Estimating of Side Weir Discharge Coefficient by Using Neuro-Fuzzy (ANFIS)

According to the side weir application, which is widely used in irrigation and drainage systems, accurate estimation of discharge coefficient is a quite important problem for many investigators. So, in this study by using laboratory data, sugeno neuro-fuzzy with either gaussian and g-bell membership function and with hybrid and back-propagation learning methods was used to analyze discharge coefficient of inclined and flat side weir crest. In this paper 90 percent of data were used for training and 10 percent for testing. The results showed that adding fuzzy logic to artificial neural networks in shape of Adaptive Network-based Fuzzy Inference System approach with Gaussian membership function and hybrid learning method that has three fuzzy rules reduce forecast mean absolute percentage error to value of 1.58 and 5.53 respectively for inclined and flat side weir. So neuro-fuzzy method can estimate discharge coefficient more accurate than other investigation methods.