Determining dynamic water level control boundaries for a multi‐reservoir system during flood seasons with considering channel storage

River flood routing computes changes in the shape of a flood wave over time as it travels downstream along a river. Conventional flood routing models, especially hydrodynamic models, require a high quality and quantity of input data, such as measured hydrologic time series, geometric data, hydraulic structures, and hydrological parameters. Unlike physically based models, machine learning algorithms, which are data-driven models, do not require much knowledge about underlying physical processes and can identify complex nonlinearity between inputs and outputs. Due to their higher performance, lower complexity, and low computation cost, researchers introduced novel machine learning methods as a single application or hybrid application to achieve more accurate and efficient flood routing. This paper reviews the recent application of machine learning methods in river flood routing.

Section: Introductionmentioning

confidence: 99%

Review of Machine Learning Methods for River Flood Routing

Li,

Jun

2024

“…This is particularly crucial when attempting to achieve flood routing for complex reaches that are difficult to model accurately using physical models. This paper compares and analyzes the effectiveness of different common ML models by using two complex reaches of the Yangtze River as case studies; some reaches of the Yangtze River have been used for flood routing by the Muskingum models [50,51], numerical methods [52,53], ML models [23,26], and so on. These two reaches were chosen because they are representative and necessary for river flood routing.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Analysis of Multiple Machine Learning Methods for Flood Routing in the Yangtze River

Zhou

Kang

2023

Obtaining more accurate flood information downstream of a reservoir is crucial for guiding reservoir regulation and reducing the occurrence of flood disasters. In this paper, six popular ML models, including the support vector regression (SVR), Gaussian process regression (GPR), random forest regression (RFR), multilayer perceptron (MLP), long short-term memory (LSTM) and gated recurrent unit (GRU) models, were selected and compared for their effectiveness in flood routing of two complicated reaches located at the upper and middle main stream of the Yangtze River. The results suggested that the performance of the MLP, LSTM and GRU models all gradually improved and then slightly decreased as the time lag increased. Furthermore, the MLP, LSTM and GRU models outperformed the SVR, GPR and RFR models, and the GRU model demonstrated superior performance across a range of efficiency criteria, including mean absolute percentage error (MAPE), root mean square error (RMSE), Nash–Sutcliffe efficiency coefficient (NSE), Taylor skill score (TSS) and Kling–Gupta efficiency (KGE). Specifically, the GRU model achieved reductions in MAPE and RMSE of at least 7.66% and 3.80% in the first case study and reductions of 19.51% and 11.76% in the second case study. The paper indicated that the GRU model was the most appropriate choice for flood routing in the Yangtze River.

“…Zhang et al [12] discussed and improved the DC-FLWL on the basis of forecast information. Gong et al [13] conceptualized river flood routing as a hypothetical reservoir based on the Muskingum model and analyzed its impact on dynamic control boundaries. Ning et al [14] assessed flood risk by considering flood forecast uncertainty and analyzed the impacts at different FLWLs.…”

Section: Introductionmentioning

confidence: 99%

Research on Flood Risk Control Methods and Reservoir Flood Control Operation Oriented towards Floodwater Utilization

Zhou,

Kang,

Hou

et al. 2023

Since improving floodwater utilization may increase flood risk, flood risk control methods for trade-offs between these factors have research value. This study presented a flood risk control method oriented towards floodwater utilization which considers multiple main flood risk factors. The proposed method not only achieves the boundaries of the flood limited water level (FLWL) under various acceptable risks but also dynamically controls the water level to enhance floodwater utilization. A case study conducted on the Danjiangkou reservoir yielded the following results: (1) The proposed method provides FLWL dynamic control boundaries under various acceptable risks. (2) The proposed method reveals the potential to raise the FLWL, with a possibility to raise it by 1.00 m above the present FLWL under the absence of flood risk. (3) The available flood resources in both the wet and dry seasons increase, on average, by 0.83 and 0.81 billion m3, and the flood risk remains within the acceptable range after raising the FLWL by 1.00 m, which contributes to enhancing floodwater utilization.