南京信息工程大学气象灾害地理信息工程实验室,江苏 南京,210044
单晶,硕士生,主要从事城市内涝数值模拟研究。2870405807@qq.com
薛丰昌,博士,教授。001968@nuist.edu.cn
收稿:2025-01-08,
纸质出版:2025-06-05
移动端阅览
单晶,薛丰昌,程瑶.城市内涝ChebNet-LSTM积水预测模型[J].武汉大学学报(信息科学版),2025,50(6):1225-1234.
SHAN Jing,XUE Fengchang,CHENG Yao.ChebNet-LSTM Waterlogging Prediction Model for Urban Flooding[J].Geomatics and Information Science of Wuhan University,2025,50(6):1225-1234.
单晶,薛丰昌,程瑶.城市内涝ChebNet-LSTM积水预测模型[J].武汉大学学报(信息科学版),2025,50(6):1225-1234. DOI:10.13203/j.whugis20250005
SHAN Jing,XUE Fengchang,CHENG Yao.ChebNet-LSTM Waterlogging Prediction Model for Urban Flooding[J].Geomatics and Information Science of Wuhan University,2025,50(6):1225-1234. DOI:10.13203/j.whugis20250005
随着城市化快速发展,极端暴雨事件频发,城市内涝问题日益严重,精准的积水深度预测对防灾减灾至关重要。现有预测方法中,统计模型难以处理复杂时空特征,而传统机器学习方法又缺乏有效的时空关联建模能力。因此,开发能够有效融合时空特征的预测方法具有重要现实意义。ChebNet利用各监测站间的节点关系矩阵挖掘空间关联,长短期记忆网络(long short-term memory network,LSTM)捕捉数据在时间序列上的变化,二者结合可以充分提取积水深度数据中的时空特征,有效解决传统模型难以兼顾空间与时间信息的局限。采用河南省开封市城区2020—2021年15个积水监测站的小时积水深度观测数据、6个气象站的小时降水数据以及15个监测站之间的距离数据,构建ChebNet-LSTM模型,并用该模型对测试集中的监测站点进行积水深度预测,预测结果显示,平均绝对误差为0.62,平均绝对百分比误差为6.23%,均方根误差为1.06。各项指标表明,ChebNet-LSTM模型在预测积水深度时具有较高的准确性和可靠性。与图注意力网络、LSTM、时空图卷积网络等模型相比,ChebNet-LSTM模型展现出较好的性能,能够有效预测积水变化趋势,为城市排水规划和防涝决策提供重要支持。
Objectives
2
With the increasing frequency of extreme rainfall events and rapid urbanization
urban waterlogging has become a growing concern. Accurate prediction of water accumulation depth is crucial for disaster prevention and mitigation. Traditional statistical models struggle to capture complex spatiotemporal features
while conventional machine learning methods lack effective modeling capabilities for spatiotemporal correlations.
Methods
2
To address these limitations
this paper proposes a ChebNet-LSTM (long short-term memory network) model that effectively integrates spatial and temporal features for improved prediction accuracy. ChebNet leverages the relational matrix between monitoring stations to extract spatial correlations
while LSTM captures temporal patterns in the data. By combining these two approaches
the proposed model fully exploits the spatiotemporal characteristics of water accumulation depth data
overcoming the shortcomings of traditional models that fail to account for both spatial and temporal information.The study utilizes hourly water depth observations from 15 monitoring stations
hourly precipitation data from six meteorological stations
and distance data between the monitoring stations in Kaifeng City
Henan Province
from 2020 to 2021.The ChebNet-LSTM model is constructed using these datasets and subsequently applied to predict water accumulation depths at monitoring stations in the test set.
Results
2
The experimental results demonstrate that the ChebNet-LSTM model achieves outstanding predictive performance, with the water depth predictions showing mean absolute error of 0.62, mean absolute percentage error of 6.23%, and root mean square error of 1.06. These evaluation metrics conclusively prove that the proposed model maintains high accuracy and reliability in practical water accumulation depth prediction tasks.
Conclusions
2
Comparative experiments with models such as graph attention network, LSTM, and spatiotemporal graph convolutional network show that ChebNet-LSTM model outperforms these alternatives, effectively predicting waterlogging trends. The findings provide valuable support for urban drainage planning and flood prevention decision-making, and offer a robust tool for managing urban waterlogging risks.
