Using semantic annotation in building databases to improve information and energy modelling: a use-case of UK domestic time-series data

Gokhan Mevlevioglu, S Natarajan, J A Padget

Research output: Working paper

17 Downloads (Pure)

Abstract

There is a increasing interest in modelling stock-level (i.e. local authority, re- gional or national) energy flows in buildings (both domestic and non-domestic) primarily as a means for technological and economic assessment of carbon abate- ment options. Modelling stock level building energy flows is a complex endeav- our that requires the bringing together of a range of different data-sets (cli- mate data, physical building data, occupant profiles, system profiles etc.) each with its own particular data-structure. Typically, this process can be time- consuming, repetitive and difficult to update. As new data is continually being produced, models can quickly become out-dated. We propose a semantically annotated database via an over-arching ontology that radically simplifies this process providing powerful new techniques to combine data-sets and query them. We demonstrate this technique through building up a full time-series of English Housing Survey (EHS) data (from 1970 onwards) which are not directly com- patible due to changes in survey methodologies over time. We then use the combined data-set to build up a picture of changing SAP levels for new build- ings over this period and plot them against mandated changes to the building regulations. The key demonstration here is the speed and the efficiency of the process rather than the data itself.
Original languageEnglish
Number of pages28
Publication statusUnpublished - 2014

Fingerprint

Time series
Semantics
Ontology
Data structures
Demonstrations
Economics
Carbon

Keywords

  • semantic annotations
  • Time series analysis
  • building energy modelling

Cite this

@techreport{5089829f286a422796c22dd536b27373,
title = "Using semantic annotation in building databases to improve information and energy modelling: a use-case of UK domestic time-series data",
abstract = "There is a increasing interest in modelling stock-level (i.e. local authority, re- gional or national) energy flows in buildings (both domestic and non-domestic) primarily as a means for technological and economic assessment of carbon abate- ment options. Modelling stock level building energy flows is a complex endeav- our that requires the bringing together of a range of different data-sets (cli- mate data, physical building data, occupant profiles, system profiles etc.) each with its own particular data-structure. Typically, this process can be time- consuming, repetitive and difficult to update. As new data is continually being produced, models can quickly become out-dated. We propose a semantically annotated database via an over-arching ontology that radically simplifies this process providing powerful new techniques to combine data-sets and query them. We demonstrate this technique through building up a full time-series of English Housing Survey (EHS) data (from 1970 onwards) which are not directly com- patible due to changes in survey methodologies over time. We then use the combined data-set to build up a picture of changing SAP levels for new build- ings over this period and plot them against mandated changes to the building regulations. The key demonstration here is the speed and the efficiency of the process rather than the data itself.",
keywords = "semantic annotations, Time series analysis, building energy modelling",
author = "Gokhan Mevlevioglu and S Natarajan and Padget, {J A}",
year = "2014",
language = "English",
type = "WorkingPaper",

}

TY - UNPB

T1 - Using semantic annotation in building databases to improve information and energy modelling: a use-case of UK domestic time-series data

AU - Mevlevioglu, Gokhan

AU - Natarajan, S

AU - Padget, J A

PY - 2014

Y1 - 2014

N2 - There is a increasing interest in modelling stock-level (i.e. local authority, re- gional or national) energy flows in buildings (both domestic and non-domestic) primarily as a means for technological and economic assessment of carbon abate- ment options. Modelling stock level building energy flows is a complex endeav- our that requires the bringing together of a range of different data-sets (cli- mate data, physical building data, occupant profiles, system profiles etc.) each with its own particular data-structure. Typically, this process can be time- consuming, repetitive and difficult to update. As new data is continually being produced, models can quickly become out-dated. We propose a semantically annotated database via an over-arching ontology that radically simplifies this process providing powerful new techniques to combine data-sets and query them. We demonstrate this technique through building up a full time-series of English Housing Survey (EHS) data (from 1970 onwards) which are not directly com- patible due to changes in survey methodologies over time. We then use the combined data-set to build up a picture of changing SAP levels for new build- ings over this period and plot them against mandated changes to the building regulations. The key demonstration here is the speed and the efficiency of the process rather than the data itself.

AB - There is a increasing interest in modelling stock-level (i.e. local authority, re- gional or national) energy flows in buildings (both domestic and non-domestic) primarily as a means for technological and economic assessment of carbon abate- ment options. Modelling stock level building energy flows is a complex endeav- our that requires the bringing together of a range of different data-sets (cli- mate data, physical building data, occupant profiles, system profiles etc.) each with its own particular data-structure. Typically, this process can be time- consuming, repetitive and difficult to update. As new data is continually being produced, models can quickly become out-dated. We propose a semantically annotated database via an over-arching ontology that radically simplifies this process providing powerful new techniques to combine data-sets and query them. We demonstrate this technique through building up a full time-series of English Housing Survey (EHS) data (from 1970 onwards) which are not directly com- patible due to changes in survey methodologies over time. We then use the combined data-set to build up a picture of changing SAP levels for new build- ings over this period and plot them against mandated changes to the building regulations. The key demonstration here is the speed and the efficiency of the process rather than the data itself.

KW - semantic annotations

KW - Time series analysis

KW - building energy modelling

M3 - Working paper

BT - Using semantic annotation in building databases to improve information and energy modelling: a use-case of UK domestic time-series data

ER -