基于BAM神经网络河床断面模式识别的中小桥水害预测

Water Hazard Prediction of Small and Medium Bridges for Pattern Recognition of Riverbed Profile Based on BAM Neural Network

针对现行的中小桥水害预测方法野外作业量大、缺乏资料无法预测等弊端,从推理公式出发,把实际中小桥河床断面分为9种类型,建立了各类断面直接从雨量到水位的水害预测模型;利用BAM神经网络对河床断面进行模式识别,以确定水害预测模型的类型;将该桥雨量—水位数据代入相应的水害预测模型,求解其中的参数,得出该桥的雨量—水位函数关系式。实例验证该方法具有可行性和有效性,大大降低了劳动强度,特别是对于缺乏资料的中小桥水害预测,尤为适用。

The traditional water hazard prediction method of small and medium bridges has many abuses,such as it needs a lot of field work and can＇t be used in the bridges by lack of data. To solve those problems ,from the reasoning formula ,the models of water hazard prediction directly from precipitation rain fall to water line are set up for the nine types of riverbeds of small and medium bridges. To predict the water line with these models, the BAM neural network are used for the pattern recognition of riverbed profile to choose the appropriate model and parameters in the model can be gotten with the data of precipitation rain fall--water line of the bridge. Then the function between precipitation rain fall and water line of the bridge are conformed. An example and its results are provided to demonstrate the feasibility and validity of the proposed method. With this method, the hard field work is largely reduced. Especially ,it is applicable to the small and medium bridges by lack of data.