Monthly flow forecast for Mississippi River basin using artificial neural networks

Chandrasekaran, S.; Vanitha, S.; Muttil, Nitin; Suganya, K.; Suji, S.; Selvi, M Thamarai; Selvi, R. Tamil; Sudha, S.

doi:10.1007/s00521-013-1419-6

Cited by 19 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many different types of training algorithms are available with various characteristics and performances, among which the back propagation algorithm is most commonly used. This algorithm updates the weights of the network by adjusting the error between observed and predicted, finally leading to a trained network after repeating this process a sufficient number of times (Haykin, 1999;ASCE Task Committee, 2000a;2000b;Nourani et al, 2011;Ajmera and Goyal, 2012;Sivapragasam et al, 2014). Many researchers have used an ANN as a "forecasting model" in the field of atmospheric sciences and meteorology (Gheiby et al, 2003;Tapiador et al, 2004;Chattopadhyay and Chattopadhyay, 2008;Dahamsheh and Aksoy, 2009;Babel et al, 2015;Collins and Tissot, 2015;Valipour, 2016;Modarres et al, 2018).…”

Section: Artificial Neural Networkmentioning

confidence: 99%

An investigation on the relationship between the Hurst exponent and the predictability of a rainfall time series

Chandrasekaran

Poomalai

Balamurali

et al. 2019

Meteorological Applications

View full text Add to dashboard Cite

Rainfall prediction is a very challenging task due to its dependence on many meteorological parameters. Because of the complex nature of rainfall, the uncertainty associated with its predictability continues to be an issue in rainfall forecasting. The Hurst exponent is considered as a measure of persistence and it is believed that if a time series has persistence (as reflected by a Hurst exponent value greater than 0.5) it is also predictable. However, very limited studies have been carried out to demonstrate this hypothesis. This study, through experimental work on hypothetical data as well as real data, demonstrates that the Hurst exponent can be taken as an indicator for predictability provided that the exponent values at “smaller levels” of the time series are also significantly greater than 0.5 together with the Hurst exponent of the overall time series. It is also shown that it is better to predict the “similarity” aspect associated with a time series (and derive the predicted rainfall) than to predict the rainfall directly.

show abstract

Section: Artificial Neural Networkmentioning

confidence: 99%

An investigation on the relationship between the Hurst exponent and the predictability of a rainfall time series

Chandrasekaran

Poomalai

Balamurali

et al. 2019

Meteorological Applications

View full text Add to dashboard Cite

show abstract

“…Thus, the inputs of the net behave like a sliding window buffer. The MLP network has been used for predicting river flow discharges [13], tourism demand to Catalonia [14], stock market [15], flow forecast for Mississippi River [16], earthquakes in Chile [17], rainfall [18], groundwater level [19] and price of oilseeds in India [20]. As mentioned above, besides static neural networks, recurrent neural networks (RNNs) have been widely used for time series prediction.…”

Section: Previous Work Donementioning

confidence: 99%

“…Forecasting from historic values[13][14][15][16][17][18][19][20][21][22][23][24][25][26]. 2 Prediction of a complete time series in a specific time interval from another distinct time series[27].3 Prediction of complete time series in a specific time interval from multiple and distinct time series (proposed solution) Neural Comput & Applic…”

mentioning

confidence: 99%

A new approach for dynamic modelling of energy consumption in the grinding process using recurrent neural networks

Arriandiaga

Portillo

Sánchez

et al. 2015

Neural Comput & Applic

View full text Add to dashboard Cite

Grinding is a critical machining process because it produces parts of high precision and high surface quality. Due to the semi-artisan production of the wheel, it is not possible to know in advance the performance of the wheel. One of the most useful parameters to characterize the grinding process is the specific grinding energy, which varies with the wear of the grinding wheel during its lifecycle. Thus, it would be useful to model the specific grinding energy in order to get information about the performance of the wheel before buying it. Unlike the typical applications of time series forecasting, in this work, a totally different issue is presented: the prediction of new and complete time series bounded in time without initial or historic values. In this context, an analysis of the effect of the time characteristics and the number of points of the time series on the prediction capabilities of the ANN is presented. The results of the analysis show that 200 points are enough to predict a complete time series up to 2000 mm 3 /mm of specific volume of material removed. Actually, it is shown that modelling the evolution of the grinding specific energy up to 2000 mm 3 /mm is possible. The net shows good capability to generalize to new grinding conditions, with errors below 23.65 %, and to new wheel characteristics, with errors below 20.01 %, which are satisfactory from the grinding process perspective.

show abstract

“…Streamflow is known as multidimensional and highly nonlinear nature, therefore forecasting accurate streamflow is very challenging [1,2,3]. In the past few decades, there are various application of AI in streamflow forecasting, for instance, artificial neural network (ANN) [4,5,6], fuzzy logic [7,8], genetic programming (GP) [9,10,11], support vector machine [12,13,14,15] and hybrid model [16,17,18,19]. ANN has the ability of mapping nonlinear data and found to be a reliable method for streamflow forecasting as it able to learn and generalize non-linear time series data [20,21,22].…”

Section: Introductionmentioning

confidence: 99%

Support Vector Machine and Neural Network based Model for Monthly Stream Flow Forecasting

Zaini¹,

Malek²,

Yusoff³

et al. 2018

IJET

View full text Add to dashboard Cite

Accurate forecasting of streamflow is desired in many water resources planning and management, flood prevention and design development. In this study, the accuracy of two hybrid model, support vector machine - particle swarm optimization (SVM-PSO) and bat algorithm – backpropagation neural network (BA-BPNN) for monthly streamflow forecasting at Kuantan River located in Peninsular Malaysia are investigated and compared to regular SVM and BPNN model. Heuristic optimization namely PSO and BA are introduced to find the optimum SVM and BPNN parameters. The input parameters to the forecasting models are antecedent streamflow, historical rainfall and meteorological parameters namely evaporation, temperature, relative humidity and mean wind speed. Two performance evaluation measure, root mean square error (RMSE) and coefficient of determination (R2) were employed to evaluate the performance of developed forecasting model. It is found that, RMSE and R2 for hybrid SVM-PSO are 24.8267 m3/s and 0.9651 respectively while general SVM model yields RMSE of 27.5086 m3/s and 0.9305 of R2 for testing phase. Besides that, hybrid BA-BPNN produces RMSE, 17.7579 m3/s and R2, 0.7740 while BPNN model produces lower RMSE and R2 of 28.1396 m3/s and 0.5015 respectively. Therefore, the results indicate that hybrid model, SVM-PSO and Bat-BPNN yield better performance as compared to general SVM and BPNN, respectively in streamflow forecasting.

show abstract

Monthly flow forecast for Mississippi River basin using artificial neural networks

Cited by 19 publications

References 20 publications

An investigation on the relationship between the Hurst exponent and the predictability of a rainfall time series

An investigation on the relationship between the Hurst exponent and the predictability of a rainfall time series

A new approach for dynamic modelling of energy consumption in the grinding process using recurrent neural networks

Support Vector Machine and Neural Network based Model for Monthly Stream Flow Forecasting

Contact Info

Product

Resources

About