Multi-resolution, multi-horizon distributed solar PV power forecasting with forecast combinations
作者：
Highlights：
• Five solar forecasting models were analysed across multiple resolutions and horizons.
• No model is consistently the best, and a forecast combination is necessary.
• A forecast combination approach based on particle swarm optimization is proposed.
• This approach can be adapted to any resolution or horizon.
• In an extensive evaluation, it outperforms individual models and other combinations.
摘要
•Five solar forecasting models were analysed across multiple resolutions and horizons.•No model is consistently the best, and a forecast combination is necessary.•A forecast combination approach based on particle swarm optimization is proposed.•This approach can be adapted to any resolution or horizon.•In an extensive evaluation, it outperforms individual models and other combinations.
论文关键词：Solar photovoltaic power forecasting,Forecast combinations,Ensembles,Particle swarm optimization,Multi-resolution,Multi-horizon
论文评审过程：Received 2 March 2022, Revised 30 April 2022, Accepted 28 May 2022, Available online 3 June 2022, Version of Record 14 June 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117690
