SAMCCTLBO: a multi-class cooperative teaching–learning-based optimization algorithm with simulated annealing

• DOI: 10.1007/s00500-015-1613-9
• Publication Date: 2015-02-13T07:37:49Z
• Authors: Debao Chen; Feng Zou; Jiangtao Wang; Wujie Yuan
• Keywords: Microteaching; Benchmark (surveying); Operator (biology)

A variant of teaching---learning-based optimization algorithm (TLBO) with multi-classes cooperation and simulated annealing operator (SAMCCTLBO) is proposed in paper. To take full advantage of microteaching, the population is divided into several sub-classes, the mean of all learners in teacher phase of original TLBO is replaced by the mean solutions of different sub-classes, the modification might make the mean solutions improved quickly for the effect of microteaching is often better than teaching in big classes. With considering the limitation of learning ability of learner, the learners in different sub-classes only learn new knowledge from others in their sub-classes in learner phase of SAMCCTLBO, and all learners are regrouped randomly after some generations to improve the diversity of the sub-classes. The diversity of the whole class is improved by simulated annealing operator. The effectiveness of the proposed algorithm is tested on several benchmark functions, the results demonstrate that SAMCCTLBO has some good performances when compared with some other EAs.