
河海大学,江苏 南京 210024
Received:23 June 2025,
Revised:2025-09-01,
Accepted:01 September 2025,
Online First:12 September 2025,
Published:25 March 2026
移动端阅览
陈厚霖,王加虎.基于机器学习的平原城市圩区内涝积水快速模拟方法研究[J].人民珠江,2026,47(3):41-53.
CHEN Houlin,WANG Jiahu.A Rapid Inundation Simulation Method for Urban Waterlogging in Plain Polder Areas Based on Machine Learning[J].PEARL RIVER,2026,47(03):41-53.
陈厚霖,王加虎.基于机器学习的平原城市圩区内涝积水快速模拟方法研究[J].人民珠江,2026,47(3):41-53. DOI: 10.3969/j.issn.1001-9235.2026.03.005.
CHEN Houlin,WANG Jiahu.A Rapid Inundation Simulation Method for Urban Waterlogging in Plain Polder Areas Based on Machine Learning[J].PEARL RIVER,2026,47(03):41-53. DOI: 10.3969/j.issn.1001-9235.2026.03.005.
近年来,江苏省常熟市城区在汛期频繁遭遇强降雨过程,局地积水严重,传统物理模型在应急响应中难以满足时效需求。为提升洪涝预测效率,亟须构建兼顾精度与速度的快速模拟方法。选取常熟主城区典型平原圩区,基于InfoWorks ICM构建一维管网-二维地表耦合模型,模拟12场不同强度降雨过程,提取高精度淹水结果作为样本。结合模拟数据,构建长短期记忆网络(Long Short-Term Memory,LSTM)用于关键节点水深序列预测,卷积神经网络(Convolutional Neural Network,CNN)用于淹水空间分布回归,实现数据驱动与物理建模的融合。模型在测试样本中预测精度较高,LSTM预测序列决定系数超过0.90,CNN预测图峰值误差控制在±5 cm内,空间交并比(Intersection over Union,IoU)达到0.87,计算效率较ICM(Integrated Catchment Modeling)提升百倍以上。深度学习代理模型能有效提升城市内涝模拟响应速度,在区域尺度的快速预警与辅助决策中具备工程应用潜力。
Frequent short-duration and high-intensity rainstorms during the monsoon season have caused recurrent pluvial flooding and severe localized ponding in the central urban district of Changshu City
Jiangsu Province. For emergency response
purely physics-based hydrodynamic models are often too slow to provide actionable forecasts. This study aims to develop a rapid simulation approach that balances accuracy and computational efficiency by integrating physics-based modeling with deep learning
thereby enabling timely warnings
situational awareness
and decision support for urban flood risk management in a typical plain polder city. A 1D pipe-2D overland coupled hydrodynamic model was built in InfoWorks ICM for the representative polderized core of Changshu
explicitly representing stormwater pipes
inlets
pumps
gates
and surface flow pathways over high-resolution terrain. Twelve design rainstorms with different intensities and temporal patterns were simulated
and the resulting high-fidelity inundation depths were extracted as training and validation samples. Two complementary surrogates were designed: a long short-term memory (LSTM) network to forecast depth time series at critical nodes for operations
and a convolutional neural network (CNN) to regress full-domain flood-depth rasters for spatial situational awareness. Preprocessing ensured mesh-raster alignment and standardized rainfall and terrain inputs. Model skill was evaluated with metrics appropriate for temporal and spatial tasks
including coefficient of determination (
R
2
)
Nash-Sutcliffe efficiency (NSE)
mean absolute error (MAE)
root-mean-square error (RMSE)
peak-depth error
and map-based indicators such as intersection-over-union (IoU)
precision
recall
and F1 score. Cross-scenario tests assessed generalization across rainfall intensities and hyetographs. The framework achieved high predictive accuracy while drastically reducing computation time. The LSTM reproduced hydrograph evolution at most monitored locations with
R
2
>
0.90
capturing both rising and receding limbs. The CNN accurately reconstructed spatial inundation patterns governed by micro-topography and road embankments; peak-depth errors were controlled within ±5 cm
and the IoU for inundation extent reached 0.87. Compared with the full InfoWorks ICM solver
the surrogate-based workflow delivered ≥100× speed-up
enabling second- to minute-level inference on standard hardware. Error analysis shows residuals concentrate near sharp elevation transitions and curbed road segments where mesh-raster misalignment is most influential. Including near-dry states and first-flush periods in training reduces false positives and improves depiction of wetting-drying fronts. A demonstration case indicates that the method provides actionable depth maps and node-level forecasts early in the event window
supporting emergency routing
temporary storage scheduling
and pump dispatch. Coupling deep learning surrogates with a calibrated 1D/2D hydrodynamic model substantially enhances the timeliness of urban flood prediction without sacrificing accuracy
meeting operational demands for rapid warning and decision support in polderized plain cities. The workflow is reproducible and transferable to similar low-relief urban regions
provided a baseline hydrodynamic model is available for scenario generation and that training covers multiple storm intensities and early wetting phases. Future work will focus on probabilistic outputs via ensemble surrogates
tighter data assimilation with real-time sensors
and streamlined deployment within municipal emergency platforms.
