ORIGINAL RESEARCH
Evaluation of ANN, GEP, and Regression Models
to Estimate the Discharge Coefficient
for the Rectangular Broad-Crested Weir
More details
Hide details
1
Faculty of Agriculture, Bu-Ali Sina University in Hamedan, Iran
2
Department of Water Engineering, Faculty of Agriculture, Ferdowsi University of Mashhad, Iran
3
Civil Engineering Department, Faculty of Engineering, University of Malaya, Malaysia
4
Department of Water Resources and Harbor Engineering, College of Civil Engineering,
Fuzhou University, Fuzhou, China
Submission date: 2022-01-26
Final revision date: 2022-03-10
Acceptance date: 2022-03-21
Online publication date: 2022-08-02
Publication date: 2022-09-28
Corresponding author
Hamed Benisi Ghadim
College of Civil Engineering, Fuzhou University, Minhou County, 350108, Fuzhou, Fujian Province, China
Pol. J. Environ. Stud. 2022;31(5):4817-4827
KEYWORDS
TOPICS
ABSTRACT
Broad-crested weirs are structures used to measure and control the water flows in rivers, canals, and
irrigation and drainage networks. Accurate estimation of spillway discharge is one of the most striking
elements in measurement structures. So far, many researchers have studied this issue based on various
experimental conditions and a specific range of optional variables. They also have presented several
relations. In the present study, 113 data sets of Bos were used for applicability of Artificial Neural
Network (ANN), Gene expression programming (GEP), regression models to estimate the discharge
coefficient for the rectangular broad-crested weirs. The effectiveness of the models was calculated using
statistical criteria, including the coefficient of determination (R2), Root Mean Square Error (RMSE),
and mean absolute error (MAE). Comparing the models showed that the ANN with the highest R2
coefficient (0.9916), lowest RMSE = 0.0012, and MAE = 0.00052 has the best discharge coefficient
estimation than GEP models, regression models, and other empirical relations for the rectangular broadcrested
weirs.