白山水库径流中长期预报研究与应用

Research and Application on Medium-term and Long-term Runoff Forecast of Baishan Reservoir

基于时间序列和物理成因分析 ,将成因分析、统计方法与人工神经网络相结合 ,挑选出影响白山水库汛期入库流量的前期大气环流影响因子 ,建立了逐步周期分析模型和逐步多元回归与人工神经网络的耦合模型。预报实践表明 ,所建模型合理 ,预报效果好 ,精度高 ,具有较高的推广和应用价值。

With the application of time series and contributing factors analysis method, a model of successive regression period analysis and a coupled model of successive regression and neural network for medium term and long term runoff forecast have been set up by organically combining cause and effect and statistical analysis with neural network. The result of calculation shows that the models are highly effective and are worth popularization and application.