Daily Discharge Prediction Using Artificial Neural Networks (ANNs) For Al Gharraf River in Thi Qar Province, Iraq

  • Ahmed A. Dakheel
Keywords: Artificial Neural Networks (ANNs), Regulator, Al Badaa, Al Gharraf River, Thi Qar Province, Iraq


In the present study an Artificial Neural Networks (ANNs) model has been developed for Al Gharraf River in Thi Qar Province, Iraq. The modeled network is trained, validated and tested using daily discharge data pertaining to 3 years (January 2014 to January 2017) for four stations on the river Al Gharraf (Regulator II, Regulator III, Regulator IIII and Al Badaa). The number of hidden neurons is estimated according to trial and error procedure. The best model is selected according to based root mean square error (RMSE), mean absolute error (MAE) and coefficient of correlation (R). The results showed the optimum numbers of neuron in hidden layer is equal to 10 and indicate that the ANNs is effective technique for forecasting the river discharge, which are utmost essential to hydrologists around the globe