GMDH-type neural network approach for modeling the discharge coefficient of rectangular sharp-crested side weirs

Estimating the discharge coefficient using hydraulic and geometrical specifications is one of the influential factors in predicting the discharge passing over a side weir. Taking into account the fact that existing equations are incapable of estimating the discharge coefficient well, artificial intelligence methods are used to predict it. In this study, Group Method of Data Handling (GMDH) was used for the purpose of predicting the discharge coefficient in a side weir. The Froude number (F1), weir dimensionless length (b/B), ratios of weir length to depth of upstream flow (b/y1) and weir height to its length (p/y1) were taken as input parameters to express a new model for predicting the discharge coefficient. Two different sets of laboratory data were used to train the artificial network and test the new model. Different statistical indexes were used to evaluate the performance of the GMDH model presented for two states, training and testing. The results indicate that the proposed model predicts the discharge coefficient precisely (MAPE = 5.263 & RMSE = 0.038) and this model is more accurate in predicting than the feed-forward neural network model and existing nonlinear regression equations.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.