基于EEMD的枯季入库径流预报分析

针对枯季径流量预测的问题，提出了一种新方法。采用集合经验模态分解（EEMD）的数据处理方法，实现对枯季径流的多层次、多时间尺度分解，获得简单且平稳性较好的固有模式分量（IMF），再利用径向基（RBF）人工神经网络对数据进行预测分析。通过以密云、潘家口水库的枯季入库径流量的数据为例,运用此方法进行预测分析，并与以EMD为基础的方法进行对比，结果表明本文提出的模型预测效果理想，精度较高，具有一定的使用价值。

Analysis and Prediction for Dry Season Reservoir Inflow Based on EEMD

For the dry season runoff prediction,this paper proposes a new method.By using a data processing method of ensemble empirical mode decomposition(EEMD),this paper aims at achieving a multi-level,multi-time scale decomposition of dry season runoff time series and obtaining a simple and stable pattern of good inherent component(IMF).Meanwhile,an analysis of the data has also been made by radial basis function(RBF) artificial neural network.Takinging this method in Miyun and Panjiakou Reservoir for examples,an analysis is made of the dry season reservoir inflow by this method and the results are compared with those of the EMD in this paper.According to the results,this paper has a greater degree of accuracy and enhancement in performance.