Next-day wind park electric energy generation forecasting using fuzzy time-series

Resumen

The integration of wind energy conversion systems in the electric power networks has become an important problem for the integral forecast of the short-term demand. Due to the intermittent nature of wind, it is difficult to forecast the electric energy generated from a wind farm for the next hours. In this paper a new model based on fuzzy time series (FTS) is used for the forecast of the electric energy generated from a wind farm in the next 24 hours. Only past values of daily electric energy generation from a real life wind farm and fuzzy real weather predictions are used. The computer results obtained using FTS models have been better than the obtained ones from autoregressive and neural networks models.