A deep learning based hybrid method for hourly solar radiation forecasting

作者：

Highlights：

• Proposed a deep learning clustering method for solar irradiance feature learning.

• Hourly solar forecasting with Feature Attention based Deep Forecasting (FADF).

• RMSE reduced by 11.88%–12.65% as compared to smart persistence.

摘要

•Proposed a deep learning clustering method for solar irradiance feature learning.•Hourly solar forecasting with Feature Attention based Deep Forecasting (FADF).•RMSE reduced by 11.88%–12.65% as compared to smart persistence.

论文关键词：Solar forecasting,Deep learning,Clustering,Feature attention

论文评审过程：Received 24 December 2020, Revised 28 February 2021, Accepted 22 March 2021, Available online 26 March 2021, Version of Record 21 April 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.114941