基于遗传算法优化RBF网络的预测混沌时间序列

Based on Genetic Algorithm Optimization RBF Neural Network for Predicting Chaotic Time Series

提出用遗传算法优化径向基函数(RBF)神经网络,使其更接近非线性映射和更快的学习收敛速度。然后用改进后的RBF神经网络预测混沌时间序列。实验结果表明,基于RBF网络的混沌时间序列具有很强的拟合能力、误差小、取得更好的效果。

This paper proposes a genetic algorithm with radial basis function neural network（RBF）,making it closer to the nonlinear mapping and faster learning speed.Then the improved RBF neural network for predicting chaotic time series.The experimental results show that,chaotic time series based on RBF network has very strong capability of fitting,small error,to obtain a better effect.