A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Discharge prediction of circular and rectangular side orifices using artificial neural networks
Abstract A side orifice created in the side of a channel is a structure for diverting some of the flow from the main channel for different purposes. The prediction of the discharge through this side structure is very important in hydraulic and irrigation engineering. In the present study, three artificial neural network models including feed forward back propagation, radial basis function, and Generalized Regression neural networks as well as a multiple non-linear regression method were used to predict the discharge coefficient for flow through both square and circular shapes of sharp-crested side orifices. The discharge coefficient was modeled as a function of five input non-dimensional variables resulted from five dimensional variables, which were the type of orifice shape, the diameter or width of the orifice, crest height, depth and velocity of approach flow. The results obtained in this study indicated that all of the neural network models could successfully predict the discharge coefficient with adequate accuracy. However, according to different performance measures, the accuracy of radial basis function approach was a bit better than two other neural network models. The neural network models predicted the discharge coefficient more accurately than the non-linear regression relation.
Discharge prediction of circular and rectangular side orifices using artificial neural networks
Abstract A side orifice created in the side of a channel is a structure for diverting some of the flow from the main channel for different purposes. The prediction of the discharge through this side structure is very important in hydraulic and irrigation engineering. In the present study, three artificial neural network models including feed forward back propagation, radial basis function, and Generalized Regression neural networks as well as a multiple non-linear regression method were used to predict the discharge coefficient for flow through both square and circular shapes of sharp-crested side orifices. The discharge coefficient was modeled as a function of five input non-dimensional variables resulted from five dimensional variables, which were the type of orifice shape, the diameter or width of the orifice, crest height, depth and velocity of approach flow. The results obtained in this study indicated that all of the neural network models could successfully predict the discharge coefficient with adequate accuracy. However, according to different performance measures, the accuracy of radial basis function approach was a bit better than two other neural network models. The neural network models predicted the discharge coefficient more accurately than the non-linear regression relation.
Discharge prediction of circular and rectangular side orifices using artificial neural networks
Eghbalzadeh, A. (author) / Javan, M. (author) / Hayati, M. (author) / Amini, A. (author)
KSCE Journal of Civil Engineering ; 20 ; 990-996
2015-06-12
7 pages
Article (Journal)
Electronic Resource
English
Discharge prediction of circular and rectangular side orifices using artificial neural networks
| Online Contents | 2016
Investigation of discharge and coefficients of small circular orifices
| Engineering Index Backfile | 1940
Discharge coefficients for pipe orifices
| Engineering Index Backfile | 1934
Discharge Coefficient of a Circular-Crested Side Weir in Rectangular Channels
| British Library Online Contents | 2016