Deep attention ConvLSTM-based adaptive fusion of clear-sky physical prior knowledge and multivariable historical information for probabilistic prediction of photovoltaic power

作者：

Highlights：

• Deep attention ConvLSTM is first proposed for photovoltaic (PV) power forecast.

• Attention mechanism is first introduced into PV power forecast.

• Attention adaptatively fuses clear-sky prior knowledge and historical information.

• Significantly improvement than conventional PV power forecast methods is achieved.

摘要

•Deep attention ConvLSTM is first proposed for photovoltaic (PV) power forecast.•Attention mechanism is first introduced into PV power forecast.•Attention adaptatively fuses clear-sky prior knowledge and historical information.•Significantly improvement than conventional PV power forecast methods is achieved.

论文关键词：deep attention ConvLSTM network,Deep learning,Photovoltaic power forecast,Solar energy,Kernel density estimation,Renewable energy

论文评审过程：Received 30 January 2022, Revised 29 March 2022, Accepted 25 April 2022, Available online 28 April 2022, Version of Record 4 May 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117335