融合简化粒子群的教与学优化算法

教与学优化算法(teaching-learning-based optimization algorithm,TLBO)是一种基于班级"教师阶段"和"学生阶段"的新型群智能优化算法.针对算法求解高维非线性复杂优化问题时精度较低的缺点,提出一种混合的教与学优化算法(HTLBO).首先,对"教师阶段"中的学生平均水平重新定义,并采用一种自适应策略根据粒子的适应度值对学习因子动态取值;然后,在迭代的过程中,根据适应度值将种群分成两个子种群,对于适应度值好的子种群采用改进的教与学优化算法(ATLBO)更新以增加种群的多样性,对于适应度值差的子种群采用简化粒子群算法(SPSO)以提升子种群的收敛性;最后,通过10个无约束优化问题进行对比测试实验,结果显示,HTLBO在探索性能和收敛速度方面优于TLBO等其他4种类型的算法.

A hybrid Method Based on Teaching-learning-based and Simplified Particle Swarm Optimization

Teaching-learning-based optimization(TLBO)is a novel swarm intelligence optimizationalgorithm.Since the low accuracy of TLBO for solving high dimensional nonlinear complex optimization problems,A hybrid TLBO algorithm(HTLBO)is proposed in this paper.Firstly,the average level of students is redefined in the"teacher stage",and the learning factor is adopted according to the fitness value of the particle with an adaptive strategy.Then,the population is divided into two sub populations according to the fitness value in the process of iteration.To the sub populations with good fitness,the improved TLBO(ATLBO)is used to increase the population diversity.The SPSO is used to improve the convergence of sub populations for the sub populations with poor fitness.Finally,through 10 unconstrained optimization problems are compared and tested,the results show that HTLBO is significantly better than the other four types of TLBO algorithm.