A hybrid forecasting model based on date-framework strategy and improved feature selection technology for short-term load forecasting

详细信息    查看全文

• 作者：Ping Jiang ; Feng Liu ; ^{liuf13@lzu.edu.cn} ; ^{feng.liu.1990@ieee.org} ; Yiliao Song
• 关键词：Electricity loads forecasting ; Date-framework strategy ; Feature selection ; Minimum subset of features ; Hybrid model
• 刊名：Energy
• 出版年：2017
• 出版时间：15 January 2017
• 年：2017
• 卷：119
• 期：Complete
• 页码：694-709
• 全文大小：3469 K
• 卷排序：119

文摘

This paper investigates a date-framework strategy to select feature candidates. The improved FS model has ability to select minimum and effect features. MAPE of the proposed model is 0.58%. The proposed model has satisfactory in-sample and out-sample forecasting accuracy. The proposed model outperforms the benchmarks in the whole year.