ニューラルネットワークによる多地点時系列データを用いた風速予測におけるVARによるデータ選択の検討

抄録

The output of the wind power depends on natural phenomena, so the influence on the connection for power system is of particular concern. Therefore, if wind power generation can be predicted, effective use of wind energy can be expected. This study proposes the method for wind speed prediction with neural network, NN, and VAR model using the time-series wind speed data of multiple neighboring sites. Also the study evaluates the wind speed prediction performance with wind data of multiple sites including meteorological towers in Tohoku district.