摘要
教与学优化算法(TLBO)是一种基于课堂教学学习过程的新型元启发式算法,在解决众多科学和工程问题方面表现出了卓越的性能。然而,许多研究表明,TLBO在问题的求解过程中搜索能力较差,算法易陷入局部最优。针对这一问题,提出了一种基于莱维飞行和高斯分布的教与学优化算法(LTLBO-GD)。一方面,将原有的TLBO算法与莱维飞行策略相融合,在教学阶段前期增强老师的学习能力,选择出教学能力突出的老师。另一方面,在学习阶段加入高斯分布局部搜索算子,引导学生进行自学习模式,高斯分布主要是对班级中某一维度进行搜索,从而增强算法单维搜索能力和收敛速度。测试实验结果表明,LTLBO-GD算法在求解精度和收敛速度上都有了显著的改善,并且在求解效果上也明显优于TLBO算法,在典型的标准测试函数优化问题中其优越性更加突出。
Teaching-learning-based optimization algorithm(TLBO)is a new meta-heuristic algorithm based on classroom teaching and learning process,which has shown excellent performance in solving many scientific and engineering problems.However,many studies have shown that TLBO has poor searching ability in the process of problem solving,and the algorithm is easy to fall into local optimum.In this paper,a TLBO based on Levy flight and Gaussian distribution(LTLBO-GD)is proposed to solve this problem.On the one hand,the original TLBO algorithm is integrated with Levy's flight strategy to enhance teachers'learning ability in the early stage of teaching and select teachers with outstanding teaching ability.On the other hand,the local search operator of Gaussian distribution is added in the learning stage to guide students to conduct self-learning mode.The Gaussian distribution is mainly used to search a certain dimension in the class,so as to enhance the single-dimensional search ability and convergence speed of the algorithm.The test results show that the LTLBO-GD algorithm has a significant improvement in solving precision and convergence speed,and it is also significantly better than the TLBO algorithm in solving performance,and its superiority is more prominent in the typical standard test function optimization problem.
作者
付帅
张晓霞
杨娇
胡银银
FU Shuai;ZHANG Xiaoxia;YANG Jiao;HU Yinyin(School of Computer Science and Software Engineering,University of Science and Technology Liaoning,Anshan 114051,China)
出处
《辽宁科技大学学报》
CAS
2021年第3期208-215,240,共9页
Journal of University of Science and Technology Liaoning
基金
辽宁省教育厅项目(2019LNJC07)。
关键词
教与学算法
莱维飞行
高斯分布
teaching-learning-based optimization algorithm
Levy flights
Gaussian distribution