基于神经网络和进化算法的混沌序列产生方法

应用具有全局最优的进化规划算法建立产生混沌序列的优化神经网络模型。该模型利用神经网络权值调整的灵活性 ,能够在同一网络结构中产生的多种混沌序列。计算机仿真结果表明 :该模型比 BP算法训练的神经网络模型能更好地重构混沌吸引子 ,调整网络权值即可产生多种混沌序列。

A METHOD OF GENERATING CHAOTIC SEQUENCES BASED ON THE NEURAL NETWORK AND EVOLUTIONARY PROGRAMMING

Based on the strong learning ability and nonlinear function approximation capacity of Multi Layer Perceptrons (MLPs), a generating chaotic sequence model is proposed in this paper. The chaos generation neural network model and synaptic weights database have been built to generate many chaotic sequences trained by the Evolutionary Programming (EP) algorithm with various discrete chaotic time series. Experimental results show that this EP trained MLP model can generate a chaotic series, whose attractor can be reconstructed better than that generated by the BP trained MLP model and which generates many chaotic sequences by changing weights of this MLPs very easily.