基于讨论组和自主学习的教与学优化算法*

教与学优化算法(teaching-learning-based optimization, TLBO) 是一种模仿教学过程的新型启发式优化算法。针对TLBO 算法寻优精度低、稳定性差的特点, 提出了基于讨论组和自主学习的教学优化算法DSTLBO(discussion group and self-learning TLBO)。在原TLBO算法的“教”阶段当中加入了小组讨论,随机将全体同学分成若干组,通过组内学生向本组中学习最好的组长学习,提高了算法的局部开发和寻优能力；组长受老师和组内同学影响进行变异,提高了算法的探索能力；在“教”、“学”阶段后,每个学生进入“自我学习”阶段,从而提高了算法的全局搜索能力。通过对8个复杂的Benchmark函数的测试表明：DSTLBO 算法与基本TLBO算法和其经典改进算法ETLBO算法相比,在寻优精度、稳定性和收敛速度方面更具优势。

Teaching-learning-based optimization algorithm based on discussion group and self-learning

Teaching-learning-based optimization (TLBO) is a new heuristic optimization algorithm that imitates the teaching process. Aiming at the low precision and poor stability of TLBO algorithm, an improved teaching-learning-based optimization algorithm named DSTLBO (Discussion Group and Self-learning TLBO) based on discussion group and autonomous learning is proposed. In the process of teaching, the group discussion mechanism was added into the TLBO, and all the students were randomly divided into several groups. The students in the group learned from the group monitor, and then which improved the local search ability of the algorithm. And the ability to explore the algorithm was enhanced by the mutation of the group monitors . After the "teaching" and "learning" phases, all students got into the "self-learning" which improved the global optimization ability of the algorithm. The proposed algorithm was tested on 8 complex benchmark functions and the performance of the algorithm was compared with that of TLBO and ETLBO.The resualt shows that DSTLBO algorithm has advantages over TLBO and ETLBO in optimizing precision, stability and convergence speed.