冰凌预报模糊优选神经网络BP方法

黄河内蒙段地处黄河流域最北端，由于它的特殊地理位置、水文气象条件、河道特性，几乎每年产生凌汛。本文在分析影响凌汛成因因素的基础上，选取合适的预报因子，基于模糊优选神经网络BP模型，提出冰凌预报方法，应用于黄河内蒙段封河、开河日期的预报。结果表明，计算简便，精度良好。

Fuzzy optimization neural network BP approach for ice forecasting

Based on the analysis on the factors affecting the formation of ice-jam flood, the most important factors for forecasting are selected. By using the fuzzy optimization neural network BP approach the forecasting of ice formation is carried out. This method is applied to forecast the freeze-up date and break-up date of the Yellow River. The result shows that the accuracy of forecasting is high and the procedure of calculation is simple.