[1]翟军昌,赵 震,张 萍.改进的教-学优化算法[J].计算机技术与发展,2019,29(08):37-41.[doi:10. 3969 / j. issn. 1673-629X. 2019. 08. 007]
 ZHAI Jun-chang,ZHAO Zhen,ZHANG Ping.An Improved Teaching-learning Based Optimization Algorithm[J].,2019,29(08):37-41.[doi:10. 3969 / j. issn. 1673-629X. 2019. 08. 007]
点击复制

改进的教-学优化算法()
分享到:

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
29
期数:
2019年08期
页码:
37-41
栏目:
智能、算法、系统工程
出版日期:
2019-08-10

文章信息/Info

Title:
An Improved Teaching-learning Based Optimization Algorithm
文章编号:
1673-629X(2019)08-0037-05
作者:
翟军昌1 ;?赵 震1 ;?张 萍2
1. 渤海大学 信息科学与技术学院,辽宁 锦州 121013; 2. 鞍山师范学院 物理科学与技术学院,辽宁 鞍山 114005
Author(s):
ZHAI Jun-chang1 ;?ZHAO Zhen1 ;?ZHANG Ping2
1. School of Information Science and Technology,Bohai University,Jinzhou 121013,China; 2. School of Physics and Technology,Anshan Normal University,Anshan 114005,China
关键词:
教学优化算法;?教学阶段;?学习阶段;?扰动
Keywords:
teaching-learning based optimization algorithm;?teaching phase;?learning phase;?perturbation
分类号:
TP301.6
DOI:
10. 3969 / j. issn. 1673-629X. 2019. 08. 007
摘要:
教-学优化算法是一种新型启发式优化算法。 针对教-学优化算法容易陷入局部最优的不足,提出了一种改进教-学优化算法(an improved teaching-learning based optimization, AITLBO)。 在教学阶段通过扰动机制提高教师的教学效果,避免算法陷入局部最优。 在学习阶段初期分别采取较差学生向优秀学生动态随机学习和优秀学生重新向教师随机学习的策略使当前解向最优方向进化,避免较差解破坏较优解的结构,提高了学习阶段学生的学习效率。 在学习阶段后期引入了学生自我反思的学习策略,实现算法对局部信息的精细搜索,提高算法对解空间信息开发的能力,避免了算法因过早收敛易陷入局部最优的不足。 将其与目前较优的几种改进 TLBO 算法和其他启发式优化算法进行性能测试对比,结果表明 AITLBO 算法具有较高的寻优精度和较快的收敛速度。
Abstract:
Teaching-learning based optimization algorithm is a new heuristic optimization algorithm. We propose an improved teaching-learning based optimization (ATLBO) for the problem of premature convergence in teaching-learning based optimization algorithm. A random perturbed scheme is employed for improving the teaching efficiency in the teaching phase,which can avoid the algorithm trapped in local optimal. In the early phase of learning,the strategies of dynamic random learning from poor students to excellent students and random learning from excellent students to teachers are adopted respectively to make the current solution evolve to the optimal direction, avoiding the destruction of the structure of the optimal solution by the poor solution,and improving the learning efficiency of students in the learning phase. In the late phase of learning, students’ self-reflection learning strategy is introduced to realize the fine search of local information of the algorithm, improving the ability of the algorithm to develop spatial information of solution, and avoiding the shortcoming of the algorithm falling into local optimal easily due to premature convergence. It is compared with several improved TLBO algorithms and other heuristic optimization algorithms,which shows that the proposed algorithm has higher convergence precision and faster convergence speed.
更新日期/Last Update: 2019-08-10