Estimating discharge coefficient of semi-elliptical side weir using ANFIS

• Published in: Journal of hydrology (Amsterdam) Vol. 426-427; pp. 55 - 62
• Main Authors: Dursun, O. Faruk, Kaya, Nihat, Firat, Mahmut
• Format: Journal Article
• Language: English
• Published: Kidlington Elsevier B.V 21.03.2012; Elsevier
• Subjects: ANFIS; Channel flow; Discharge coefficient; Earth sciences; Earth, ocean, space; Exact sciences and technology; Hydrology. Hydrogeology; Labyrinth; Semi-elliptical; Side weir; Labyrinth; Semi-elliptical; ANFIS; Discharge coefficient; Channel flow; Side weir; models; channels; regression; discharge; sills; performances; flow
• ISSN: 0022-1694; 1879-2707
• DOI: 10.1016/j.jhydrol.2012.01.010

► We examine discharge capacity of semi-elliptical side weirs. ► ANFIS method was applied for estimation of discharge coefficient. ► ANFIS model was trained and tested using 675 laboratory data. ► ANFIS model has shown a good performance for estimation of discharge coefficient. ► It was seen that performance of ANFIS model was better than the regression models. A labyrinth weir is defined as a weir crest that is not straight in planform. The increased sill length provided by the semi-elliptical labyrinth side weirs effectively reduces upstream head to the particular discharge. They can therefore be used to particular advantage where the width of a channel is restricted and a weir is required to pass a range of discharges with a limited variation in upstream water level. In this study, the discharge capacity of semi-elliptical side weirs is estimated by using Adaptive-Neuro Fuzzy Inference System (ANFIS). 675 Laboratory test results are used for determining discharge coefficient of semi-elliptical labyrinth side weirs. The performance of the ANFIS model is compared Multiple Linear Regression (MLR) and Nonlinear Regression (NLR) models based on performance evaluation parameters. Comparison results indicated that the ANFIS technique could be successfully employed in modeling discharge coefficient.