Limited water resources and increased water demand in recent decades have caused irreparable damage to the country's water resources. One of the important components in surface water optimization and management is long-term and short-term river flow forecasts. The aim of the present study is to compare the performance of two Bayesian BN network models with probabilistic approach and MLP neural network. Then selecting the best structural model for flow prediction 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 entering the Taleghan Dam from 2006 to 2018 was 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. In the next step, using the hydrological method of Tennant, The environmental discharge was calculated And the probability of these discharges occurring in the registration data and seventeen patterns in the Easyfit software environment was calculated. Then comparing the selected pattern according to the probability of occurrence and the criteria of the index, Nash-Sutcliffe coefficient (NS), root mean square error (RMSE) and mean absolute prediction error (MAPE) was performed. 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 comparing the environmental probabilities of the two models in the top five patterns, the BN model has an acceptable accuracy . The basin was also found to be at environmental risk.
Manuscript profile
