A new approach to wind turbine power generation forecasting, using weather radar data based on Hidden Markov Model

The wind is one of the most important and affecting phenomena and is known as one of the significant clean resources of energy. Apart from other atmospheric parameters, the wind has complex behavior and intermittent characteristics. Local phenomena can be accompanied by the wind, which is strong, non-predicted, and damaging.  Weather radars are capable of detecting and displaying storm-related turbulence as well as precipitation in a relatively wide area. This capability can improve the quality of the wind forecast. In this paper, a method is presented and implemented to forecast the probability of strong wind in the next five hours based on the Hidden Markov Model (HMM). The method is expanded to find out the forecast of wind turbine output power and reliability as well. Achieved results show that about 67% of strong winds are correctly forecasted