Side weirs are flow-regulating devices commonly encountered in hydraulic engineering. A side weir is an over flow weir framed in the side of a channel over which lateral outflow takes place when the water surface in the channel rises above the weir crest. They are widely used for flow diversion in irrigation, drainage, urban sewage systems and also in intake structures. It is essential to correctly predict the weir discharge coefficient for hydraulic engineers regarding to the technical and economical design of side weirs. Although, the discharge coefficient for flow over side weirs was investigated experimentally by many investigators, but an overall acceptable formula or a complete analytical solution of the equation governing the flow does not exist. In this study, the discharge coefficient (C M ) of triangular labyrinth and sharp crested side weirs are estimated by using artificial neural networks (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models. In this study both the discharge coefficient and the discharge capacity of side weirs are used as the target outputs. The results show that the ANN and ANFIS models predict more accurate results than those of other investigators. Also, the results show that ANN model is more capable for predicting the real discharge coefficient of sharp crested side weirs where as for the triangular labyrinth side weirs the results of discharge coefficient by ANFIS model are more accurate. Keywords: Sharp-Crested Side Weir; Labyrinth Side Weir; Discharge Coefficient; Artificial Neural Network (ANN); Adaptive Neuro-Fuzzy Inference System (ANFIS).
