Regression-Based Models for Predicting Discharge Coefficient of Triangular Side Orifice

- This study introduced another technique to predict the discharge coefficient (Cd) of the triangular side orifice (TSO). This technique is based on the SPSS software as multiple linear regression (MLR) and multiple nonlinear regression (MNLR) models. These models were established using 570 experimental datasets, 70 and 30% for calibration and testing stages, respectively. These sets considered five non-dimensional parameters, including (orifice crest height, orifice length, orifice height, upstream flow depth, and Froude number of the main channel). Results showed that the MLR and MNLR models in the calibrating stage had higher determination coefficients and lower errors. In addition, the importance of the input parameters was investigated, showing that the orifice crest height and Froude number highly affect the discharge coefficient value by 36%. In the testing stage, the estimated discharge coefficient by the MLR and MNLR models stayed within the range ±12 and ‡5%, respectively, of the experimental values. The MNLR model demonstrated a high level of equivalence compared to previous studies, which provided a mathematical expression to easily predict the TSO\u27s discharge coefficient