A Multiple Model Approach for Flood Forecasting, Simulation, and Evaluation Coupling in Zhouqu County

Flood disasters are considered to be one of the ten natural disasters that threaten the survival of mankind. They occur frequently and have a serious impact on the national economy. For quicker response to the sudden flood, in this paper, the relevant characteristics of flood forecasting and disaste...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Water (Basel) Jg. 15; H. 24; S. 4246
Hauptverfasser: Li, Yongfeng, Liu, Yi, Liu, Xiaoming, Shen, Chao
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Basel MDPI AG 01.12.2023
Schlagworte:
Accuracy
Algorithms
Analysis
China
Disasters
Emergency communications systems
Flood forecasting
Floods
Forecasting
Fuzzy sets
Genetic algorithms
humans
hydrologic models
Hydrology
Methods
Natural disasters
Neural networks
Optimization
Risk assessment
rivers
Runoff
Simulation
Time series
water
Watersheds
China
ISSN:2073-4441, 2073-4441
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Flood disasters are considered to be one of the ten natural disasters that threaten the survival of mankind. They occur frequently and have a serious impact on the national economy. For quicker response to the sudden flood, in this paper, the relevant characteristics of flood forecasting and disaster assessment are comprehensively studied to establish the corresponding models, and a multi-objective culture shuffled complex differential evolution (MOCSCDE) algorithm is proposed to optimize the model parameters. It can achieve better convergence and significantly improve the model accuracy. Then, a river hydrodynamic model is established to simulate the flooding process, and the characteristics of flood evolution, such as water depth, flow speed, duration, and submerged area, are analyzed. Third, based on the above-mentioned flood forecasting and flood evolution calculations, the relative membership function (VFS) is determined via the set pair analysis method (SPA), and the variable fuzzy set model (SPAVFS) is used for flood risk assessment. Finally, through the study of flow forecasting at Zhouqu hydrological station, it is found that the accuracy of the forecast result of the built model is best compared with LSTM and XAJ model, the mean relative error is only 7.6%, and the certainty coefficient can reach 0.96, which surpass the baselines by 20% and 7.9%.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2073-4441
2073-4441
DOI:10.3390/w15244246