黄震 , 张学全 , 张庭瑞 , 等 . 暴雨内涝环境下城市交通脆弱路段预测方法 [J/OL]. 武汉理工大学学报(交通科学与工程版) , [ 2024-12-17 ]. http://kns.cnki.net/kcms/detail/42.1824.U.20240102.1045.002.html http://kns.cnki.net/kcms/detail/42.1824.U.20240102.1045.002.html .
HUANG Zhen , ZHANG Xuequan , ZHANG Ting-rui , et al . Prediction Method of Urban Traffic Vulnerable Sections in Rainstorm Waterlogging Environment [J/OL]. Journal of Wuhan University of Technology(Transportation Science & Engineering) , [ 2024-12-17 ]. http://kns.cnki.net/kcms/detail/42.1824.U.20240102.1045.002.html http://kns.cnki.net/kcms/detail/42.1824.U.20240102.1045.002.html .
刘爱丽 . 城市洪涝模拟模型构建及其应用研究: 以北京市凉水河流域为例 [D]. 郑州 : 华北水利水电大学 , 2023 .
LIU Aili . Research on the Construction and Application of Urban Flood Simulation Model: Taking the Liangshui River Basin in Beijing as an Example [D]. Zhengzhou : North China University of Water Resources and Electric Power , 2023 .
KAUR J , PARMAR K S , SINGH S . Autoregressive Models in Environmental Forecasting Time Series: A Theoretical and Application Review [J]. Environmental Science and Pollution Research International , 2023 , 30 ( 8 ): 19617 - 19641 .
YUNUS K , THIRINGER T , CHEN P Y . ARIMA-Based Frequency-Decomposed Modeling of Wind Speed Time Series [J]. IEEE Transactions on Power Systems , 2016 , 31 ( 4 ): 2546 - 2556 .
MASUM M H , ISLAM R , HOSSEN M A , et al . Time Series Prediction of Rainfall and Temperature Trend Using ARIMA Model [J]. Journal of Scientific Research , 2022 , 14 ( 1 ): 215 - 227 .
MUKHERJEE F , SINGH D . Detecting Flood Prone Areas in Harris County: A GIS Based Analysis [J]. GeoJournal , 2020 , 85 ( 3 ): 647 - 663 .
DING Y K , ZHU Y L , FENG J , et al . Interpretable Spatiotemporal Attention LSTM Model for Flood Forecasting [J]. Neurocomputing , 2020 , 403 : 348 - 359 .
张响亮 , 许文华 , 杨明灿 , 等 . 基于机器学习的内涝积水深度预报方法研究 [J]. 海峡科学 , 2022 ( 3 ): 36 - 39 .
ZHANG Xiangliang , XU Wenhua , YANG Mingcan , et al . Research on Prediction Method of Waterlogging Depth Based on Machine Learning [J]. Straits Science , 2022 ( 3 ): 36 - 39 .
潘昊 . 基于卷积神经网络的城市内涝积水分布快速预测模型研究 [D]. 天津 : 天津大学 , 2022 .
PAN Hao . Study on Rapid Predictive Model of Urban Waterlogging Distribution Based on Convolutional Neural Network [D]. Tianjin : Tianjin University , 2022 .
王安琪 , 易民 , 赵含雪 , 等 . 基于LSTM神经网络模型的河道洪水反流向演算 [J]. 水文 , 2025 , DOI: 10.19797/j.cnki.1000-0852.20240148 http://dx.doi.org/10.19797/j.cnki.1000-0852.20240148 .
WANG Anqi , YI Min , ZHAO Hanxue , et al . Flood Inverse-Routing in River Channel Based on LSTM Neural Network [J]. Journal of China Hydrology , 2025 , DOI: 10.19797/j.cnki.1000-0852.20240148 http://dx.doi.org/10.19797/j.cnki.1000-0852.20240148 .
GRAVES A . Generating Sequences with Recurrent Neural Networks [EB/OL]. ( 2023-12-04 ). https://ui.adsabs.harvard.edu/ abs/2013arXiv1308.0850G/abstract https://ui.adsabs.harvard.edu/abs/2013arXiv1308.0850G/abstract .
