Efficiency Comparison of Bayesian and MLP Neural Networks in Predicting Runoff to the Taleghan Dam

The importance of regulating the supply and demand regime shows the need for planning in the exploitation of surface water resources. The aim of the present study is to compare the performance of two Bayesian BN network models with a probabilistic approach and MLP neural network for flow prediction. Then selecting the best structural model is another goal of the present study. Monthly meteorological data including precipitation, monthly average temperature, evaporation, and also the volume of water transferred from five hydrometric stations were introduced as input data to the models and runoff to the dam was considered as predictable. Then, with the aim of estimating the best prediction pattern structure, input data with different layouts were introduced to the models. The results of comparing the selected pattern were performed according to the criteria of the index, Nash-Sutcliffe coefficient (NS), mean square error (MSE), root mean square error (RMSE), and mean absolute prediction error (MAPE). The best model in BN model with 43.3% similarity and index criteria were estimated to be -3.98, 300, 17.3, and 0.06, respectively. MLP model with 80% similarity and index criteria were introduced as -10.3, 8266, 23.9, and 122.3 in the best model, respectively. As a result, both models performed well in runoff estimation, but the accurate BN model is much better at being prefabricated. Finally, a structural pattern with acceptable results was identified in both BN and MLP models.