Holistic distributed stream clustering for smart grids

• Type: Article in International Conference Proceedings Book
• Date: 2012

Authors

Rodrigues, PP (Author) FMUP; Gama, J (Author) Other

Conference proceedings International

Title: CEUR Workshop Proceedings

Pages: 18-22

Workshop on Ubiquitous Data Mining, UDM 2012 - In Conjunction with the 20th European Conference on Artificial Intelligence, ECAI 2012

27 August 2012 through 31 August 2012

Indexing

ISI Web of Knowledge

Scopus - 4 Citations

Other information

Authenticus ID: P-00A-8QK

Abstract (EN): Smart grids consist of millions of automated electronic meters that will be installed in electricity distribution networks and connected to servers that will manage grid supervision, billing and customer services. World sustainability regarding energy management will definitely rely on such grids, so smart grids need also to be sustainable themselves. This sustainability depends on several research problems that emerge from this new setting (from power balance to energy markets) requiring new approaches for knowledge discovery and decision support. This paper presents a holistic distributed stream clustering view of possible solutions for those problems, supported by previous research in related domains. The approach is based on two orthogonal clustering algorithms, combined for a holistic clustering of the grid. Experimental results are included to illustrate the benefits of each algorithm, while the proposal is discussed in terms of application to smart grid problems. This holistic approach could be used to help solving some of the smart grid intelligent layer research problems, thus improving global sustainability.

Language: English

Type (Professor's evaluation): Scientific