Determining the Discharge Coefficient of One-Cycle Sharp-Crested U-Shape Weirs Using Kernel-Based SVM Approach

weirs are one of the hydraulic and very vital structures from various economic, environmental, safety and security aspects. In addition to passing and flow measurement, they are the controlling structures in the floods. The main purpose of the present research was to predict the discharge coefficient to method of intelligent support vector machine using experimental data as well as analysis of the hydraulic performance of weirs. The geometric variables of the weirs were used 10, 12.5, and 15 cm in height, the length of different archs were 40.82, 45, and 48.10 cm. To analysis the performance, the dimensional analysis was used by the 𝜋 Bakingham method, then by intelligent method, considering different models, the effect of different parameters in determining the discharge coefficient were investigated. As the length of the weirs arch increases, the discharge efficiency decreases, and as the length of the weirs arch increases, the crest length increases. The results of sensitivity analysis also showed that by removing the Ht/P parameter, the model error rate increases up to 14% for training data series and up to 20% for test data series, and this parameter is the most effective variable in estimating discharge coefficient in in U-shaped labyrinth weirs.