苏峰山 , 禄小敏 , 叶云会 , 等 . 基于图卷积自编码器的道路网相似度计算方法 [J]. 武汉大学学报(信息科学版) , 2024 , DOI: 10.13203/j.whugis20230303 http://dx.doi.org/10.13203/j.whugis20230303 .
SU Fengshan , LU Xiaomin , YE Yunhui , et al . A Road Networks Similarity Calculation Method Based on Graph Convolutional Autoencoder [J]. Geomatics and Information Science of Wuhan University , 2024 , DOI: 10.13203/j.whugis20230303 http://dx.doi.org/10.13203/j.whugis20230303 .
章超钦 . 基于深度学习的天气雷达回波外推算法研究 [D]. 南京 : 南京信息工程大学 , 2023 .
ZHANG Chaoqin . Research on Extrapolation Algorithm of Weather Radar Echo Based on Deep Learning [D]. Nanjing : Nanjing University of Information Science & Technology , 2023 .
李晗旭 , 李昕 , 黄观文 , 等 . 基于速度分类的LSTM神经网络辅助GNSS/SINS车载定位方法 [J]. 武汉大学学报(信息科学版) , 2024 , DOI: 10.13203/j.whugis20230061 http://dx.doi.org/10.13203/j.whugis20230061 .
LI Hanxu , LI Xin , HUANG Guanwen , et al . LSTM Neural Network Assisted GNSS/SINS Vehicle Positioning Method Based on Velocity Classification [J]. Geomatics and Information Science of Wuhan University , 2024 , DOI: 10.13203/j.whugis20230061 http://dx.doi.org/10.13203/j.whugis20230061 .
SERGIO R P , WACHS-LOPES G , MORELLO S R , et al . A q -Extension of Sigmoid Functions and the Application for Enhancement of Ultrasound Images [J]. Entropy , 2019 , 21 ( 4 ): 430 .
LI Y N . Performance Evaluation of Tanh and AWBM Rainfall-Runoff Models [J]. IOP Conference Series: Earth and Environmental Science , 2021 , 768 ( 1 ): 012048 .
MAAS A L , HANNUN A Y , NG A Y . Rectifier Nonlinearities Improve Neural Network Acoustic Models [C]// The 30th International Conference on Machine Learning , Atlanta, USA , 2013 .
王梅 , 任怡果 , 刘勇 , 等 . 求解多核学习的自适应随机递归梯度下降法 [J]. 计算机技术与发展 , 2025 , DOI: 10.20165/j.cnki.ISSN1673-629X.2025.0038 http://dx.doi.org/10.20165/j.cnki.ISSN1673-629X.2025.0038 .
WANG Mei , REN Yiguo , LIU Yong , et al . Adaptive Stochastic Recursive Gradient Algorithm for Multiple Kernel Learning [J]. Computer Technology and Development , 2025 , DOI: 10.20165/j.cnki.ISSN1673-629X.2025.0038 http://dx.doi.org/10.20165/j.cnki.ISSN1673-629X.2025.0038 .
刘俊奇 , 涂文轩 , 祝恩 . 图卷积神经网络综述 [J]. 计算机工程与科学 , 2023 , 45 ( 8 ): 1472 - 1481 .
LIU Junqi , TU Wenxuan , ZHU En . Survey on Graph Convolutional Neural Network [J]. Computer Engineering & Science , 2023 , 45 ( 8 ): 1472 - 1481 .
YU B , YIN H T , ZHU Z X . Spatiotemporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting [C]// The 27th International Joint Conference on Artificial Intelligence , Stockholm, Sweden , 2018 .
刘燕 , 李恒如 , 谷卫 . 基于多分支ResCovLSTM的城市轨道交通短时客流预测模型 [J]. 现代城市轨道交通 , 2025 ( 2 ): 130 - 139 .
LIU Yan , LI Hengru , GU Wei . Short-Term Passenger Flow Prediction Model for Urban Rail Transit Based on Multi Branch ResCovLSTM [J]. Modern Urban Transit , 2025 ( 2 ): 130 - 139 .
0
浏览量
52
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621