基于反向学习的微种群教与学优化算法(opposition-based learning teaching-learning-based optimization algorithm with a micro population,OBL-μTLBO)。在所提算法中,利用小的种群规模(微种群)来提高教与学优化算法的收敛速率和降低对计算机内存的要求。同时,OBL-μTLBO算法使用反向学习所得“教师”来指引种群进化,以此提高算法的全局探索能力和避免其陷入局部最优。仿真结果表明,OBL-μTLBO算法不仅具有较好的寻优性能,而且还具有较快的收敛速度。最后,将OBL-μTLBO算法应用于求解非合作博弈纳什均衡问题,取得令人满意的结果。To improve the convergence speed and reliability of Teaching-learning-based optimization(TLBO)algorithm,an opposition-based learning TLBO with a micro population(OBL-μTLBO)was proposed in the current study.In the proposed algorithm,a small population size(i.e.,micro population)was used to speed up the convergence of TLBO and was able to reduce the computer memory requirements.Moreover,a“teacher”achieved by an opposition-based learning in the proposed algorithm was utilized to improve the global exploration capability and avoid trapping into local optimum region.Simulation results indicated that OBL-μTLBO not only had better overall performance,but also had more quick convergence speed when compared with other competitors.Finally,OBL-μTLBO was used to solve two Nash equilibrium problems of non-cooperative game,and satisfactory results were achieved.

关 键 词：教与学优化 微种群 反向学习 非合作博弈 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]