Data mining for energy systems: Review and prospect. (24th March 2021)

• Record Type: Journal Article
• Authors: Liu, Wenxuan; Zhao, Junhua; Wang, Dianhui
• Is Part Of: Wiley interdisciplinary reviews. Volume 11:Number 4(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-03-24
• Subjects: electricity market -- experimental economics -- machine learning; Data mining -- Periodicals; 006.31205
• DOI: 10.1002/widm.1406
• Languages: English
• ISSNs: 1942-4787

Abstract

Abstract: An in‐depth study on big data mining is urgently needed for the next‐generation energy systems, which are characterized by a deep integration of cyber, physical, and social components. This paper presents an initial discussion on big data mining and its applications in intelligent energy systems. New progress in big data mining, such as deep learning, transfer learning, randomized learning, granular computing, and multisource data fusion, is introduced first. Some applications of data mining in energy systems, such as load forecasting and modeling, integrated power and transportation system, and electricity market forecasting and simulation, are discussed then. Moreover, some research problems in energy system data mining, such as cyber–physical–social system modeling and super‐resolution perception for smart meter data, which require further attention in the future, are also discussed. This article is categorized under: Application Areas > Business and Industry Abstract : This paper reviews some machine learning techniques for power big data mining, such as deep learning, transfer learning, randomized learning, granular computing and multi‐source data fusion. Some typical applications, such as load forecasting and modelling, integrated power and transportation system, and electricity market forecasting, are discussed.