A dynamic artificial neural network model for forecasting nonlinear processes

Ghiassi, M.; Nangoy, Stanley

doi:10.1016/j.cie.2008.11.027

Cited by 40 publications

(17 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ANNs are statistical models built and maintained through an iterative training process. The ANN accumulates knowledge at each model layer through a self‐learning process until a model is created that accurately captures the behavior of the process being modeled and can be used to forecast future values [ Ghiassi and Nangoy , 2009]. ANNs have been offered as effective alternatives to traditional linear modeling approaches because of their ability to explicitly analyze nonlinear time series events.…”

Section: Methodological Reviewmentioning

confidence: 99%

“…Research involving forecasting of peak weekly water demand in Nicosia, Crete, found that the ANN model utilizing the Levenberg‐Marquardt learning algorithm was the most accurate for short‐term forecasting when compared to several other ANNs using different learning algorithms [ Adamowski and Karapataki , 2010]. One ANN, the dynamic architecture for artificial neural networks (DAN2), models nonlinearity through a transfer function of a weighted and normalized sum of the input variables [ Ghiassi and Nangoy , 2009]. DAN2 performance was compared to ARIMA for modeling future water demand at multiple temporal scales and was found to perform significantly better than the ARIMA method [ Ghiassi and Nangoy , 2009].…”

Section: Methodological Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Urban water demand modeling: Review of concepts, methods, and organizing principles

House‐Peters

Chang

2011

Water Resources Research

338

184

View full text Add to dashboard Cite

[1] In this paper, we use a theoretical framework of coupled human and natural systems to review the methodological advances in urban water demand modeling over the past 3 decades. The goal of this review is to quantify the capacity of increasingly complex modeling techniques to account for complex human and natural processes, uncertainty, and resilience across spatial and temporal scales. This review begins with coupled human and natural systems theory and situates urban water demand within this framework. The second section reviews urban water demand literature and summarizes methodological advances in relation to four central themes: (1) interactions within and across multiple spatial and temporal scales, (2) acknowledgment and quantification of uncertainty, (3) identification of thresholds, nonlinear system response, and the consequences for resilience, and (4) the transition from simple statistical modeling to fully integrated dynamic modeling. This review will show that increasingly effective models have resulted from technological advances in spatial science and innovations in statistical methods. These models provide unbiased, accurate estimates of the determinants of urban water demand at increasingly fine spatial and temporal resolution. Dynamic models capable of incorporating alternative future scenarios and local stochastic analysis are leading a trend away from deterministic prediction.Citation: House-Peters, L. A., and H. Chang (2011), Urban water demand modeling: Review of concepts, methods, and organizing principles, Water Resour. Res., 47, W05401,

show abstract

Section: Methodological Reviewmentioning

confidence: 99%

Section: Methodological Reviewmentioning

confidence: 99%

Urban water demand modeling: Review of concepts, methods, and organizing principles

House‐Peters

Chang

2011

Water Resources Research

338

184

View full text Add to dashboard Cite

show abstract

“…In a method called additive or constructive, the model starts with a minimal network consisting of a single hidden layer, and, gradually, hidden layers and neurons are added and the e ectiveness of the obtained model is evaluated using the evaluation instruments. In the second approach, the model starts with a very large network; moreover, pruning algorithm is used to reduce the size of the model [68].…”

Section: Arti Cial Neural Network Developmentmentioning

confidence: 99%

Predicting Shear Wave Velocity of Soil Using Multiple Linear Regression Analysis and Artificial Neural Networks

et al. 2017

View full text Add to dashboard Cite

Abstract. This paper investigates the correlation between shear wave velocity and some of the index parameters of soils, including Standard Penetration Test blow counts (SPT), FineContent (FC), soil moisture (W ), Liquid Limit (LL), and Depth (D). The study attempts to show the application of arti cial neural networks and multiple regression analysis to the prediction of the shear wave velocity (V S ) value of soils. New prediction equations are suggested to correlate VS with the mentioned parameters based on a dataset collected from Mashhad city in the north east of Iran. The results suggest that, in the case of ANN method use, highly accurate correlations in the estimation of VS are acquired. The predicted values using ANN method are checked against the real values of V S to evaluate the performance of this method. The minimum correlation coe cient obtained in ANN method is higher than the maximum correlation coe cient obtained from the MLR. In addition, the value of estimation error in the ANN method is much less than that in the MLR method, indicating the role of higher con dence coe cient of the ANN in estimating VS of soil.

show abstract

“…In this study, a feed-forward back-propagation-based ANN solver, although there are fewer guidelines on the optimal number of hidden nodes [38] . To find the optimal network that allows most accurate ANNs for fitting target, different numbers of hidden layers (1 and 2) and nodes (9)(10)(11)(12)(13)(14)(15) in every hidden layer were tested, and then, the bests were selected.…”

Section: Artificial Neural Network (Ann)mentioning

confidence: 99%

“…The inputs were control factors from A to E in the levels of 1, 2 or 3 and the target was the shrinkage percentage, warpage or average of normalized shrinkage and warpage. [38] . The training R 2 based on mean squared error was calculated by the software but the forecasting R 2 was calculated using Eq.…”

Section: Artificial Neural Network (Ann)mentioning

confidence: 99%

Warpage and Shrinkage Optimization of Injection-Molded Plastic Spoon Parts for Biodegradable Polymers Using Taguchi, ANOVA and Artificial Neural Network Methods

Oliaei,

Heidari,

Davachi

et al. 2016

Journal of Materials Science & Technology

View full text Add to dashboard Cite

A dynamic artificial neural network model for forecasting nonlinear processes

Cited by 40 publications

References 23 publications

Urban water demand modeling: Review of concepts, methods, and organizing principles

Urban water demand modeling: Review of concepts, methods, and organizing principles

Predicting Shear Wave Velocity of Soil Using Multiple Linear Regression Analysis and Artificial Neural Networks

Warpage and Shrinkage Optimization of Injection-Molded Plastic Spoon Parts for Biodegradable Polymers Using Taguchi, ANOVA and Artificial Neural Network Methods

Contact Info

Product

Resources

About