Forecasting daily river flow of Ahar Chay River using Artificial Neural Networks (ANN) and Comparation with Adaptive Neuro Fuzzy Inference System (ANFIS)

Abdollahpour, Mohammadreza; Sattari, Mohammad

Forecasting daily river flow of Ahar Chay River using Artificial Neural Networks (ANN) and Comparation with Adaptive Neuro Fuzzy Inference System (ANFIS)

Document Type : Complete scientific research article

Authors

U tabriz

Abstract

In recent years application of intelligent methods has been considered in forecasting hydrologic processes. In this research, daily river flow of Ahar Chay, a river located in East-Azerbaijan province at the north-west of Iran, was forecasted using Artificial Neural Networks and Adaptive Neuro Fuzzy Inference System methods in Orang, Bermis and Tazekand stations. For the modeling of the next day flow, daily flow discharge data was used which was collected between 2002 and 2009 years. So that 6-year data set was selected as the training data and the rest as a test data. Determination coefficient (R2( and Root Mean Squared Error (RMSE) statistical criteria were used to evaluate the performance of the obtained results. The results showed that Adaptive Neuro Fuzzy Inference System (ANFIS) method gives a better daily river flow forecasting in Ahar Chay with R2=0.94 and RMSE=0.0318 m3/sec compared to Artificial Neural Networks (ANN) method with R2=0.92 and RMSE=0.0378 m3/sec.

Keywords