PHYSICAL, CHEMICAL SCIENCE AND ENGINEERING: ARTIFICIAL NUREL NETWORK MODELLING OF FLOW OVER WEIR.

Prediction and modeling of hydraulic phenomenon is an important part of hydraulic engineering activities. One of the applications of prediction and modeling is estimating the discharge coefficient for hydraulic structures. Weirs and spillways build for passing water flow in critical conditions or for regulating the water surface elevation. The most common types of weir crest in the practice are broad-crested weir, step broad crested weir and ogee crest weir. These structures are installed for measuring or regulating rate of flow in open channels. A broad crested weir is an overflow structure with horizontal crest above which the streamlines are nearly straight and parallel. This study presents a series of experimental investigations on flow over broad crested weir to find the hydraulic characteristics of the flow. Computer modeling of hydraulic characteristics for weir includes the calculation of water surface profiles and estimating the discharge coefficient (Cd). There are several ways to estimate the (Cd), such as experimental formulas and computational intelligence techniques. For this purpose, the physical model of broad-crested weir with rectangular channel cross-sections was tested for a different range of discharge values. The experimental results were yielded to find discharge coefficient by using the dimensional analysis. Then, the results obtained were applied with artificial neural network (ANN) techniques to investigate the applicability, ability, and accuracy of these procedures. Comparison of result from ANN procedures with these of experimental results clearly indicates that the ANN technique is efficient for the determination of discharge coefficient