Analysis of building performance data

Simple item page
dc.check.embargoformatEmbargo not applicable (If you have not submitted an e-thesis or do not want to request an embargo)en
dc.check.infoNot applicableen
dc.check.opt-outNot applicableen
dc.check.reasonNot applicableen
dc.check.typeNo Embargo Required
dc.contributor.advisorMenzel, Karstenen
dc.contributor.advisorBrown, Kenen
dc.contributor.authorHoerster, Stephan Carlo
dc.contributor.funderHigher Education Authorityen
dc.contributor.funderBilfinger HSG FMen
dc.date.accessioned2018-07-31T10:53:04Z
dc.date.available2018-07-31T10:53:04Z
dc.date.issued2018
dc.date.submitted2018
dc.description.abstractIn recent years, the global trend for digitalisation has also reached buildings and facility management. Due to the roll out of smart meters and the retrofitting of buildings with meters and sensors, the amount of data available for a single building has increased significantly. In addition to data sets collected by measurement devices, Building Information Modelling has recently seen a strong incline. By maintaining a building model through the whole building life-cycle, the model becomes rich of information describing all major aspects of a building. This work aims to combine these data sources to gain further valuable information from data analysis. Better knowledge of the building’s behaviour due to high quality data available leads to more efficient building operations. Eventually, this may result in a reduction of energy use and therefore less operational costs. In this thesis a concept for holistic data acquisition from smart meters and a methodology for the integration of further meters in the measurement concept are introduced and validated. Secondly, this thesis presents a novel algorithm designed for cleansing and interpolation of faulty data. Descriptive data is extracted from an open meta data model for buildings which is utilised to further enrich the metered data. Additionally, this thesis presents a methodology for how to design and manage all information in a unified Data Warehouse schema. This Data Warehouse, which has been developed, maintains compatibility with an open meta data model by adopting the model’s specification into its data schema. It features the application of building specific Key Performance Indicators (KPI) to measure building performance. In addition a clustering algorithm, based on machine learning technology, is developed to identify behavioural patterns of buildings and their frequency of occurrence. All methodologies introduced in this work are evaluated through installations and data from three pilot buildings. The pilot buildings were selected to be of diverse types to prove the generic applicability of the above concepts. The outcome of this work successfully demonstrates that the combination of data sources available for buildings enable advanced data analysis. This largely increases the understanding of buildings and their behavioural patterns. A more efficient building operation and a reduction of energy usage can be achieved with this knowledge.en
dc.description.sponsorshipPRTLI-5en
dc.description.statusNot peer revieweden
dc.description.versionAccepted Version
dc.format.mimetypeapplication/pdfen
dc.identifier.citationHoerster, S. C. 2018. Analysis of building performance data. PhD Thesis, University College Cork.en
dc.identifier.endpage200en
dc.identifier.urihttps://hdl.handle.net/10468/6555
dc.language.isoenen
dc.publisherUniversity College Corken
dc.relation.projectHigher Education Authority (PRTLI-5 program)en
dc.rights© 2018, Stephan Carlo Hoerster.en
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/en
dc.subjectBIMen
dc.subjectIFCen
dc.subjectClusteringen
dc.subjectMeteringen
dc.subjectMonitoringen
dc.subjectData warehouseen
dc.thesis.opt-outfalse
dc.titleAnalysis of building performance dataen
dc.typeDoctoral thesisen
dc.type.qualificationlevelDoctoralen
dc.type.qualificationnamePhDen
ucc.workflow.supervisork.menzel@ucc.ie
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2018-02-03_PhD_Thesis_FINAL_READYTOPRINT-sho.pdf
Size:
4.21 MB
Format:
Adobe Portable Document Format
Description:
Full Text E-thesis
Download
License bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
5.62 KB
Format:
Item-specific license agreed upon to submission
Description:
Download
Collections
Doctoral Theses
Civil and Environmental Engineering - Doctoral Theses
College of Science, Engineering and Food Science - Doctoral Theses