AUTHOR=He HongYing , Fu FangYu , Luo DianSheng TITLE=Multiplex parallel GAT-ALSTM: A novel spatial-temporal learning model for multi-sites wind power collaborative forecasting JOURNAL=Frontiers in Energy Research VOLUME=Volume 10 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.974682 DOI=10.3389/fenrg.2022.974682 ISSN=2296-598X ABSTRACT=In order to improve the accuracy of wind power output forecast and ensure the reliability of power grid, Multiplex Parallel GAT-ALSTM, a spatial-temporal learning model for multi-sites wind power collaborative forecasting is proposed in this paper. Topography is generated by using the geographic information (longitude and latitude) obtained from the wind power generation sites. GAT layer is used to capture the spatial correlation characteristics of multi-sites wind power. Feature dimension enhancement of each wind power generation site is achieved by aggregating the information of the adjacent sites. ALSTM layer is used to capture the temporal correlation of each power output time series. The multiplex parallel structure of the model is designed to satisfied fast prediction of large-scale distributed wind power generation. The validity of the proposed Multiplex Parallel GAT-ALSTM is confirmed by comparing with the forecast results obtained by RNN, LSTM, ALSTM and GNN-ALSTM. The testing results show that compared with the RNN, LSTM, ALSTM and GNN-ALSTM, the forecast results of the Multiplex Parallel GAT-ALSTM have lowest mean absolute value error and highest accuracy.