International Journal of Agriculture and Crop Sciences. Available online at www.ijagcs.com IJACS/2013/5-23/2804-2811 ISSN 2227-670X ©2013 IJACS Journal Development and evaluating of two-neural network model (MLP 1 and SVM 2 ) to estimate the Side weir discharge coefficient Abbas Parsaie 1 , Amir Hamzeh Haqiabi 2 1. Ph.D. student of hydro structures, Department of water Engineering, Lorestan University 2. Associate professor of water Engineering Lorestan University (Haghiabi@yahoo.com) Corresponding author email: Abbas_Parsaie@yahoo.com ABSTRACT: Prediction and modeling of hydraulic phenomenon are an important part of hydraulic engineering activities. One of the applications of Prediction and modeling is to estimate the discharge coefficient for hydraulic structures. Side weirs are widely used for flow diversion in irrigation, land drainage, urban sewage systems and also in intake structures. There are many ways to calculate flow discharge coefficient as experimental methods and computational intelligence. Experimental methods have considered due to error. Therefore using of soft computing methods to estimate the flow rate are inevitable. In this study to estimate the side weir flow coefficient two models of neural network (MLP and SVM) are developed. Training and simulation process of these models are conducted in the Matlab software environment. The results show that the MLP and SVM developed models in comparison with other empirical equations, has favorable accuracy. the Borghei equation is the best accurate equation amonge the empirical equation( 2 0.83 R  ). the performance of MLP model in the stage of preparation is better than the emperical equation( 2 0.99 Train MLP R   and 2 0.93 Validation MLP R   ) . the performance of MLP model is better than SVM model( 2 0.92 Train SVM R   and 2 0.90 Validation SVM R   ). Keywords: Side weir discharge coefficient, MLP, SVM. INTRODUCTION A side weir is an overflow and set into the side of a channel (Chow 1959). This structure is used to remove excess flow therefore it used in most water engineering project such as irrigation network, land drainage, urban sewage systems and sometimes used as an overflow dam (Ghodsian, 2004). When the water level reaches above the side weir crest it diverts a certain amount of discharge (May et al, 2003). Like all typical weir, side weir can be, sharp, broad crested with various geometric and also the flow in the main channel can be sub-critical and supercritical (Subramanya and Awasthy, 1976). Many studies have been done on defining the hydraulic behavior of this type of weir, which are often experimental method (Borghei and Parvaneh, 2011; Kumar et al, 2011; Kaya, 2011). The flow over a side weir is a typical case of spatially varied flow (SVF) with decreasing discharge (AL- TAEE., 2011). De Marchi with assuming constant energy, obtained equation of the (SVF) and to calculate the outflow discharge form the side, he solved analytically the (SVF) Finally, he proposed an equation for side weir discharge coefficient that today is known as the De Marchi coefficient (Haddadi and Rahimpour, 2012). Sharp crested rectangular side weirs is studied extensively by many researchers and usually used in water engineering project (Emiroglu et al, 2011). More recently, Laboratory Studies have on various types of this structure and also to improve the performance of these structures have been done. Rahimpour et al (2011) and Haddadi and Rahimpour and et al. (2012) done an experimental and Theoretical investigation on the flow over trapezoidal sharp a broad- 1 Multilayer Perceptron 2 Support vector machine