Reference : Towards a Plug-and-Play and Holistic Data Mining Framework for Understanding and Faci...

	Document type :	Reports : Internal report
	Discipline(s) :	Engineering, computing & technology : Computer science
	Focus Areas :	Computational Sciences; Sustainable Development
	To cite this reference:	http://hdl.handle.net/10993/32961

Title :	Towards a Plug-and-Play and Holistic Data Mining Framework for Understanding and Facilitating Operations in Smart Buildings
Language :	English
Author, co-author :	Li, Daoyuan [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > >]
	Bissyande, Tegawendé François D Assise [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > >]
	Klein, Jacques [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) > Computer Science and Communications Research Unit (CSC) >]
	Le Traon, Yves [University of Luxembourg > Faculty of Science, Technology and Communication (FSTC) > Computer Science and Communications Research Unit (CSC) >]
	Schummer, Paul []
	Muller, Ben []
	Solvi, Anne-Marie []
Publication date :	18-Oct-2017
Publisher :	SnT
ISBN :	978-99959-58-01-5
Report number :	TR-SNT-2017-5
Keywords :	[en] Data mining for smart buildings ; time series mining ; outlier detection
Abstract :	[en] Nowadays, a significant portion of the total energy consumption is attributed to the buildings sector. In order to save energy and protect the environment, energy consumption in buildings must be more efficient. At the same time, buildings should offer the same (if not more) comfort to their occupants. Consequently, modern buildings have been equipped with various sensors and actuators and interconnected control systems to meet occupants’ requirements. Unfortunately, so far, Building Automation Systems data have not been well-exploited due to technical and cost limitations. Yet, it can be exceptionally beneficial to take full advantage of the data flowing inside buildings in order to diagnose issues, explore solutions and improve occupant-building interactions. This paper presents a plug-and-play and holistic data mining framework named PHoliData for smart buildings to collect, store, visualize and mine useful information and domain knowledge from data in smart buildings. PHoliData allows non technical experts to easily explore and understand their buildings with minimum IT support. An architecture of this framework has been introduced and a prototype has been implemented and tested against real-world settings. Discussions with industry experts have suggested the system to be extremely helpful for understanding buildings, since it can provide hints about energy efficiency improvements. Finally, extensive experiments have demonstrated the feasibility of such a framework in practice and its advantage and potential for buildings operators.
Target :	Researchers ; Professionals ; Students ; General public
Permalink :	http://hdl.handle.net/10993/32961

	File(s) associated to this reference
	Fulltext file(s): File Commentary Version Size Access Open access pholidata_paper.pdf Author preprint 2.02 MB View/Open

All documents in ORBi^lu are protected by a user license.

 

University of Luxembourg Library |  Feedback |  Legal notices Site Map