A river channel terrain reconstruction method for flood simulations based on coarse DEMs

Hou, Jingming; Ma, Yuchi; Li, Bingyao; Li, Xuan; Feng, Wang; Jin, Songli; Ma, Houyu

doi:10.1016/j.envsoft.2021.105035

Cited by 10 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conventionally, hydrodynamic models have been employed for applications such as flood inundation simulations, the assessment of mitigation, and adaptation measures (Wolfs and Willems, 2013;Wang et al, 2021;Li et al, 2019). Despite the fact that the physically based models can simulate the drainage and surface flooding processes well, they usually require a large number of inputs to describe the model structure and parameters and are often computationally intensive, especially with the adoption of two-dimensional calculations (Yin et al, 2020;Jamali et al, 2018;Ziliani et al, 2019;Hou et al, 2021b). Meanwhile, there is an inevitable need for conceptualization and simplification in the physical model, and the relevant calibration and validation procedures are also quite challenging (Davidsen et al, 2017;Coulthard et al, 2013;Wu et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

A deep-learning-technique-based data-driven model for accurate and rapid flood predictions in temporal and spatial dimensions

Zhou

Situ

Liao

et al. 2023

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. An accurate and rapid urban flood prediction model is essential to support decision-making for flood management. This study developed a deep-learning-technique-based data-driven model for flood predictions in both temporal and spatial dimensions, based on an integration of long short-term memory (LSTM) network, Bayesian optimization, and transfer learning techniques. A case study in northern China was applied to test the model performance, and the results clearly showed that the model can accurately predict the maximum water depths and flood time series for various hyetograph inputs, with substantial improvements in the computation time. The model predicted flood maps 19 585 times faster than the physically based hydrodynamic model and achieved a mean relative error of 9.5 %. For retrieving the spatial patterns of water depths, the degree of similarity of the flood maps was very high. In a best case scenario, the difference between the ground truth and model prediction was only 0.76 %, and the spatial distributions of inundated paths and areas were almost identical. With the adoption of transfer learning, the proposed model was well applied to a new case study and showed robust compatibility and generalization ability. Our model was further compared with two baseline prediction algorithms (artificial neural network and convolutional neural network) to validate the model superiority. The proposed model can potentially replace and/or complement the conventional hydrodynamic model for urban flood assessment and management, particularly in applications of real-time control, optimization, and emergency design and planning.

show abstract

Section: Introductionmentioning

confidence: 99%

A deep-learning-technique-based data-driven model for accurate and rapid flood predictions in temporal and spatial dimensions

Zhou

Situ

Liao

et al. 2023

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…In this case, the hydrodynamic models, based on the numerical solutions to the 2-D shallow water equations, can reliably predict the physical processes of ood and output temporally and spatially varying ows, water depth and velocity, which can be then used to quantify the ood impact in the objective functions (Li et al, 2021;Xia et al, 2019). Some studies have greatly improved the computational e ciency of simulating the ood propagation by coupling a graphics processing unit (GPU) accelerated hydrodynamic model (Hou et al, 2021;Xia & Liang, 2016), but they still cannot meet the requirements of application in EAs. The directly in combination with ood hydrodynamics propagation is still not a common practice in a ood control optimal operation system.…”

Section: Introductionmentioning

confidence: 99%

Hydrodynamic Model and Data Driven Evolutionary Algorithm Based Coordinated Optimal Operation Method of a Multi-Dam Flood Control System

Hou

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Coordinated flood control operation of dams in a river is an optimization problem with competing decision objectives and introduces risks for decision making. Taking into account upstream and downstream flood risks, a novel method for multi-objective dam optimal operation is proposed in this study based on the hydrodynamic model and data driven evolutionary algorithm. In the proposed method, a flood control system based on hydrodynamic model was developed to simulate flood propagation and provide hydraulic characteristics, and were combined with the multi-objective particle swarm optimization (MOPSO) algorithm to design operation rules. In addition, to improve computational efficiency of the optimal operation method, the Kriging model was coupled into the optimization process to assist in the selection of potential non-dominated solutions for exact objective function evaluation. This drives the MOPSO algorithm for solution and greatly reduces unnecessary reliability evaluation. The proposed method was applied to a real flood control system of the rivers in the urban city of Chengdu, China. The results show that the developed optimal operation method is able to accurately evaluate the flood propagation using the hydrodynamic model and provide optimal operation schedules. The Pareto front of the proposed objectives demonstrates the trade-off between the competing objectives to minimize risks while satisfying flood protection requirements. The new method could provide reliable guidance for flood control scheduling of dams in flood season, and the decision makers can select the operation rules according to their risk preference.

show abstract

“…Conventionally, hydrodynamic models have been employed for applications such as flood inundation simulations, assessment of mitigation and adaptation measures (Wolfs and Willems, 2013;Wang et al, 2021;Li et al, 2019). Despite the fact that the physically-based models can well simulate the 30 drainage and surface flooding processes, they require a large number of inputs to describe the model structure and parameters, and are often computational intensive, especially with the two-dimensional calculations (Yin et al, 2020;Jamali et al, 2018;Ziliani et al, 2019;Hou et al, 2021b). Meanwhile, there is an inevitable need for conceptualization and simplification in the physical model, and the relevant calibration and validation procedures are also quite challenging (Davidsen et al, 2017;35 Coulthard et al, 2013;Wu et al, 2018).…”

mentioning

confidence: 99%

A deep learning technique-based data-driven model for accurate and rapid flood prediction

Zhou

Liao

Situ

et al. 2022

Preprint

View full text Add to dashboard Cite

Abstract. An accurate and rapid urban flood prediction model is essential to support decision-making on flood management, especially under increasing extreme precipitation conditions driven by climate change and urbanization. This study developed a deep learning technique-based data-driven flood prediction model based on an integration of LSTM network and Bayesian optimization. A case study in north China was applied to test the model performance and the results clearly showed that the model can accurately predict flood maps for various hyetograph inputs, meanwhile with substantial improvements in computation time. The model predicted flood maps 19,585 times faster than the physical-based hydrodynamic model and achieved a mean relative error of 9.5 %. For retrieving the spatial patterns of water depths, the degree of similarity of the flood maps was very high. In a best case, the difference between the ground truth and model prediction was only 0.76 % and the spatial distributions of inundated paths and areas were almost identical. The proposed model showed a robust generalizability and high computational efficiency, and can potentially replace and/or complement the conventional hydrodynamic model for urban flood assessment and management, particularly in applications of real time control, optimization and emergency design and plan.

show abstract

A river channel terrain reconstruction method for flood simulations based on coarse DEMs

Cited by 10 publications

References 37 publications

A deep-learning-technique-based data-driven model for accurate and rapid flood predictions in temporal and spatial dimensions

A deep-learning-technique-based data-driven model for accurate and rapid flood predictions in temporal and spatial dimensions

Hydrodynamic Model and Data Driven Evolutionary Algorithm Based Coordinated Optimal Operation Method of a Multi-Dam Flood Control System

A deep learning technique-based data-driven model for accurate and rapid flood prediction

Contact Info

Product

Resources

About