Deep learning models demonstrate impressive performance in rapidly predicting urban floods,but there are still limitations in enhancing physical connectivity and interpretability.This study proposed an innovative mode...Deep learning models demonstrate impressive performance in rapidly predicting urban floods,but there are still limitations in enhancing physical connectivity and interpretability.This study proposed an innovative modeling approach that integrates convolutional neural networks with weighted cellular automaton(CNN-WCA)to achieve the precise and rapid prediction of urban pluvial flooding processes and enhance the physical connectivity and reliability of modeling results.The study began by generating a rainfall-inundation dataset using WCA and LISFLOOD-FP,and the CNN-WCA model was trained using outputs from LISFLOOD-FP and WCA.Subsequently,the pre-trained model was applied to simulate the flood caused by the 20 July 2021 rainstorm in Zhengzhou City.The predicted inundation spatial distribution and depth by CNN-WCA closely aligned with those of LISFLOOD-FP,with the mean absolute error concentrated within 5 mm,and the prediction time of CNN-WCA was only 0.8%that of LISFLOOD-FP.The CNN-WCA model displays a strong capacity for accurately predicting changes in inundation depths within the study area and at susceptible points for urban flooding,with the Nash-Sutcliffe e fficiency values of most flood-prone points exceeding 0.97.Furthermore,the physical connectivity of the inundation distribution predicted by CNN-WCA is better than that of the distribution obtained with a CNN.The CNN-WCA model with additional physical constraints exhibits a reduction of around 34%in instances of physical discontinuity compared to CNN.Our results prove that the CNN model with multiple physical constraints has signifi cant potential to rapidly and accurately simulate urban flooding processes and improve the reliability of prediction.展开更多
Levee or dam failure can cause a significant disaster in most cases. A good prediction of the flood process especially in a real complex terrain is necessary for working out emergency plans for levee or dam breaches. ...Levee or dam failure can cause a significant disaster in most cases. A good prediction of the flood process especially in a real complex terrain is necessary for working out emergency plans for levee or dam breaches. Numerical simulations of levee or dam breach flow were carried out often with constant flow parameters and in relatively simple channels rather than in natural rivers with complex boundaries. This article presents our dedicated studies on the 2-D numerical model of levee or dam breach hydraulics with finite difference schemes. The good performance of the model is demonstrated by comparisons with the theoretical solution of an idealized dam-break flow over a frictionless flat rectangular channel. The model is also validated through its stability and conservation properties. The model is applied to simulate the flood propagation under complex boundary conditions, and the unsteady flood process in a river and in the dry floodplain with a complex bed terrain simultaneously. Furthermore, with respect to engineering practice, the numerical solutions can give special guidance to the effects of parameters such as the flood depth at different sites and the inundated area at different time periods after the levee breach and the travel time of the flood waves, which may be very important for practicing engineers in an efficient flood management.展开更多
基金supported by the General Program of National Natural Science Foundation of China(Grant No.42377467)。
文摘Deep learning models demonstrate impressive performance in rapidly predicting urban floods,but there are still limitations in enhancing physical connectivity and interpretability.This study proposed an innovative modeling approach that integrates convolutional neural networks with weighted cellular automaton(CNN-WCA)to achieve the precise and rapid prediction of urban pluvial flooding processes and enhance the physical connectivity and reliability of modeling results.The study began by generating a rainfall-inundation dataset using WCA and LISFLOOD-FP,and the CNN-WCA model was trained using outputs from LISFLOOD-FP and WCA.Subsequently,the pre-trained model was applied to simulate the flood caused by the 20 July 2021 rainstorm in Zhengzhou City.The predicted inundation spatial distribution and depth by CNN-WCA closely aligned with those of LISFLOOD-FP,with the mean absolute error concentrated within 5 mm,and the prediction time of CNN-WCA was only 0.8%that of LISFLOOD-FP.The CNN-WCA model displays a strong capacity for accurately predicting changes in inundation depths within the study area and at susceptible points for urban flooding,with the Nash-Sutcliffe e fficiency values of most flood-prone points exceeding 0.97.Furthermore,the physical connectivity of the inundation distribution predicted by CNN-WCA is better than that of the distribution obtained with a CNN.The CNN-WCA model with additional physical constraints exhibits a reduction of around 34%in instances of physical discontinuity compared to CNN.Our results prove that the CNN model with multiple physical constraints has signifi cant potential to rapidly and accurately simulate urban flooding processes and improve the reliability of prediction.
基金supported by the National Basic Research and Development Program of China (973 Program,Grant No.2007CB714100)
文摘Levee or dam failure can cause a significant disaster in most cases. A good prediction of the flood process especially in a real complex terrain is necessary for working out emergency plans for levee or dam breaches. Numerical simulations of levee or dam breach flow were carried out often with constant flow parameters and in relatively simple channels rather than in natural rivers with complex boundaries. This article presents our dedicated studies on the 2-D numerical model of levee or dam breach hydraulics with finite difference schemes. The good performance of the model is demonstrated by comparisons with the theoretical solution of an idealized dam-break flow over a frictionless flat rectangular channel. The model is also validated through its stability and conservation properties. The model is applied to simulate the flood propagation under complex boundary conditions, and the unsteady flood process in a river and in the dry floodplain with a complex bed terrain simultaneously. Furthermore, with respect to engineering practice, the numerical solutions can give special guidance to the effects of parameters such as the flood depth at different sites and the inundated area at different time periods after the levee breach and the travel time of the flood waves, which may be very important for practicing engineers in an efficient flood management.
