A Hybrid Model based on Deep Learning and Cross-attention for Short-term Wind Power Prediction

New energy is a substitute for traditional energy in the coming decades, but its stability is poor. Power generation forecasting is an effective way to mitigate its negative effects. This paper proposed a wind power generation prediction model algorithm independent of meteorological data and propose a new way to build the cross-attention mechanism and perform a better result. We introduce a three-year wind farm dataset and apply out model to it, which facilitate feasibility analysis.