A hybridization of teaching–learning-based optimization and differential evolution for chaotic time series prediction

摘要: Chaotic time series prediction problems have some very interesting properties and their has received increasing interest in the recent years. Prediction of chaotic based on phase space reconstruction theory been applied many research fields. It is well known that a system nonlinear, multivariable multimodal optimization problem for which global techniques are required order to avoid local optima. In this paper, new hybrid algorithm named teaching---learning-based (TLBO)---differential evolution (DE), integrates TLBO DE, proposed solve prediction. DE incorporated into update previous best positions individuals force jump out stagnation, because its strong searching ability. The speeds up convergence improves algorithm's performance. To demonstrate effectiveness our approaches, ten benchmark functions three typical nonlinear used simulating. Conducted experiments indicate TLBO---DE performs significantly better than, or at least comparable to, other algorithms.