| 陈一梅,徐造林.基于神经网络的河道浅滩演变预测模型[J].水利学报,2002,33(8):0068-0073 |
| 基于神经网络的河道浅滩演变预测模型 |
| Model based on neural network for predicting the evolution of shoal in river |
| |
| DOI： |
| 中文关键词: BP神经网络 浅滩演变 样本 预测模型 |
| 英文关键词: BP neural network shoal evolution sample prediction model |
| 基金项目: |
|
| 摘要点击次数: 1626 |
| 全文下载次数: 50 |
| 中文摘要: |
| 河道浅滩演变是一个复杂的非线性动力学过程, 作者借助神经网络处理非线性问题的优势, 在分析影响河道浅滩演变因素的基础上, 建立了预测河道浅淮演变的BP网络模型, 并对模型中的输入因子和样本的提取进行了探讨. 以闽江竹岐至侯官河段为实例,用“试控法”给出了BP网络模型的建模方案, 用正交设计原理选取相应的训练样本集, 利用该样本集对网络进行学习和训练, 并用训练好的BP网络模型预测浅滩上年内最小水深和年平均淤积厚度. 计算结果表明: 模型预测结果与实际值吻合良好. 这为河道浅滩演变预测研究提供了新方法. |
| 英文摘要: |
| A back propagation (BP) network model for predicting the evolution of shoal in a river is established based on the analysis of the factors affecting the shoal evolution. The approach of extracting the factors and samples is investigated. The modeling scheme based upon BP neural network is obtained from “trial-test” method. Samples for training are chosen according to the principle of orthogonal method and the model is trained by the samples. The model is applied to predict the minimum water depth in a year and the mean annual depth of deposition on shoal. The prediction of an example shows that the agreement between calculation and prototype is good. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |