[1]薛海朋,刘玉,陈书宁,等.基于改进深度学习的水闸扬压力异常分级预警方法[J].江苏水利,2026,(04):6-11.
 XUE Haipeng,LIU Yu,CHEN Shuning,et al.Graded early warning method for abnormal uplift pressure in sluice based on improved deep learning[J].JIANGSU WATER RESOURCES,2026,(04):6-11.
点击复制

基于改进深度学习的水闸扬压力异常分级预警方法()

《江苏水利》[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2026年04期
页码:
6-11
栏目:
水利信息化
出版日期:
2026-04-22

文章信息/Info

Title:
Graded early warning method for abnormal uplift pressure in sluice based on improved deep learning
文章编号:
1007-7839(2026)04-0006-0006
作者:
薛海朋1刘玉2陈书宁1王东3乔佳1
1.江苏省秦淮河水利工程管理处,江苏 南京 210022;2.江苏南水科技有限公司,江苏 南京 210012;3.北京飞舟空间科技有限公司,北京 100076
Author(s):
XUE Haipeng1 LIU Yu2 CHEN Shuning1 WANG Dong3 QIAO Jia1
1.Management Division of Qinhuai River Hydraulic Engineering of Jiangsu Province, Nanjing 210022, China; 2.Jiangsu Nanshui Technology Co., Ltd., Nanjing 210012, China; 3.Beijing Feizhou Space Technology Co., Ltd., Beijing 100076, China
关键词:
水闸扬压力改进深度学习分级预警实测数据
Keywords:
sluice uplift pressure improved deep learning graded Warning measured Data
分类号:
TV698.1
文献标志码:
A
摘要:
水闸所受扬压力是水闸抗滑和抗浮稳定的重要指标,为实现水闸扬压力的分级预警,可将预警问题转化为基于实测扬压力数据与基于正常状态预测值进行对比问题。为有效提高预警准确性首先必须提高预测精确性,针对水闸扬压力测值序列存在强烈的非平稳性特性,提出了一种基于改进深度学习的水闸扬压力异常分级方法。结果表明,所提预测模型预测精度和泛化能力显著优于改进支持向量机、改进神经网络、长短时记忆神经网络以及改进长短时记忆神经网络等模型,其结果可更好地用于水闸扬压力异常分级预警。
Abstract:
The uplift pressure on a sluice gate is a critical indicator for its anti-sliding and anti-floating stability. To achieve graded early warning for sluice uplift pressure, the warning problem can be transformed into a problem of comparing measured uplift pressure data with predicted values under normal conditions. To effectively improve the accuracy of early warning, it is first necessary to improve prediction precision. Aiming at the strong non-stationary characteristics of the measured uplift pressure sequence of sluice gates, an improved deep learning-based abnormal graded early warning method for sluice uplift pressure is proposed. The results show that the proposed prediction model is significantly superior to the improved support vector machine (SVM), improved neural network, long short term memory (LSTM) neural network, and improved LSTM neural network in terms of prediction accuracy and generalization ability. Its results can be better used for the abnormal graded early warning of sluice uplift pressure.

参考文献/References:

[1]方卫华,张慧,徐孟启. 基于边缘计算的闸坝扬压力在线分级预警方法[J]. 人民黄河,2023,45(9):151-156.
[2]孙嘉良,郑春锋. 平原河网地区水闸工程扬压力监测数据分析与思考[J]. 中国水能及电气化,2025(6):55-60.
[3]季洪涛. 低水头挡水闸下游排水孔失效对闸基渗流场影响分析[J]. 水利建设与管理,2016,36(1):23-26.
[4]刘灿,丁健睿,王山东,等. 水闸工程观测数据分析与预警研究[J]. 黑龙江科学,2025,16(6):75-78.
[5]胡璟,王豹,王璐,等. 水闸扬压力混合预测模型构建与解释[J]. 水力发电,2025(3):56-63.
[6]翟发刚,杨鹏. 基于改进SVM法的水库闸渗流及稳定安全分析[J]. 水利技术监督,2023(3):121-123,194.
[7]乔广超,杨明祥,刘琦,等. 基于PSO-SVR-ANN的丹江口水闸秋汛期月尺度径流预报模型[J]. 水利水电技术(中英文),2021,52(4):69-78.
[8]陶驷骥,李彬权,陈云瑶,等. 新安江岩溶模型与LSTM模型在洪水模拟中的应用比较[J]. 水电能源科学,2025,43(10):6-9,55.
[9]张运鑫,郭邦,王恰恰,等. 基于改进灰狼算法优化BP模型的滨州市用水量预测[J]. 中国农村水利水电,2025(6):95-101.
[10]吕国旭,陈波,张孟昕,等. 基于VMD-AOA-BiLSTM的大坝变形预测模型[J]. 水电能源科学,2025,43(7):148-152.
[11]魏海红. 基于机器学习的水闸变形预测研究[J]. 陕西水利,2025(9):162-164.
[12]刘东,汝文超,张亮亮,等. 应用鱼鹰优化算法改进随机森林模型对区域水-能源-粮食关联系统安全评价[J]. 东北农业大学学报,2025,56(4):64-74.
[13]张挺,马睿佳,程泳铭,等. 基于VMD-LSTM模型的流域径流预报技术[J]. 南水北调与水利科技(中英文),2025,23(5):1196-1203.
[14]滕志军,付雨珊,谷梁岑,等. 融合动态权重系数与Levy飞行的黏菌优化算法[J]. 陕西科技大学学报,2024,42(4):191-198.

相似文献/References:

[1]张 羽.结合水利部安全生产考核浅谈上海市 水闸运行管理中存在的不足[J].江苏水利,2016,(07):69.
 ZHANG Yu.Discussion on the shortage existing in the operation and management of the sluice in Shanghai City, combined with the assessment of the safety production by Ministry of Water Resources[J].JIANGSU WATER RESOURCES,2016,(04):69.
[2]胡峥嵘,张胤,吴利华.万寿河除险加固分析及治理方案比选[J].江苏水利,2016,(10):18.
 HU Zhengrong,ZHANG Yin,WU Lihua.Analysis of reinforcement and control scheme selection of WanshouRiver[J].JIANGSU WATER RESOURCES,2016,(04):18.
[3]邱乾勇,蔡慧娟,汪庆发.钢板桩防渗墙在新禹河闸除险加固工程中的应用[J].江苏水利,2016,(12):28.
 QIU Qianyong,CAI Huijuan,WANG Qingfa.The application of steel sheet pile cutoff wall in Xinyuhe sluice reinforcement project[J].JIANGSU WATER RESOURCES,2016,(04):28.
[4]张 羽,沈 强.超声波及浮子式水位计在感潮河道水闸泵站中的应用与探讨[J].江苏水利,2017,(06):6.
 ZHANG Yu,SHEN Qiang.Application and discussion of ultrasonic and float type water level gauge in sluice pump station of tidal river[J].JIANGSU WATER RESOURCES,2017,(04):6.
[5]梁豪杰,李 猛,刘 涛.水闸工程观测典型年数据处理与成果分析[J].江苏水利,2017,(07):52.
 LIANG Haojie,LI Meng,LIU Tao.Typical year data processing and results analysis of sluice engineering observation[J].JIANGSU WATER RESOURCES,2017,(04):52.
[6]冯佳佳,程 博,何仕强,等.侯家坝闸钢闸门结构流激振动试验研究[J].江苏水利,2018,(08):7.
 FENG Jiajia,CHENG Bo,HE Shiqiang,et al.Experimental study on flow-induced vibration of steel gate structure in Houjiaba sluice[J].JIANGSU WATER RESOURCES,2018,(04):7.
[7]周和平,徐 铭,周建方.水闸基于检修闸门系统的事故闸门研究与应用[J].江苏水利,2019,(10):69.
 ZHOU Heping,XU Ming,ZHOU Jianfang.Research and application of emergency gate based on bulkhead gate system[J].JIANGSU WATER RESOURCES,2019,(04):69.
[8]陈 华,缪慧丽,田磊磊.双节顶升式液压启闭机在金湾闸的应用检修及技术改进[J].江苏水利,2020,(09):68.
 CHEN Hua,MIAO Huili,TIAN Leilei.Application and technical improvement of double-section jacking-up hydraulic hoist in overhaul of Jinwan Sluice[J].JIANGSU WATER RESOURCES,2020,(04):68.
[9]李学德,秦景洪,赵津磊,等.基于有限元仿真的闸室结构内力计算分析[J].江苏水利,2020,(10):10.
 LI Xuede,QIN Jinghong,ZHAO Jinlei,et al.Calculation and analysis of internal force of lock chamber structure based on finite element simulation[J].JIANGSU WATER RESOURCES,2020,(04):10.
[10]王亚乒,陈 飞,曹晓东,等.基于四类水闸安全管理探讨——以常州小河水闸为例[J].江苏水利,2021,(05):69.
 WANG Yaping,CHEN Fei,CAO Xiaodong,et al.Discussion on safety management of four-type sluices——Taking Xiaohe Sluice in Changzhou as an example[J].JIANGSU WATER RESOURCES,2021,(04):69.

备注/Memo

备注/Memo:
收稿日期:2026-01-04
基金项目:江苏省水利科技项目(2024041)
作者简介:薛海朋(1985—),男,高级工程师,本科,主要从事大型水利枢纽运行管理工作。E-mail:279252241@qq.com
更新日期/Last Update: 2026-04-01