Visible to the public Mobile Crowdsourcing of Data for Fault Detection and Diagnosis in Smart Buildings

TitleMobile Crowdsourcing of Data for Fault Detection and Diagnosis in Smart Buildings
Publication TypeConference Paper
Year of Publication2016
AuthorsLazarova-Molnar, Sanja, Logason, Halldór Þór, Andersen, Peter Grønb\textbackslasha ek, Kj\textbackslasha ergaard, Mikkel Baun
Conference NameProceedings of the International Conference on Research in Adaptive and Convergent Systems
PublisherACM
Conference LocationNew York, NY, USA
ISBN Number978-1-4503-4455-5
KeywordsBuildings, crowdsourcing, Data collection, energy performance, fault detection and diagnosis, Human Behavior, Metrics, multiple fault diagnosis, occupants, pubcrawl, Resiliency
Abstract

Energy use of buildings represents roughly 40% of the overall energy consumption. Most of the national agendas contain goals related to reducing the energy consumption and carbon footprint. Timely and accurate fault detection and diagnosis (FDD) in building management systems (BMS) have the potential to reduce energy consumption cost by approximately 15-30%. Most of the FDD methods are data-based, meaning that their performance is tightly linked to the quality and availability of relevant data. Based on our experience, faults and relevant events data is very sparse and inadequate, mostly because of the lack of will and incentive for those that would need to keep track of faults. In this paper we introduce the idea of using crowdsourcing to support FDD data collection processes, and illustrate our idea through a mobile application that has been implemented for this purpose. Furthermore, we propose a strategy of how to successfully deploy this building occupants' crowdsourcing application.

URLhttp://doi.acm.org/10.1145/2987386.2987416
DOI10.1145/2987386.2987416
Citation Keylazarova-molnar_mobile_2016