Show simple item record

Interactive visualization for NILM in large buildings using non-negative matrix factorization

dc.contributor.authorGarcía Pérez, Diego 
dc.contributor.authorDíaz Blanco, Ignacio 
dc.contributor.authorPérez López, Daniel 
dc.contributor.authorCuadrado Vega, Abel Alberto 
dc.contributor.authorDomínguez González, Manuel
dc.contributor.authorMorán, Antonio
dc.date.accessioned2018-10-15T08:44:55Z
dc.date.available2018-10-15T08:44:55Z
dc.date.issued2018-10
dc.identifier.citationEnergy and Buildings, 176, p. 95-108 (2018); doi:10.1016/j.enbuild.2018.06.058
dc.identifier.issn0378-7788
dc.identifier.urihttp://hdl.handle.net/10651/49010
dc.descriptionArtículo publicado en abierto mediante APC Elsevier Open Accessspa
dc.description.abstractNon-intrusive load monitoring (NILM) techniques have recently attracted much interest, since they allow to obtain latent patterns from power demand data in buildings, revealing useful information to the expert user. Unsupervised methods are specially attractive, since they do not require labeled datasets. Particularly, non-negative matrix factorization (NMF) methods decompose a single power demand measurement over a certain time period into a set of components or “parts” that are sparse, non-negative and sum up the original measured quantity. Such components reveal hidden temporal patterns which may be difficult to interpret in complex systems such as large buildings. We suggest to integrate the knowledge of the user into the analysis in order to recognize the real events inside the electric network behind the learnt patterns. In this paper, we integrate the available domain knowledge of the user by means of a visual analytics web application in which an expert user can interact in a fluid way with the NMF outcome through visual approaches such as barcharts, heatmaps or calendars. Our approach is tested with real electric power demand data from a hospital complex, showing how the interpretation of the decomposition is improved by means of interactive data cube visualizations, in which the user can insightfully relate the NMF components to characteristic demand patterns of the hospital such as those derived from human activity, as well as to inefficient behaviors of the largest systems in the hospitalspa
dc.description.sponsorshipPlan Nacional I+D+i (DPI2015-69891-C2-2-R)spa
dc.format.extentp. 95-108spa
dc.language.isoengspa
dc.publisherElsevierspa
dc.relation.ispartofEnergy and Buildings, 176spa
dc.rightsCC Reconocimiento 4.0 Internacional
dc.rights© 2018 Autores. Publicado por Elsevier B.V.
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectEnergy efficiencyspa
dc.subjectVisual analyticsspa
dc.subjectBuilding energy consumptionspa
dc.subjectNMFspa
dc.titleInteractive visualization for NILM in large buildings using non-negative matrix factorizationspa
dc.typeinfo:eu-repo/semantics/articlespa
dc.identifier.doi10.1016/j.enbuild.2018.06.058
dc.type.dcmitextspa
dc.relation.projectIDMINECO/DPI2015-69891-C2-2-Rspa
dc.relation.publisherversionhttps://doi.org/10.1016/j.enbuild.2018.06.058spa
dc.rights.accessRightsinfo:eu-repo/semantics/openAccessspa


Files in this item

untranslated

This item appears in the following Collection(s)

Show simple item record

CC Reconocimiento 4.0 Internacional
This item is protected with a Creative Commons License