基于相空间重构与模糊神经网络耦合的海温预测模型

Application of phase space reconstruction and ANFIS model in SST forecasting

在海温预报中引入混沌理论,将相空间重构理论与模糊神经网络相结合,建立了海温预测模型。通过相空间重构,把一维海温时间序列拓展为多维序列,而多维序列包含着各态历经的信息,从而挖掘出了丰富的海温变化空间的信息,有利于模糊神经网络的训练。利用建立好的模糊神经网络模型,对海温预报问题进行了建模、训练和预测。实际的预测结果表明,该模型预报精度较高,预测结果可以为业务工作提供一定的参考与借鉴。

The chaos theory is introduced into SST （sea surface temperatare） forecasting, and the phase reconstruction and ANFIS model are combined to establish the forecasting model of SST. Based on the phase reconstruction, the time series of SST is expended to multivariate time series, which includes ergodic information, so that more abundant information can be found in favor of ANFIS training. The model, integrating the phase reconstruction and ANFIS, can work well in the model establishing, training and forecasting. The method evidently improves the prediction precision of SST.