Teaching–learning-based optimization with differential and repulsion learning for global optimization and nonlinear modeling

摘要: Teaching–learning-based optimization (TLBO) algorithm is one of the recently proposed algorithms. It has been successfully used for solving problems in continuous spaces. To improve performance TLBO algorithm, a modified with differential and repulsion learning (DRLTLBO) presented this paper. In evolution (DE) operators are introduced into teacher phase DRLTLBO to increase diversity new population. learner DRLLBO, local method or adopted according certain probability make learners search knowledge from different directions. method, learn not only best but also another random their neighbors. keep away worst Moreover, self-learning exploitation ability when they changed some generations. decrease blindness history information corresponding generations phase. Furthermore, all regrouped after iterations learners. end, tested on 32 benchmark functions characteristics two typical nonlinear modeling problems, comparison results show that shown interesting outcomes aspects.