基于深度学习的城市积水深度预报研究
作者:
作者单位:

1.南京信息工程大学 气象灾害预报预警与评估协同创新中心/气象灾害教育部重点实验室;2.南京信息工程大学

基金项目:

名称和编号:国家重点研发计划重点专项“重大灾害性天气的短时短期精细化无缝隙预报技术研究”(2017YFC1502000)


A study on urban water level prediction at water-logging stationsbased on deep learning approach
Author:
Affiliation:

1.Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters CIC-FEMD,Nanjing University of Information Science Technology;2.Nanjing University of Information Science &3.Technology

  • 摘要
  • | |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • | |
  • 文章评论
    摘要:

    随着全球气候变化的不断加剧和城市化的快速发展,极端降雨过程导致的城市积涝灾害愈演愈烈,已成为世界各地许多城市面临的严重挑战. 基于2021年5-8月浙江省诸暨市75个国家自动气象观测站的降雨量数据和典型积水点的积水深度数据,使用深度学习模型长短时记忆网络LSTM(Long Short Term Memory,LSTM)构建降雨量与积水深度的关系模型,提供未来间隔15min的2h内城市积涝水位预报,并与随机森林RF(Random Forest,RF)和人工神经网络ANN(Artificial Neural Network,ANN)模型预报结果进行对比. 预报结果表明,LSTM使用前4h的积水与降雨量资料进行未来2h积水预报的结果最优,均方根误差(RMSE)小于5.6cm,相关系数(CC)在 0.93以上,纳什效率系数(NSE)值在0.86以上,预报效果优于RF和ANN,表明所构建的积水预报人工智能模型具有较好的预报效果.

    Abstract:

    With the continuous intensification of global climate change and the rapid development of urbanization, urban waterlogging disasters caused by extreme rainfall processes have become increasingly severe and have become a serious challenge faced by many cities around the world. Based on the rainfall data from 75 national automatic meteorological observation stations in Zhuji City, Zhejiang Province from May to August 2021 and the depth data of typical waterlogging points, a deep learning model Long Short Term Memory (LSTM) network is used to construct a relationship model between rainfall and waterlogging depth, providing a 2-hour urban waterlogging level forecast with an interval of 15 minutes in the future, The prediction results are compared with those of random forest (RF) and Artificial Neural Network (ANN) models. The prediction results show that LSTM has the best performance in predicting water accumulation in the next 2 hours using the water accumulation and rainfall data from the first 4 hours, with a root mean square error (RMSE) of less than 5.6cm, a correlation coefficient (CC) of over 0.93, and a Nash efficiency coefficient (NSE) of over 0.86. The prediction effect is superior to RF and ANN, indicating that the constructed artificial intelligence model for water accumulation prediction has good prediction performance.

