融入教育心理学的SBO算法

SBO algorithm integrated into educational psychology

针对SBO(school based optimization)算法搜索性能差、易陷入局部最优等缺陷,提出融入教育心理学的SBO算法(SBO based on educational psychology,SBO-EP)。在教阶段,引入最近发展区理论,对学生进行分组动态教学,提高算法的探索能力;引用成就动机理论加入自学阶段,针对每组学生的成就动机设计动态自学方式,提高算法的开发能力;在每轮学习过程结束后参考同伴效应设置班级重组操作,增加解的多样性。采用40个CEC2021测试函数和20个其他类型测试函数进行数值实验,并将SBO-EP算法与蚁群优化算法、基于球形矢量的粒子群优化算法、阿基米德优化算法、灰狼优化算法、教与学优化算法、融合认知心理学的教与学优化算法、学生心理学优化算法进行对比分析。结果表明,SBO-EP算法在收敛速度、寻优精度及稳定性上优势明显。最后,对三种策略的组合进行对比实验,验证了改进策略的有效性。

Aiming at the shortcomings of SBO algorithm,such as poor search performance and local optimization,this paper proposed SBO algorithm integrated with educational psychology.The teaching stage used the theory of zone of proximal deve-lopment to carry out dynamic teaching for students in groups to improve the exploration ability of algorithms.It introduced the theory of achievement motivation into the self-study stage,and designed dynamic self-study methods according to the achievement motivation of each group of students to improve the development ability of algorithms.After each round of learning process,refer to the peer effect theory,it set up the class reorganization operation to increase the diversity of solutions.This paper used 40 CEC2021 test functions and 20 other types of test functions for numerical experiments,and compared SBO-EP algorithm with ant colony optimization algorithm,spherical vector-based particle swarm optimization,Archimedes optimization algorithm,gray wolf optimization algorithm,teaching and learning algorithm,cognitive psychology teaching-learning-based optimization and student psychology based optimization algorithm.The results show that SBO-EP algorithm has obvious advantages in convergence speed,optimization accuracy and stability.Finally,it conducts a comparative experiment on the combination of the three strategies,verifies the effectiveness of the improved strategy.