遗传门限回归模型在河道洪水预报中的应用

提出了基于遗传算法的一套操作简便、通用的ＴＲ建模方案，并成功地将ＴＲ模型用于相应流量法去预测洪水，实例计算的结果说明其可行性和有效性。

Application of Genetic Threshold Regressive Model to Forecast River Flood

Threshold regressive(TR)model is suggested to describe the complex nonlinear relations between upper reach flood peak discharge and lower one of a river,and a simple and general scheme is presented to deduce TR model.Both threshold values and regressive coefficients of TR model can be optimized with accelerating genetic algorithm proposed by the authors,and the difficulty problem of modeling of TR can be resolved,which gives a strong tool for wide use of modeling of TR can be resolved,which gives a strong tool for wide use of TR model.The example shows that the scheme is practical and efficient,and that TR model can effectively control and reduce model errors,and ensure good forecasting stability and accuracy of the model by means of the threshold valves.The method can be applied to forecast nonlinear time series in various engineering practices.