摘要
通过对影响挟沙能力因素的分析 ,认为挟沙力的武汉水院公式和窦国仁公式在神经网络原理的基础上是一致的。找出了进行BP神经网络训练时应考虑的水沙要素 ,采用C ++Builder语言编制的程序 ,对 3 0组高、中、低含沙量的水槽试验资料进行训练 ,发现训练值与实测值符合较好 ,说明得到的连接权能反映实际情况 ;用该连接权对 4组试验数据进行了预测 ,预测结果与实测值相差较小 ,表明用BP神经网络方法确定挟沙能力的可行性。
According to the analysis of factors influencing on the sediment-carrying capacity, it is concluded that formulas of Wuhan Univerisity and Dou Guoren's formula are identical from the viewpoint of Artificial Neural Network (ANN). The factors which should be considered in BP training are found and program is compiled by C++ Builder. Good agreements are obtained by training 30 runs of lower concentration, middle concentration and hyproconcentration flow. The conjunctional functions are memoried and used to predict the other 4 runs' sediment-carrying capacity. The differences between the predicted and measured are also slight. It is recommended that using ANN to estimate sediment-carrying capacity be feasible.
出处
《泥沙研究》
CSCD
北大核心
2004年第1期29-34,共6页
Journal of Sediment Research
关键词
BP神经网络
水流
挟沙力
含沙量
水沙要素
artificial neural network
sediment-carrying capacity
training
predict