Artificial Intelligence (AI) Studies in Water Resources

Ay, Murat; Özyıldırım, Serhat

doi:10.28978/nesciences.424674

Cited by 8 publications

(6 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Equations (2)- (7) describe the processes in an LSTM block. Equations (2)-(4) represent the input, output and forget gate.…”

Section: Deep Learning Based On Lstm Neural Networkmentioning

confidence: 99%

“…h t = o t × tanh(c t ) (7) Water 2019, 11 The classical LSTM block structure shown in Figure 2b consists of different processes called gates [37]. These gates compute the desired output from a new input data at a time , along with elements obtained from the previous time step − 1.…”

Section: Deep Learning Based On Lstm Neural Networkmentioning

confidence: 99%

“…Furthermore, an early warning system should be able to give accurate warning for extreme events, but also should be able to give accurate forecasts for small flows to avoid false positive warnings that eventually reduce the reliability of the system [6]. Given that the early warning systems give more importance to the simplicity and robustness of the forecasting model rather than an accurate description of the various internal sub-processes, it is certainly worth considering hybrid models that combine physical-based models with data-driven approaches for improving real-time runoff forecasts [7][8][9][10]. Another difficulty in predicting hydrological extremes is that to get ahead of an extreme event requires the implementation of weather forecasting models coupled to hydrological models.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Hydrological Early Warning System Based on a Deep Learning Runoff Model Coupled with a Meteorological Forecast

Fuente

Meruane

2019

Water

View full text Add to dashboard Cite

The intensification of the hydrological cycle because of global warming raises concerns about future floods and their impact on large cities where exposure to these events has also increased. The development of adequate adaptation solutions such as early warning systems is crucial. Here, we used deep learning (DL) for weather-runoff forecasting in región Metropolitana of Chile, a large urban area in a valley at the foot of the Andes Mountains, with more than 7 million inhabitants. The final goal of this research is to develop an effective forecasting system to provide timely information and support in real-time decision making. For this purpose, we implemented a coupled model of a near-future global meteorological forecast with a short-range runoff forecasting system. Starting from a traditional hydrological conceptual model, we defined the hydro-meteorological and geomorphological variables that were used in the data-driven weather-runoff forecast models. The meteorological variables were obtained through statistical scaling of the Global Forecast System (GFS), thus enabling near-future prediction, and two data-driven approaches were implemented for predicting the entire hourly flow time-series in the near future (3 days), a simple Artificial Neural Networks (ANN) and a Deep Learning (DL) approach based on Long-Short Term Memory (LSTM) cells. We show that the coupling between meteorological forecasts and data-driven weather-runoff forecast models are able to satisfy two basic requirements that any early warning system should have: The forecast should be given in advance, and it should be accurate and reliable. In this context, DL significantly improves runoff forecast when compared with a traditional data-driven approach such as ANN, being accurate in predicting time-evolution of output variables, with an error of 5% for DL, measured in terms of the root mean square error (RMSE) for predicting the peak flow, compared to 15.5% error for ANN, which is adequate to warn communities at risk and initiate disaster response operations.

show abstract

“…Equations (2)- (7) describe the processes in an LSTM block. Equations (2)-(4) represent the input, output and forget gate.…”

Section: Deep Learning Based On Lstm Neural Networkmentioning

confidence: 99%

Section: Deep Learning Based On Lstm Neural Networkmentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Hydrological Early Warning System Based on a Deep Learning Runoff Model Coupled with a Meteorological Forecast

Fuente

Meruane

2019

Water

View full text Add to dashboard Cite

show abstract

“…It is stated that water crisis due to flood and drought could be solved Research Article effectively with adaptive and integrated reservoir management approach [8]. Consequently, optimal longterm reservoir management through modern Artificial Intelligence (AI) technologies as well as the best operational practice driven by up-to-date reservoir operating policy have been proposed and brought into action to cope fruitfully with natural disasters [9].…”

Section: Introductionmentioning

confidence: 99%

Re–operating the Bhumibol and Sirikit Dams Using Hybrid Neuro–Fuzzy Technique to Solve the Water Scarcity and Flooding Problems in the Chao Phraya River Basin

Muyar Kyaw,

Rittima,

Phankamolsil

et al. 2024

App. Envi. Res.

