Rapid simulation of flood routing using deep learning and hydrodynamic model
LIAO Yaoxing1,2, GAO Weizhi1,2, ZHANG Xuan3, LAI Chengguang1,2, WANG Zhaoli1,2
1. School of Civil Engineering & Transportation, South China University of Technology, Guangzhou 510641; 2. Guangdong Artificial Intelligence and Digital Economy Laboratory(Guangzhou), Guangzhou 510330; 3. The National Key Laboratory of Water Disaster Prevention, Nanjing Hydraulic Research Institute, Nanjing 210029
Abstract：The rapid simulation and early warning of flood disasters are crucial for flood prevention and mitigation. However, the current simulation efficiency of urban flood models based on physical mechanisms remains low. In this study, a deep learning model based on convolutional neural network(CNN) is constructed by combining flood inundation data generated by hydrodynamic model and deep learning techniques to rapidly simulate urban flood routing. The results show that the developed CNN model can effectively simulate the flood inundation, with a peak water depth prediction error within 8%, and a good performance in simulating the inundation extent. The CNN model demonstrates a significantly higher efficiency in flood inundation simulation, achieving approximately 400 times faster computation while maintaining comparable accuracy to hydrodynamic models. This study can provide valuable insights for rapid simulation of urban flood inundation, early warning and forecasting of flood disasters, and the development of digital twin basins.
[1] Bruno W,William R,Benjamin A,et al.Impact of warmer climate periods on flood hazard in the European Alps[J].Nature Geoscience,2022,15(2):118-123. [2] 张建云,王银堂,贺瑞敏,等.中国城市洪涝问题及成因分析[J].水科学进展,2016,27(4):485-491. [3] 《中国水旱灾害防御公报》编写组.《中国水旱灾害防御公报2022》概要[J].中国防汛抗旱,2023,33(10):78-82. [4] 徐卫红,刘昌军,吕娟,等.郑州主城区2021年"7·20"特大暴雨洪涝特征及应对策略[J].中国防汛抗旱,2022,32(5):5-10. [5] 侯精明,张兆安,马利平,等.基于GPU加速技术的非结构流域雨洪数值模型[J].水科学进展,2021,32(4):567-576. [6] 王兆礼,陈昱宏,赖成光.基于TELEMAC-2D和SWMM模型的城市内涝数值模拟[J].水资源保护,2022,38(1):117-124. [7] 王昕泽,侯精明,王峰,等.城市雨洪内涝三维实景动态模拟表达方法研究[J].中国防汛抗旱,2023,33(10):52-57. [8] 黄国如,陈志威,曾博威.城市洪涝模型及CPU-GPU异构并行计算技术研究进展[J].水利学报,2023,54(6):654-665. [9] LIAO Y X,WANG Z L,LAI C G,et al.A framework on fast mapping of urban flood based on a multi-objective random forest model[J].International Journal of Disaster Risk Science,2023,14(2):253-268. [10] 杨哲,杨昆,吕娟,等.基于改进的三层机器学习搜索法极端降雨区域频率估计[J].中国防汛抗旱,2023,33(1):44-51. [11] Niels F,WANG Q J,WU W Y,et al.Supercharging hydrodynamic inundation models for instant flood insight[J]. Nature Water,2023,1(10):835-843. [12] XU K,HAN Z T,XU H S,et al.Rapid Prediction Model for Urban Floods Based on a Light Gradient Boosting Machine Approach and Hydrological-Hydraulic Model[J]. International Journal of Disaster Risk Science,2023,14(1):79-97. [13] 周聂,侯精明,陈光照,等.基于机器学习的山洪灾害快速预报方法[J].水资源保护,2022,38(2):32-40,111. [14] 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):284-295. [15] HOU J M,ZHOU N,CHEN G Z,et al.Rapid forecasting of urban flood inundation using multiple machine learning models[J].Natural Hazards,2021,108(2):2335-2356. [16] 徐宗学,叶陈雷.城市暴雨洪涝模拟:原理、模型与展望[J].水利学报,2021,52(4):381-392. [17] ZHOU Y R,WU W Y,Rory N,et al.Deep learning-based rapid flood inundation modeling for flat floodplains with complex flow paths[J].Water Resources Research,2022,58(12):e2022WR033214. [18] KAO I F,LIOU J Y,LEE M H,et al.Fusing stacked autoencoder and long short-term memory for regional multistep-ahead flood inundation forecasts[J].Journal of Hydrology,2021(14):126371. [19] CHU H B,WU W Y,WANG Q J,et al.An ANN-based emulation modelling framework for flood inundation modelling:Application,challenges and future directions[J]. Environmental Modelling&Software,2020,124:104587. [20] Kazem S M,Georges K,Ahmed C A,et al.LISFLOOD-FP 8.1:new GPUaccelerated solvers for faster fluvial/pluvial flood simulations[J].Geoscientific Model Development,2023,16(9):2391-2413. [21] Serkan K,Onur A,Osama A J,et al.1D convolutional neural networks and applications:A survey[J].Mechanical Systems and Signal Processing,2021,151:107398.
2024, Vol. 34 
