鄱阳湖水位动态预测模型

鄱阳湖集江西省内五河(赣江、抚河、信江、饶河和修水)之来水，全省各地降水的多寡对鄱阳湖水位的影响很大。1998年汛期全省出现了两次连续大暴雨过程。第二次大暴雨过程致使鄱阳湖区洪水泛滥，洪涝成灾，造成了国家和人们的重大损失。为了提前做好预防，我们根据历史数值，采用逐步回归^1,2,3]的方法建立鄱阳湖区域及上游降水与鄱阳湖水位关系模型。通过对相应假设进行F检验以及使用模型进行鄱阳湖短期水位的预测检验表明，模型选择的因子较为合理，模型的预测精度较高，效果较好。因此我们认为模型是合理可行的。

Dynamic Prediction Model of Poyang Lake's Water Level

The water of five rivers in Jiangxi flows into Poyang lake,and these five rivers are GanJang?FuHe?XinJiang?RaoHe and XiouHe.The precipitation of every place in Jiangxi province has great impact on Poyang lake's water level.Whole province occurred continuous heavy rain two times at the flood season in 1998.The second heavy rain caused flooding and became disaster,which brought the nation and people to great damage.For the purpose of prevention,we adopt stepwise regression to make relational Model with Poyang lake's Water Level according to the precipitation of the district in Poyang lake and upper stream.We carry out an F_test of the related hypotheses and use this model to forecast the water level of Poyang lake.The result indicated that the factors we had selected were rational,the precision of forecast was high,and the effect was good.So we think that the model is rational and feasible.