A Hybrid Wavelet and Neuro-Fuzzy Model for Forecasting the Monthly Streamflow Data

摘要: Researchers have studied to forecast the streamflow in order develop water usage policy. They used not only traditional methods, but also computer aided methods. Some black-box models, like Adaptive Neuro Fuzzy Inference Systems (ANFIS), became very popular for hydrologic engineering, because of their rapidity and less variation requirements. Wavelet Transform has become a useful tool analysis variations time series. In this study, hybrid model, Wavelet-Neuro (WNF), been data 5 Flow Observation Stations (FOS), which belong Sakarya Basin Turkey. evaluate accuracy performance Auto Regressive Integrated Moving Average (ARIMA) model with same sets. The comparison made by Root Mean Squared Errors (RMSE) models. Results showed that WNF forecasts more accurately than ARIMA model.