Multi-step-ahead monthly streamflow forecasting by a neurofuzzy network model

Analysis and forecast of seasonal streamflow series are of utmost importance in the operation planning of water resources systems. One of the greatest difficulties in forecasting of those series is the seasonality nature of streamflow series due to wet and dry periods of the year. Moreover, the real world data are noisy, and may contain contradictions and imperfections. Tolerance for imprecision and uncertainty is also required to achieve tractability and robustness. Fuzzy sets based data analysis models have been especially suitable for these purposes. This suggests the application of neurofuzzy network models to seasonal streamflow forecasting. In this work, a class of neurofuzzy network is applied to the problem of seasonal streamflow forecasting. This model is based on a constructive, competition learning method where neurons groups compete when the network receives a new input, so that it learns the essential parameters to model a fuzzy system, which are the fuzzy rules and membership functions. Database of average monthly inflows from Brazilian hydroelectric plant was used. The performance of the model was compared with conventional approaches and the results show that the model proposed here provide a better performance than the others methodology considering one-step-ahead forecasting and multi-step-ahead forecasting.