Real‐time recurrent learning neural network for stream‐flow forecasting

Chang, Fi‐John; Chang, Li-Chiu; Huang, Hau-Lung

doi:10.1002/hyp.1015

Cited by 109 publications

(64 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Artificial neural networks (ANN) have been recently accepted as an efficient alternative tool for modeling of complex hydrologic system to the conventional methods and widely used for prediction. Some specific applications of ANN to hydrology include modeling rainfallrunoff process (Sajikumar et al, 1999), river flow forecasting (Dibike et al, 2001;Chang et al, 2002;Sudheer and Jain;Dawson et al, 2002), sediment transport prediction (Firat and Güngör, 2004), and sediment concentration estimation (Nagy et al, 2002). The ASCE Task Committee reports (2000) did a comprehensive review of the applications of ANN in hydrological forecasting context.…”

Section: Introductionmentioning

confidence: 99%

Comparison of Artificial Intelligence Techniques for river flow forecasting

Fırat

2008

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. The use of Artificial Intelligence methods is becoming increasingly common in the modeling and forecasting of hydrological and water resource processes. In this study, applicability of Adaptive Neuro Fuzzy Inference System (ANFIS) and Artificial Neural Network (ANN) methods, Generalized Regression Neural Networks (GRNN) and Feed Forward Neural Networks (FFNN), and Auto-Regressive (AR) models for forecasting of daily river flow is investigated and Seyhan River and Cine River was chosen as case study area. For the Seyhan River, the forecasting models are established using combinations of antecedent daily river flow records. On the other hand, for the Cine River, daily river flow and rainfall records are used in input layer. For both stations, the data sets are divided into three subsets, training, testing and verification data set. The river flow forecasting models having various input structures are trained and tested to investigate the applicability of ANFIS and ANN and AR methods. The results of all models for both training and testing are evaluated and the best fit input structures and methods for both stations are determined according to criteria of performance evaluation. Moreover the best fit forecasting models are also verified by verification set which was not used in training and testing processes and compared according to criteria. The results demonstrate that ANFIS model is superior to the GRNN and FFNN forecasting models, and ANFIS can be successfully applied and provide high accuracy and reliability for daily river flow forecasting.

show abstract

Section: Introductionmentioning

confidence: 99%

Comparison of Artificial Intelligence Techniques for river flow forecasting

Fırat

2008

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…However, even cross referencing and the use of citations is not a foolproof method, since previous errors and omissions can be propagated throughout the field e.g. incorrect bracket notation in a published equation (MAE;Chang et al, 2002). It is better to be consistent and transparent and to promote collective development activities, that are open to inspection, and that would support and encourage maximum potential testing, correctness and peer group endorsement.…”

mentioning

confidence: 99%

HydroTest: A web-based toolbox of evaluation metrics for the standardised assessment of hydrological forecasts

Dawson

Abrahart

See

2007

Environmental Modelling & Software

448

244

View full text Add to dashboard Cite

“…Indeed they have been successfully tested as R-R models by, for example, Hsu et al (1997), Coulibaly et al (2000), Chang et al (2002) and Chiang et al (2004), but the number of applications using feedforward ANNs dwarfs those with recurrent ANNs. The main reason for this is that recurrency in ANNs causes increased complexity of the training procedure as a result of the cyclical network connections, and subsequent convergence problems for training algorithms (Atiya and Parlos, 2000;Lukoševičius and Jaeger, 2009).…”

Section: Introductionmentioning

confidence: 99%

Echo state networks as an alternative to traditional artificial neural networks in rainfall–runoff modelling

Vos

2013

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. Despite theoretical benefits of recurrent artificial neural networks over their feedforward counterparts, it is still unclear whether the former offer practical advantages as rainfall-runoff models. The main drawback of recurrent networks is the increased complexity of the training procedure due to their architecture. This work uses the recently introduced and conceptually simple echo state networks for streamflow forecasts on twelve river basins in the Eastern United States, and compares them to a variety of traditional feedforward and recurrent approaches. Two modifications on the echo state network models are made that increase the hydrologically relevant information content of their internal state. The results show that the echo state networks outperform feedforward networks and are competitive with stateof-the-art recurrent networks, across a range of performance measures. This, along with their simplicity and ease of training, suggests that they can be considered promising alternatives to traditional artificial neural networks in rainfallrunoff modelling.

show abstract

Real‐time recurrent learning neural network for stream‐flow forecasting

Cited by 109 publications

References 12 publications

Comparison of Artificial Intelligence Techniques for river flow forecasting

Comparison of Artificial Intelligence Techniques for river flow forecasting

HydroTest: A web-based toolbox of evaluation metrics for the standardised assessment of hydrological forecasts

Echo state networks as an alternative to traditional artificial neural networks in rainfall–runoff modelling

Contact Info

Product

Resources

About