Evaluation of Data-Driven and Process-Based Real-Time Flow Forecasting Techniques for Informing Operation of Surface Water Abstraction

Yassin, Mohammed; Asfaw, Alemayehu; Speight, Vanessa; Shucksmith, James

doi:10.1061/(asce)wr.1943-5452.0001397

Cited by 8 publications

(2 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Dissolved Oxygen (DO) prediction, Radial Basis Function neural networks (RBNN) model contributed higher accuracy than Multilayer Perceptron (MLP) [70]. The model's result in water quality and water quantity can be overfitting or underfitting which deteriorate model's capacity to forecast data or select trend of data based on details parameters [139]. However, with additional method from the Complete Ensemble Empirical Mode Decomposition Algorithm with Adaptive Noise (CEEMDAN) decomposition, the convolutional neural network (CNN), long short-term memory neural network (LSTM), and hybrid CNN-LSTM with different input data is analyzed.…”

Section: B Applied Knowledge Modelsmentioning

confidence: 99%

Surface Water Monitoring Systems—The Importance of Integrating Information Sources for Sustainable Watershed Management

2023

View full text Add to dashboard Cite

The complex interactions from anthropogenic activities, climate change, sedimentation and the input of wastewater has significantly affected the aquatic environment and entire ecosystem. Over the years, the researchers have investigated water monitoring approaches in terms of traditional monitoring or even integrated systems to handle such an environmental assessment and predictions based on warning systems. However, research into the selection and optimisation of water monitoring systems by the combination of parallel approach in terms of sampling techniques, process analysis and results is limited. The research objectives of the present study are to evaluate the existing water monitoring systems based on the latest approach and then provide insights into factors affecting sensor implementation at sampling locations. Here we summarize the advancement and trends of various water monitoring systems as well as the suitability of sensor placement in the area by reviewing more than 300 papers published between 2011 and 2022. The research highlights the urgency of an integrative approach with regard to water monitoring systems including water quality model and water quantity model. A framework is proposed to incorporate all water monitoring approaches, sampling techniques, and predictive models to provide comprehensive information about environmental assessment. It was observed that the urgency of model-based approaches as verification and fusion of data assemble has the ability to improve the performances of the systems. Furthermore, integrated systems with the inclusion of a separate modeling approach through integrated -semi-mechanistic models, data science and artificial intelligence are recommended in the future. Overall, this study provides guidelines for achieving standardized water management by implementing integrated water monitoring systems INDEX TERMS infrastructure, integrated monitoring, water monitoring system, modelling, water quality

show abstract

Section: B Applied Knowledge Modelsmentioning

confidence: 99%

Surface Water Monitoring Systems—The Importance of Integrating Information Sources for Sustainable Watershed Management

2023

View full text Add to dashboard Cite

show abstract

“…Unlike the process-based models which simulate the river flow conditions according to the physical or semi-physical equations that take into account various processes of the hydrological cycle, the data-based models can learn the relationships between variables and relate inputs to the outputs without a detailed understanding of the physical processes (EL Bilali et al 2020;Rokoni et al 2022). One of the advantages of these data-based models over other methods is their capability to learn long-term time dependency on the data, which has shown accurate river flow prediction (Yassin et al 2021).…”

Section: Introductionmentioning

confidence: 99%

Estimation of river high flow discharges using friction-slope method and hybrid models

Shirazi

Zahiri

Piri

et al. 2023

Preprint

View full text Add to dashboard Cite

Severe flood is considered as one of the most important hydro-geomorphic hazards in catchment and urban scales. The correct estimation of flood flow in rivers is an important issue and plays a significant role in the optimal use of water resources, operation of dam reservoirs, and the design and planning of flood control projects. Despite the advances in hydrological models for flood prediction, these models suffer from many drawbacks and still have errors for the prediction. Hydraulic models also face with many uncertainties especially in riverbed Manning roughness coefficient and energy slope of the river. In this study, flood discharge estimation based on a new method of Friction-Slope or parameter α has been discussed for three alluvial rivers located in Golestan province in Iran. The calculation method is based on the Manning formula and takes into account the effects of energy slope of the river, as well as Manning roughness coefficient of the riverbed. For calculation of parameter α based on the simple and frequently available input variables of the river cross section (flow depth, flow area and hydraulic radius), hybrid methods of Particle Swarm Optimization-Support Vector Regression (SVR-PSO), Grey Wolf Optimization- Support Vector Regression (SVR- GWO), and Response Surface Method-Support Vector Regression (SVR-RSM) have been used and finally river flow discharges have been calculated and compared with the measured data. The optimum structures of hybrid models were determined using statistical evaluation criteria such as coefficient of determination (R²), mean absolute error (MAE), root mean square error (RMSE) and conformity index (d). The results showed that the SVR-RSM model has the best accuracy with R2=0.97, MAE=0.22, RMSE=1.66 and d=0.99 in the test phase. After determining parameter α using the RSM-SVR model, the river flow rates were calculated and compared with the observed values.

show abstract

Estimation of River High Flow Discharges Using Friction-Slope Method and Hybrid Models

Shirazi,

Zahiri,

Piri

et al. 2024

Water Resour Manage

View full text Add to dashboard Cite

Evaluation of Data-Driven and Process-Based Real-Time Flow Forecasting Techniques for Informing Operation of Surface Water Abstraction

Cited by 8 publications

References 51 publications

Surface Water Monitoring Systems—The Importance of Integrating Information Sources for Sustainable Watershed Management

Surface Water Monitoring Systems—The Importance of Integrating Information Sources for Sustainable Watershed Management

Estimation of river high flow discharges using friction-slope method and hybrid models

Estimation of River High Flow Discharges Using Friction-Slope Method and Hybrid Models

Contact Info

Product

Resources

About