Wind speed and the related electrical power of wind turbines are forecasted. The work is focused on the operation of power systems with integrated wind parks. Artificial neural networks models are proposed for forecasting average values of the following 10 min or Ih. Input quantities for the prediction are wind speeds and their derivatives. Also,,spatial correlation of wind speeds and its use for forecasting, are investigated. The methods are tested using data collected over seven years at six different sites on islands of the South and Central Aegean Sea in Greece.