混沌序列优化预测模型研究

Study and simulation application of optimize prediction model of chaotic time series

混沌是一种普遍的非线性动力学行为,大多数情况下是有害的。针对混沌时间序列的难以预测和控制的问题,提出了基于趋势的混沌预测模型,用改进的最优化方法来估计模型的参数,在其相空间中对时序的未来值进行预测。算例表明,选取最佳的模型阶数能增加预测的准确程度,预测效果比统计方法好。

Considering the difficulty in prediction of chaotic time series and the problems in the control, the chaotic prediction model based on the trend is proposed. The model parameters are estimated based on the improved optimized model. The future values of the cycle time in its mutual space are predicted. The calculating example show that the proposed model performs better than the statisticsmethod.