夏军 , 石卫 . 变化环境下中国水安全问题研究与展望 [J]. 水利学报 , 2016 , 47 ( 3 ): 10 - 16 . doi: 10.13243/j.cnki.slxb.20150937 http://dx.doi.org/10.13243/j.cnki.slxb.20150937
XIA J , SHI W . Perspective on water security issue of changing environment in China [J]. Journal of Hydraulic Engineering , 2016 , 47 ( 3 ): 10 - 16 . (in Chinese)
ZHOU Q Q , LENG G Y , SU J H , et al . Comparison of urbanization and climate change impacts on urban flood volumes: importance of urban planning and drainage adaptation [J]. Science of the Total Environment , 2019 , 658 : 24 - 33 .
HALLEGATTE S , GREEN C , NICHOLLS R J , et al . Future flood losses in major coastal cities [J]. Nature Climate Change , 2013 , 3 : 802 - 806 .
MAHMOUD S H , GAN T Y . Urbanization and climate change implications in flood risk management: developing an efficient decision support system for flood susceptibility mapping [J]. Science of the Total Environment , 2018 , 636 : 152 - 167 .
ARNONE E , PUMO D , FRANCIPANE A , et al . The role of urban growth, climate change, and their interplay in altering runoff extremes [J]. Hydrological Processes , 2018 , 32 ( 12 ): 1755 - 1770 .
HUONG H T L , PATHIRANA A . Urbanization and climate change impacts on future urban flooding in Can Tho City, Vietnam [J]. Hydrology and Earth System Sciences Discussions , 2011 , 8 ( 6 ): 10781 - 10824 .
张启义 . 北京房山区“7·21”特大暴雨灾害的成因及启示 [J]. 中国防汛抗旱 , 2023 , 33 ( 10 ): 43 - 47 .
ZHANG Q Y . The cause and enlightenment of the “23·7” extremely rainstorm disaster in Fangshan District of Beijing City [J]. China Flood & Drought Management , 2023 , 33 ( 10 ): 43 - 47 . (in Chinese)
刘家宏 , 裴羽佳 , 梅超 , 等 . 郑州“7·20”特大暴雨内涝成因及灾害防控 [J]. 郑州大学学报(工学版) , 2023 , 44 ( 2 ): 38 - 45 . doi: 10.13705/j.issn.1671-6833.2023.02.019 http://dx.doi.org/10.13705/j.issn.1671-6833.2023.02.019
LIU J H , PEI Y J , MEI C , et al . Waterlogging Cause and Disaster Prevention and Control of “7·20” Torrential Rain in Zhengzhou [J]. Journal of Zhengzhou University (Engineering Science) , 2023 , 44 ( 2 ): 38 - 45 . (in Chinese)
高俊峰 , 韩昌来 . 太湖地区的圩及其对洪涝的影响 [J]. 湖泊科学 , 1999 ( 2 ): 105 - 109 . doi: 10.18307/1999.0203 http://dx.doi.org/10.18307/1999.0203
GAO J F , HAN C L . Polders in Taihu Lake region and its effects on floods [J]. Journal of Lake Sciences , 1999 ( 2 ): 105 - 109 . (in Chinese)
单玉书 , 蔡文婷 , 薛宣 , 等 . 环太湖城市群防洪大包围建设影响及对策 [J]. 中国防汛抗旱 , 2018 , 28 ( 2 ): 56 - 59,65 . doi: 10.16867/j.cnki.cfdm.20180115.001 http://dx.doi.org/10.16867/j.cnki.cfdm.20180115.001
SHAN Y S , CAI W T , XUE X , et al . The impacts and countermeasures of developing urban flood control systems around Taihu Lake [J]. China Flood & Drought Management , 2018 , 28 ( 2 ): 56 - 59, 65 . (in Chinese)
夏军 , 王惠筠 , 甘瑶瑶 , 等 . 中国暴雨洪涝预报方法的研究进展 [J]. 水利学报 , 2022 , 53 ( 3 ): 284 - 295 .
XIA J , WANG H Y , GAN Y Y , et al . Research progress in forecasting methods of rainstorm and flood disaster in China [J]. Journal of Hydraulic Engineering , 2022 , 53 ( 3 ): 284 - 295 . (in Chinese)
WEI H B , WU H , ZHANG L Y , et al . Urban flooding simulation and flood risk assessment based on the InfoWorks ICM model: A case study of the urban inland rivers in Zhengzhou, China [J]. Water Science and Technology , 2024 , 90 ( 4 ): 1338 - 1358 .
