基于EEMD的黄河上游主要来水区年来水量预测

Prediction of Annual Water in the Main Inflow Zone in the Upper Yellow River Based on EEMD

为了提高径流预测的精度,采用EEMD将非线性、非平稳的径流时间序列分解为若干固有模态分量和趋势项分量,高频分量采用GA-SVM模型进行预测,低频分量采用GA-BP模型进行预测,趋势项采用RBF模型进行预测,然后对各分量进行重构,从而建立了EEMD组合预测模型,并应用于黄河上游主要来水区年来水量预测。结果表明:黄河上游主要来水区年来水量预测误差小于20%的预报合格率为100%,预测精度高,具有较高的实用价值。

Runoff forecasting is of great significance to the rational dispatch and optimal allocation of water resources.In order to improve the accuracy of runoff prediction, ensemble empirical mode decomposition （ EEMD） was used, which decomposed non-linear and non-stationar^^ runoff time series into several intrinsic mode component and trend component,because of the different forecasting results of various single prediction models and in view of the high frequency component of the GA-SVM model to predict,low frequency component of the GA-BP model to forecast, the trend term of the RBF model to forecast, then reconstruction of each component, so as to establish a prediction model based on the combination of EEMD,and applied to forecast annual water in the main inflow zone in the upper Yellow River.The results show that the water in the main inflow zone in the upper reaches of the Yellow River forecast prediction of 100% pass rate,which has a high pre-diction accuracy,with high practical value.