Wind Farm Ultra Short Term Power Prediction Considering Attention Mechanism and Historical Data of Wind Farm Internal Units

Accurately predicting the active power output of wind farms is an important means to ensure the stable operation of new power systems. This article proposes a sequence to sequence ultra short term power prediction model for wind farms, combining Luong attention mechanism and Gated Recurrent Unit (GRU). Secondly, the models of wind turbines within the wind farm are used as model inputs to increase the dimensionality of the input data; Then, the Luong attention mechanism is used to perform attention calculations on the temporal features extracted by GRU, eliminating the distance dependence of the model on information at different times; Finally, simulation shows that the proposed model is effective in predicting the ultra short term power of wind farms.