YIN W H , LIU J , FENG H T , et al . Study of flooding hazards in highly urbanized area based on a coupled one- and two-dimensional hydrological-hydrodynamic model [J]. Desalination and Water Treatment , 2025 , 321 . DOI: 10.1016/j.dwt.2024.100933 http://dx.doi.org/10.1016/j.dwt.2024.100933 .
GUO Z F , LEITÃO J P , SIMÕES N E , et al . Data-driven flood emulation: Speeding up urban flood predictions by deep convolutional neural networks [J]. Journal of Flood Risk Management , 2020 , 14 ( 1 ). DOI: 10.1111/jfr3.12684 http://dx.doi.org/10.1111/jfr3.12684 .
DANG T Q , LE T H , TRINH A D , et al . Application of machine learning-based surrogate models for urban flood depth modeling in Ho Chi Minh City, Vietnam [J]. Applied Soft Computing , 2024 , 150 . DOI: 10.1016/j.asoc.2023.111031 http://dx.doi.org/10.1016/j.asoc.2023.111031 .
BERKHAHN S , NEUWEILER I . Data driven real-time prediction of urban floods with spatial and temporal distribution [J]. Journal of Hydrology , 2023 , 22 . DOI: 10.1016/j.hydroa.2023.100167 http://dx.doi.org/10.1016/j.hydroa.2023.100167 .
刘媛媛 , 刘业森 , 郑敬伟 , 等 . BP神经网络和数值模型相结合的城市内涝预测方法研究 [J]. 水利学报 , 2022 , 53 ( 3 ): 276 - 283 . doi: 10.13243/j.cnki.slxb.20210611 http://dx.doi.org/10.13243/j.cnki.slxb.20210611
LIU Y Y , LIU Y S , ZHENG J W , et al . Intelligent rapid prediction method of urban flooding based on BP neural network and numerical simulation model [J]. Journal of Hydraulic Engineering , 2022 , 53 ( 3 ): 276 - 283 . (in Chinese)
梁汝豪 , 兰甜 , 林凯荣 , 等 . 基于SWMM的城市雨洪径流模拟研究:以广州市猎德涌流域为例 [J]. 人民珠江 , 2018 , 39 ( 6 ): 1 - 5,15 . doi: 10.3969/j.issn.1001-9235.2018.06.001 http://dx.doi.org/10.3969/j.issn.1001-9235.2018.06.001
LIANG R H , LAN T , LIN K R , et al . Simulation of urban rainfall-runoff based on SWMM:the case of Liede River Basin in Guangzhou [J]. Pearl River , 2018 , 39 ( 6 ): 1 - 5, 15 . (in Chinese)
陈泽明 , 方序鸿 , 李家叶 , 等 . 机器学习模型在城市内涝模拟预报中的应用综述 [J]. 人民珠江 , 2025 , 46 ( 1 ): 9 - 22 . doi: 10.3969/j.issn.1001-9235.2025.01.002 http://dx.doi.org/10.3969/j.issn.1001-9235.2025.01.002
CHEN Z M , FANG X H , LI J Y , et al . An overview of application of machine learning models in urban flood simulation and forecasting [J]. Pearl River , 2025 , 46 ( 1 ): 9 - 22 . (in Chinese)
WANG F F , WANG Y , LI B , et al . Hierarchical terrain attention and multi-scale rainfall guidance for flood image prediction [C]// 2023 IEEE International Conference on Image Processing (ICIP) , Kuala Lumpur, Malaysia : 2370 - 2374 .
周添红 , 唐佐槐 , 褚俊英 , 等 . 极端暴雨条件下城市内涝模拟研究进展与展望 [J]. 人民长江 , 2025 , 56 ( 5 ): 14 - 22, 30 .
ZHOU T H , TANG Z H , CHU J Y , et al . Progress and prospects of urban waterlogging simulation under extreme rainstorm conditions [J]. Yangtze River , 2025 , 56 ( 5 ): 14 - 22, 30 . (in Chinese)
薛晓萌 , 李红旮 , 黄晓霞 , 等 . 基于高分遥感的城市内涝数值仿真与风险评价研究 [J]. 遥感技术与应用 , 2024 , 39 ( 4 ): 971 - 986 . doi: 10.11873/j.issn.1004-0323.2024.4.0971 http://dx.doi.org/10.11873/j.issn.1004-0323.2024.4.0971
XUE X M , LI H G , HUANG X X , et al . Numerical simulation and risk evaluation of urban flooding based on high-resolution remote sensing [J]. Remote Sensing Technology and Application , 2024 , 39 ( 4 ): 971 - 986 . (in Chinese)
0
Views
1122
下载量
0
CSCD
Publicity Resources
Related Articles
Related Author
Related Institution
京公网安备11010802024621