A relevant data selection method for energy consumption prediction of low energy building based on support vector machine

详细信息    查看全文

• 作者：Subodh Paudel^a ; ^b ; ^c ; Mohamed Elmitri^b ; Sté ; phane Couturier^b ; Phuong H. Nguyen^c ; René ; Kamphuis^c ; ¹ ; Bruno Lacarriè ; re^a ; Olivier Le Corre^a ; ^{Olivier.Lecorre@mines-nantes.fr} ; ^{lecorre@mines-nantes.fr} ; ^{ollecorre@gmail.com}
• 关键词：Building energy consumption ; Prediction ; Low energy building ; Support vector machine ; Online and offline learning
• 刊名：Energy and Buildings
• 出版年：2017
• 出版时间：1 March 2017
• 年：2017
• 卷：138
• 期：Complete
• 页码：240-256
• 全文大小：3556 K
• 卷排序：138

文摘

A novel relevant data selection method is developed which uses small representative day dataset for model building. Comparison of developed method with all data modeling approach which uses all available data for model training. Faster model training CPU-time to update the model parameters by considering dynamic environment of prediction day to day conditions. Applicable for Energy Services Company (ESCOs) or Building Energy Management System (BEMS) for planning and control purposes.