    参考文献
    [1] Reference
    [2] [1]Masson V, Lemonsu A, Hidalgo J, et al. Urban climates and climate change[J]. Annual Review of Environment and Resources, 2020, 45: 411-444.
    [3] [2]Xie J, Chen H, Liao Z, et al. An integrated assessment of urban flooding mitigation strategies for robust decision making[J]. Environmental Modelling Software, 2017, 95: 143-155.
    [4] [3]Wang J, Hu C, Ma B, et al. Rapid urbanization impact on the hydrological processes in Zhengzhou, China[J]. Water, 2020, 12(7): 1870.
    [5] [4]Thanvisitthpon N, Shrestha S, Pal I, et al. Assessment of flood adaptive capacity of urban areas in Thailand[J]. Environmental Impact Assessment Review, 2020, 81: 106363.
    [6] [5]Wang H, Wang H, Wu Z, et al. Using multi-factor analysis to predict urban flood depth based on Naive Bayes[J]. Water, 2021, 13(4): 432.
    [7] [6]Gu X, Li N, Zhou Y, et al. Thinking on urban waterlogging disaster defense initiated by “7. 21” extraordinary rainstorm in Beijing[J]. Journal of Natural Disasters, 2013, 2: 1-6.
    [8] [7]赵凡, 赵常军, 苏筠. 北京“7.21”暴雨灾害前后公众的风险认知变化[J]. 自然灾害学报, 2014, 23(4): 38-45.
    [9] ZHAO Fan, ZHAO Changjun, SU Yun. Changes in risk public perception before and after the
    [10] “7.21”rainstorm disaster in Beijing[J]. Journal Of Natural Disasters, 2014, 23(4): 38-45.
    [11] [8]Zhou Y, Wu Z, Jiang M, et al. Real-time prediction and ponding process early warning method at urban flood points based on different deep learning methods[J]. 2022
    [12] [9]Xue F, Huang M, Wang W, et al. Numerical simulation of urban waterlogging based on floodarea model[J]. Advances in Meteorology, 2016, 2016: 1-9.
    [13] [10]Mu D, Luo P, Lyu J, et al. Impact of temporal rainfall patterns on flash floods in Hue City, Vietnam[J]. Journal of Flood Risk Management, 2021, 14(1).
    [14] [11]Wu Z, Zhou Y, Wang H. Real-Time Prediction of the Water Accumulation Process of Urban Stormy Accumulation Points Based on Deep Learning[J]. IEEE Access, 2020, 8: 151938-151951.
    [15] [12]Vojinovic Z, Tutulic D. On the use of 1D and coupled 1D-2D modelling approaches for assessment of flood damage in urban areas[J]. Urban water journal, 2009, 6(3): 183-199.
    [16] [13]张冬冬, 严登华, 王义成, 等. 城市内涝灾害风险评估及综合应对研究进展[J]. 灾害学, 2014, 29(01): 144-149.
    [17] ZHANG Dongdong, YAN Denghua, WANG Yicheng, et al. Research Progress on Risk Assessment and Integrated Strategies for Urban Pluvial Flooding[J]. Journal of Catastrophology, 2014, 29(1) : 144-149.
    [18] [14]Behrouz M S, Zhu Z, Matott L S, et al. A new tool for automatic calibration of the Storm Water Management Model (SWMM)[J]. Journal of Hydrology, 2020, 581: 124436.
    [19] [15]Zheng Z, Zhang J, Gong H. Uncertainty analysis of MIKE SHE hydrologic model parameter[J]. Yellow River, 2015, 37(1): 23-26.
    [20] [16]Rui X, Jiang C, Chen Q, et al. Principle analysis and application of storm water management model on stimulating rainfall-runoff[J]. Advances in Science and Technology of Water Resources, 2015, 35(4): 1-5.
    [21] [17]房国良, 解以扬, 李培彦, 等. 上海城市暴雨积涝预警系统研究[J]. 大气科学研究与应用, 2009 (2): 32-41.
    [22] FANG Guoliang, XIE Yiyang, LI Peiyan, et al. Research on the Early Warning System for Rainstorm Waterlogging in Shanghai City[J]. Atmospheric Science Research And Applications, 2009(2) : 32-41.
    [23] [18]辜晓青, 章毅之, 殷剑敏, 等. 南昌市城市积涝预警系统研究[J]. 江西农业大学学报, 2005, 27(3): 477-480.
    [24] GU Xiaoqing, ZHANG Yizhi, YIN Jianmin, et al. A Study on the EarlyWarning System for Rainstorm Waterlogging in Nanchang City[J]. Acta Agriculturae Universitatis Jiangxiensis, 2005, 27(3): 477-480.
    [25] [19]张萍萍, 王丽, 张宁. 武汉市城市积涝预警系统及其仿真模拟效果[J]. 暴雨灾害, 2010, 29(1):71-75+88.
    [26] ZHANG Pingping,WANG Li,ZHANGNing. Study on the Early Warning System for Rainstorm Waterlogging in Wuhan City and its Simulated Results[J]. Torrential Rain And Disasters, 2010, 29(1):71-75+88.
    [27] [20]Molinari D, De Bruijn K M, Castillo-Rodríguez J T, et al. Validation of flood risk models: Current practice and possible improvements[J]. International journal of disaster risk reduction, 2019, 33: 441-448.
    [28] [21]Jiang J, Liu J, Cheng C, et al. Automatic estimation of urban waterlogging depths from video images based on ubiquitous Reference objects[J]. Remote Sensing, 2019, 11(5): 587.
    [29] [22]Bubeck P, Aerts J C J H, de Moel H, et al. Preface: Flood-risk analysis and integrated management[J]. Natural Hazards and Earth System Sciences, 2016, 16(4): 1005-1010.
