Computer Engineering and Applications ›› 2016, Vol. 52 ›› Issue (5): 36-40.

Previous Articles     Next Articles

Teaching-learning-based optimization algorithm by using differential mutation

LI Huirong, QIAO Ximin, ZHAO Pengjun   

  1. School of Mathematics and Computer Application, Institute of Mathematics, Shangluo University, Shangluo, Shaanxi 726000, China
  • Online:2016-03-01 Published:2016-03-17

融合差分变异的教-学优化算法

李会荣,乔希民,赵鹏军   

  1. 商洛学院 数学与计算机应用学院,数学研究所,陕西 商洛 726000

Abstract: Teaching-Learning-Based Optimization (TLBO) algorithm is an new intelligent optimization algorithm which simulates the teaching process and learning process of a classroom. In order to improve the performance of teaching- learning-based optimization algorithm, an improved teaching-learning-based optimization algorithm by using differential mutation(DMTLBO) is proposed. The adaptive teaching factor is proposed, the iterative equation of learner phase is modified by using mutation strategy in the differential evolution algorithm, so the learners increase their knowledge not only by the interaction between themselves, but also through input from the best learner in the current iteration. Simulation results show that DMTLBO outperforms TLBO, PSO and DE in terms of convergence speed and search precision.

null

摘要: 教-学优化算法(TLBO)是模拟班级中的教学过程和学习过程而提出的一种新型智能优化算法。为了改善教-学优化算法的性能,结合差分进化算法,提出一种融合差分变异的教-学优化算法(DMTLBO)。该算法提出自适应的教学因子,根据差分进化算法中变异策略,对学习阶段迭代方程进行改进,使得学习者的学习能力不仅受到学习者之间的相互影响,而且还受到当前最好学习者的影响,提高了算法的收敛速度。仿真实验表明,该算法的收敛速度和寻优精度均优于TLBO算法、PSO算法以及DE算法。

关键词: 智能优化算法, 教学优化, 差分变异, 教学因子