Binding data mining and expert knowledge for one-day-ahead prediction of hourly global solar radiation
作者：
Highlights：
• New data-driven models to predict one-day-ahead hourly solar radiation.
• 2-phases hybrid model using clustering, regression and classification methods.
• 14 different types of day have been identified; and then validated by experts.
• New proposed methodology in the context of eXplainable Artificial Intelligence.
• Relevant information can be extracted by experts from induced models.
摘要
•New data-driven models to predict one-day-ahead hourly solar radiation.•2-phases hybrid model using clustering, regression and classification methods.•14 different types of day have been identified; and then validated by experts.•New proposed methodology in the context of eXplainable Artificial Intelligence.•Relevant information can be extracted by experts from induced models.
论文关键词：Data mining,One-day-ahead prediction,Hourly global solar radiation,Expert systems
论文评审过程：Received 24 June 2020, Revised 17 October 2020, Accepted 20 October 2020, Available online 27 October 2020, Version of Record 10 February 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.114147