    [30] [23]Nearing G S, Kratzert F, Sampson A K, et al. What role does hydrological science play in the age of machine learning?[J]. Water Resources Research, 2021, 57(3).
    [31] [24]Mcgovern A, Lagerquist R, Gagne D J, et al. Making the Black Box More Transparent: Understanding the Physical Implications of Machine Learning[J]. Bulletin of the American Meteorological Society, 2019, 100(11): 2175-2199.
    [32] [25]Rasp S, Thuerey N. Data‐Driven Medium‐Range Weather Prediction With a Resnet Pretrained on Climate Simulations: A New Model for WeatherBench[J]. Journal of Advances in Modeling Earth Systems, 2021, 13(2).
    [33] [26] 智协飞,张珂珺, 田烨, 等. 基于神经网络和地理信息的华东及华南地区降水概率预报[J]. 大气科学学报, 2021, 44(3) : 381-393.
    [34] ZHI Xiefei, ZHANG Kejun, TIAN Ye, et al.,.Probabilistic precipitation forecast in East and South China based on neural network and geographic information[J]. Trans Atmos Sci, 2021, 44(3) : 381-393.
    [35] [27]智协飞, 王田, 季焱. 基于深度学习的中国地面气温的多模式集成预报研究[J]. 大气科学学报, 2020, 43(3) : 435-446.
    [36] ZHI XieFei, WANG Tian, JI Yan. Multimodel ensemble forecasts of surface air temperature over China based on deep learning approach[J]. Trans Atmos Sci, 2020, 43(3) : 435-446.
    [37] [28]Mosavi A, Ozturk P, Chau K. Flood prediction using machine learning models: Literature review[J]. Water, 2018, 10(11): 1536.
    [38] [29]Ding Y, Zhu Y, Feng J, et al. Interpretable spatio-temporal attention LSTM model for flood forecasting[J]. Neurocomputing, 2020, 403: 348-359.
    [39] [30]Granata F, Gargano R, De Marinis G. Support vector regression for rainfall-runoff modeling in urban drainage: A comparison with the EPA’s storm water management model[J]. Water, 2016, 8(3): 69.
    [40] [31]倪汉杰, 蒋仲廉, 初秀民, 等. 基于DWT-LSTM的航道水位智能预测模型研究[J]. 中国航海, 2021, 44(2): 97-102.
    [41] NI Hanjie, JIANG Zhonglian, CHU Xiumina, et al. DWT-LSTM based Intelligent Water Level PredictionModel for Navigable Waters[J]. Navigation Of China, 2021, 44(2): 97-102.
    [42] [32]周勇强, 朱跃龙. 基于SFLA-CNN和LSTM组合模型的水位预测[J]. 计算机与现代化, 2021(4): 1-7.
    [43] ZHOU Yongqiang,ZHU Yuelong. Water Level Prediction Based on SFLA-CNN and LSTM Combined Model[J]. Jisuanji Yu Xiadaihua, 2021(4): 1-7.
    [44] [33]Chiang Y M, Chang L C, Tsai M J, et al. Dynamic neural networks for real-time water level predictions of sewerage systems-covering gauged and ungauged sites[J]. Hydrology and Earth System Sciences, 2010, 14(7): 1309-1319.
    [45] [34]Lees T, Reece S, Kratzert F, et al. Hydrological Concept Formation inside Long Short-Term Memory (LSTM) networks[J]. Hydrology and Earth System Sciences Discussions, 2021: 1-37.
    [46] [35]王亦斌, 孙涛, 梁雪春,等. 基于EMD-LSTM模型的河流水量水位预测[J]. 水利水电科技进展, 2020, 40(6): 40-47.
    [47] WANG Yibin, SUN Tao, LIANG Xuechun, et al. Prediction of river water flow and water level based on EMD-LSTM model[J]. Advances in Science and Technology of Water Resources, 2020, 40(6): 40-47.
    [48] [36]刘亚新, 樊启祥, 尚毅梓, 等. 基于 LSTM 神经网络的水电站短期水位预测方法[J]. 水利水电科技进展, 2019, 39(2): 56-60.
    [49] LIU YaXin, FAN QiXiang, SHANG YiZi, et al. Short-term water level prediction method for hydropower station based on LSTM neural network[J]. Advances in Science and Technology of Water Resources, 2019, 39(2): 56-60.
    [50] [37]Liu J, Shao W W, Xiang C, et al. Uncertainties of urban flood modeling: Influence of parameters for different underlying surfaces[J]. Environmental research, 2020, 182: 108929.
    [51] [38]Chen Y, Liu R, Barrett D, et al. A spatial assessment framework for evaluating flood risk under extreme climates[J]. Science of the Total Environment, 2015, 538: 512-523.
    [52] [39]Berkhahn S, Fuchs L, Neuweiler I. An ensemble neural network model for real-time prediction of urban floods[J]. Journal of hydrology, 2019, 575: 743-754.
    [53] [40]Wang H, Zhao Y, Zhou Y, et al. Prediction of urban water accumulation points and water accumulation process based on machine learning[J]. Earth Science Informatics, 2021, 14(4): 2317-2328.
    [54] [41]王云. 诸暨市重大气象灾害应急响应保障机制的探索与实践[J]. 安徽农学通报, 2018, 24(06): 141-142+145.
    [55] WANG Yun. Exploration and practice of emergency response guarantee mechanism for major meteorological disasters in Zhuji City[J]. Anhui Agri Sci Bull, 2018, 24 (06): 141-142+145.
    相似文献
    引证文献
引用本文

智协飞,崔碧瑶,季焱.基于深度学习的城市积水深度预报研究[J].南京信息工程大学学报,,():

复制
分享
文章指标
  • 点击次数:644
  • 下载次数: 0
  • HTML阅读次数: 0
  • 引用次数: 0
历史
  • 收稿日期:2023-02-11
  • 最后修改日期:2023-05-08
  • 录用日期:2023-05-09

地址:江苏省南京市宁六路219号    邮编:210044

联系电话:025-58731025    E-mail:nxdxb@nuist.edu.cn

南京信息工程大学学报 ® 2025 版权所有  技术支持:北京勤云科技发展有限公司