View full text Add to dashboard Cite

The decision support system to reservoir re–operation using Artificial Intelligence has been broadly studied and proven in term of the operational performances for both single and multiple reservoir system, this study applied Adaptive Neuro Fuzzy Inference System (ANFIS) technique for reservoir re–operation in Chao Phraya River Basin aiming to reduce water scarcity and flooding problems in the central region of Thailand. ANFIS is an integrated approach in which neural networks are utilized to enhance the fuzzy inference system and create fuzzy “IF–Then” reservoir operational guidelines with proper membership functions for reservoir re–operation. In this study, ANFIS operating rules were trained using two different datasets; long–term dataset (scenario 1) and water year–based dataset (scenario 2). It is revealed that the extent of yearly water deficit in critical dry years are totally reduced to nearly zero when re–operating with ANFIS operation rules, except in the year 2012. However, the yearly water deficit in year 2012 is also substantially reduced from 504 MCM by the current operation to 127 and 119 MCM for scenario 1 and scenario 2, respectively when two scenarios of ANFIS–based reservoir re–operation model were performed. Moreover, considerable total amount of spilled water from BB and SK Dams is definitely declined to 0 and 37 MCM in years 2002 and 2011, respectively when water year–based ANFIS model was implemented. In addition, it is expressed that average water storages of two main dams obtained from two scenarios of ANFIS model are substantially increased up to +6.08% and +6.94% for BB Dam and +0.09% and +1.62% for SK Dam in comparison with the current operation. This signifies that supplying water from dams to meet the target water demand through adaptive fuzzy–rules can be well handled and flooded water can be minimized.

show abstract

“…Ay and Kişi [19] used ANN and ANFIS to estimate DO concentration, which was compared with the multiple linear regressions. The models are compared among one other and results indicated that the ANN model was close to accuracy to determine monthly mean DO concentration, thus making artificial intelligence suitable to study water resources [20].…”

Section: Introductionmentioning

confidence: 99%

Performance assessment of data driven water models using water quality parameters of Wangchu river, Bhutan

et al. 2022

View full text Add to dashboard Cite

Multifarious anthropogenic activities triggered by rapid urbanization has led to contamination of water sources at unprecedented rate, with less surveillance, investigation and mitigation. The use of artificial intelligence (AI) in tracking and predicting water quality parameters has surpassed the use of other conventional methods. This study presents the assessment of three main models: adaptive neuro fuzzy inference system (ANFIS), artificial neural network (ANN) and multiple linear regression (MLR) on water quality parameters of Wangchu river located at capital city of Bhutan. The performance and predictive ability of these models are compared and the optimal model for predicting the parameters are recommended based on the coefficient correlation (CC), root mean square error (RMSE) and Nash–Sutcliffe efficiency (NSE) evaluation criteria. Overall NSE and RMSE, the ANN model predicted parameters with maximum efficiency of 97.3 percent and minimum error of 8.57. The efficiency of MLR and ANFIS model are 95.9 percent and 94.1 percent respectively. The overall error generated by MLR and ANFIS are 10.64 and 12.693 respectively. From the analysis made, the ANN is recommended as the most suitable model in predicting the water quality parameters of Wangchu river. From the six-training function of ANN, trainBR (Bayesian Regularization) achieved the CC of 99.8%, NSE of 99.3% and RMSE of 9.822 for next year data prediction. For next location prediction, trainBR achieved CC of 99.2%, NSE of 98.4% and RMSE of 6.485, which is the higher correlation and maximum efficiency with less error compared to rest of the training functions. The study represents first attempt in assessing water quality using AI technology in Bhutan and the results showed a positive conclusion that the traditional means of experiments to check the quality of river water can be substituted with this reliable and realistic data driven water models. Article highlights Total dissolved solids (TDS), electrical conductivity (EC), potential of hydrogen (pH) and dissolved oxygen (DO) are selected as main water quality parameters as data for modeling. Artificial neural network model gives highest efficiency and accuracy compared to MLR and ANFIS model. Use of artificial intelligence shows better performance to provide water quality and future predictions over conventional methods leading to conservation of water resources and sustainability.

show abstract

Artificial Intelligence (AI) Studies in Water Resources

Cited by 8 publications

References 55 publications

Hydrological Early Warning System Based on a Deep Learning Runoff Model Coupled with a Meteorological Forecast

Hydrological Early Warning System Based on a Deep Learning Runoff Model Coupled with a Meteorological Forecast

Re–operating the Bhumibol and Sirikit Dams Using Hybrid Neuro–Fuzzy Technique to Solve the Water Scarcity and Flooding Problems in the Chao Phraya River Basin

Performance assessment of data driven water models using water quality parameters of Wangchu river, Bhutan

Contact Info

Product

Resources

About