[1]吴峥,毛媛媛,兰林,等.基于BP神经网络算法的平原感潮港闸流量计算模型研究[J].江苏水利,2025,(04):11-17.
　WU Zheng,MAO Yuanyuan,LAN Lin,et al.Research on the flow calculation model of plain tidal sluice based on the BP neural network algorithm[J].JIANGSU WATER RESOURCES,2025,(04):11-17.
点击复制

基于BP神经网络算法的平原感潮港闸流量计算模型研究()
分享到：

《江苏水利》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2025年04期
页码:
11-17
栏目:
水利规划与设计
出版日期:
2025-04-15

文章信息/Info

Title:
Research on the flow calculation model of plain tidal sluice based on the BP neural network algorithm
文章编号:
1007-7839（2025）04-0011-0007
作者:
吴峥1; 毛媛媛1; 兰林1; 曾贤敏2; 卢知是1
（1. 江苏省水利工程规划办公室，江苏 南京 210029；2. 河海大学 水文水资源学院，江苏 南京 210098）
Author(s):
WU Zheng1;  MAO Yuanyuan1;  LAN Lin1;  ZENG Xianmin2;  LU Zhisi1
（1. Jiangsu Water Conservancy Project Planning Office, Nanjing 210029, China；2. College of Hydrology and Water Resources, Hohai University, Nanjing 210098, China）
关键词:
感潮港闸; 水位流量关系; BP人工神经网络; 输入条件; 闸下潮位
Keywords:
tidal sluice;  stage-discharge relation;  BP artificial neural network;  input conditions;  tidal level below the sluice
分类号:
TV66
文献标志码:
A
摘要:
沿海港闸排水过程受到径流、潮流、闸门开启情况等多重因素共同影响，水位流量呈现复杂的非线性关系，传统堰流公式无法准确描述港闸出流实际情况。采用BP神经网络算法开展了平原地区感潮水闸水位流量关系研究。结合闸下潮位特征分析，提出采用计算时刻及其之前12个整点时刻的闸上、闸下水位作为神经网络模型改进输入条件，结果表明：相较仅采用计算时刻闸上、闸下水位作为输入条件而言，改进输入方式后，里下河4个港计算流量与实测值的均方根误差RMSE减小了52%~85%，决定系数R2增加了0.50~0.55，表明模型计算精度明显提高；对比传统堰流公式计算所得港闸流量结果，基于改进BP神经网络算法的计算精度同样显著提升，均方根误差RMSE减小44%~69%。研究成果为进一步研究沿海港闸出流、精确模拟计算港闸流量和排水量提供了新的方法和思路。
Abstract:
The drainage process of coastal port gates is influenced by multiple factors such as runoff, tidal current, and gate opening, resulting in a complex nonlinear relationship between water level and flow rate. Traditional weir flow formulas cannot accurately describe the actual outflow situation of port gates. The BP neural network algorithm was used to study the relationship between water level and flow rate of tidal gates in plain areas. Combined with the analysis of the tidal level characteristics below the sluice, it is proposed to use the water levels above and below the sluice at the calculation time and the previous twelve whole-hour moments as the improved input conditions for the neural network model. The results show that compared with only using the water levels above and below the gate at the calculation time as the input conditions, after improving the input method, the root mean square error （RMSE） between the calculated flow and the measured values of the four ports in Lixia River decreased by 52%~85%, and the coefficient of determination （R2） increased by 0.50~0.55, indicating a significant improvement in the calculation accuracy of the model; Compared with the traditional weir flow formula, the calculation accuracy of the port gate flow rate based on the improved BP neural network algorithm is also significantly improved, with a reduction of 44%~69% in root mean square error （RMSE）. The research results provide new methods and ideas for further studying the outflow of coastal port gates, accurately simulating and calculating the flow and discharge of port gates.

参考文献/References:

[1]顾正华，李荣. 感潮水闸流量神经网络计算模型研究[J]. 海洋工程，2007，25（3）：109-114.
[2]付晓花，董增川，韩锐，等. 复杂河网地区气候-水文-水动力耦合模型模拟[J]. 水资源保护，2023，39（3）：162-169.
[3]周全，杨笛音，孟钲秀. 平原感潮河网地区边界水闸流量计算方法[J]. 水电能源科学，2014，32（5）：19-22.
[4]王新强. 基于MIKE21水动力模型计算感潮河段闸下水位流量关系[J]. 广东水利水电，2021，308（10）：42-47.
[5]王炎，王船海，王妮，等. BP神经网络在感潮河段水动力洪水预报中的应用[J]. 水力发电，2016，42（2）：21-25.
[6]李宛东. 节制闸过闸流量规律的研究与应用[D]. 郑州：华北水利水电大学，2018.
[7]乔光全，麦宇雄，徐润刚. 利用高低潮推算乘潮水位的方法[J]. 水运工程，2022（1）：35-40.
[8]乔光全，徐润刚，卢永昌，等. 高低潮位扩展至逐时潮位 的插值方法比较[J]. 水运工程，2019（9）：6-12.
[9]刘田甲. 潮汐数据插值方法研究[D]. 南京：南京师范大学，2017.
[10]刘平，刘美华，吴海军. 南通沿海潮位资料插值方法研究[J]. 水资源开发与管理，2016（1）：68-72.

相似文献/References:

[1]徐立燕,周云,李巍,等.连云港市临洪水文站水位流量关系浅析[J].江苏水利,2016,(10):42.
　XU Liyan,ZHOU Yun,LI Wei,et al.Brief analysis of stage discharge relationship about Linhong hydrological station in Lianyungang[J].JIANGSU WATER RESOURCES,2016,(04):42.
[2]夏明嫣.长江下游潮汐河口段水位流量关系分析[J].江苏水利,2018,(05):27.
　XIA Mingyan.Analysis on the stage-discharge relation in the tidal estuary of the lower reaches of the Yangtze River[J].JIANGSU WATER RESOURCES,2018,(04):27.
[3]胡尊乐,纪小敏,闫 浩,等.基于CftooL 拟合工具箱的中田舍河水位流量关系暨产汇流模型的构建[J].江苏水利,2018,(06):1.
　HU Zunle,JI Xiaomin,YAN Hao,et al.Study on the construction of stage-discharge relation and runoff generation and concentration model of Zhongtianshe River based on CftooL fitting toolbox[J].JIANGSU WATER RESOURCES,2018,(04):1.
[4]孙正兰,王 成,李江艳,等.宝应抽水站抗旱调水时三阳河输水流量率定分析[J].江苏水利,2020,(06):40.
　SUN Zhenglan,WANG Cheng,LI Jiangyan,et al.Analysis on the flow rate of the Sanyang River during pumping for drought resistance in the Baoying Station[J].JIANGSU WATER RESOURCES,2020,(04):40.

备注/Memo

备注/Memo:
收稿日期：2025-02-10
基金项目：江苏省水利科技项目（2022003）
作者简介：吴峥（1989—），男，工程师，博士，主要从事水力学及河流动力学、水利规划等相关领域研究工作。E-mail：whowuzheng@163.com
更新日期/Last Update: 2025-04-15