Identifying the time profile of everyday activities in the home using smart meter data

Charlie Wilson, Lina Stankovic, Vladimir Stankovic, Jing Liao, Michael Coleman, Richard Hauxwell-Baldwin, Tom Kane, Steven Firth, Tarek Hassan

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

Activities are a descriptive term for the common ways households spend their time. Examples include cooking, doing laundry, or socialising. Smart meter data can be used to generate time profiles of activities that are meaningful to households’ own lived experience. Activities are therefore a lens through which energy feedback to households can be made salient and understandable. This paper demonstrates a multi-step methodology for inferring hourly time profiles of ten household activities using smart meter data, supplemented by individual appliance plug monitors and environmental sensors. First, household interviews, video ethnography, and technology surveys are used to identify appliances and devices in the home, and their roles in specific activities. Second, ‘ontologies’ are developed to map out the relationships between activities and technologies in the home. One or more technologies may indicate the occurrence of certain activities. Third, data from smart meters, plug monitors and sensor data are collected. Smart meter data measuring aggregate electricity use are disaggregated and processed together with the plug monitor and sensor data to identify when and for how long different activities are occurring. Sensor data are particularly useful for activities that are not always associated with an energy-using device. Fourth, the ontologies are applied to the disaggregated data to make inferences on hourly time profiles of ten everyday activities. These include washing, doing laundry, watching TV (reliably inferred), and cleaning, socialising, working (inferred with uncertainties). Fifth, activity time diaries and structured interviews are used to validate both the ontologies and the inferred activity time profiles. Two case study homes are used to illustrate the methodology using data collected as part of a UK trial of smart home technologies. The methodology is demonstrated to produce reliable time profiles of a range of domestic activities that are meaningful to households. The methodology also emphasises the value of integrating coded interview and video ethnography data into both the development of the activity inference process.
Original languageEnglish
Title of host publicationEceee Summer Study proceedings
Subtitle of host publicationFirst fuel now
EditorsTherese Laitinen Lindström
Pages933-946
Publication statusPublished - Jun 2015
EventECEEE-2015 - France, Toulon/Hyères, United Kingdom
Duration: 1 Jun 20156 Jun 2015

Conference

ConferenceECEEE-2015
CountryUnited Kingdom
CityToulon/Hyères
Period1/06/156/06/15

Fingerprint

Smart meters
Ontology
Laundries
Sensors
Cooking
Washing
Cleaning
Lenses
Electricity
Feedback

Keywords

  • information anc communication technologies
  • smart metering
  • activity patterns
  • energy data
  • households

Cite this

Wilson, C., Stankovic, L., Stankovic, V., Liao, J., Coleman, M., Hauxwell-Baldwin, R., ... Hassan, T. (2015). Identifying the time profile of everyday activities in the home using smart meter data. In T. L. Lindström (Ed.), Eceee Summer Study proceedings: First fuel now (pp. 933-946). [5-046-15]
Wilson, Charlie ; Stankovic, Lina ; Stankovic, Vladimir ; Liao, Jing ; Coleman, Michael ; Hauxwell-Baldwin, Richard ; Kane, Tom ; Firth, Steven ; Hassan, Tarek. / Identifying the time profile of everyday activities in the home using smart meter data. Eceee Summer Study proceedings: First fuel now. editor / Therese Laitinen Lindström. 2015. pp. 933-946
@inproceedings{309c3dc6f0c345719abda5e92034d8c5,
title = "Identifying the time profile of everyday activities in the home using smart meter data",
abstract = "Activities are a descriptive term for the common ways households spend their time. Examples include cooking, doing laundry, or socialising. Smart meter data can be used to generate time profiles of activities that are meaningful to households’ own lived experience. Activities are therefore a lens through which energy feedback to households can be made salient and understandable. This paper demonstrates a multi-step methodology for inferring hourly time profiles of ten household activities using smart meter data, supplemented by individual appliance plug monitors and environmental sensors. First, household interviews, video ethnography, and technology surveys are used to identify appliances and devices in the home, and their roles in specific activities. Second, ‘ontologies’ are developed to map out the relationships between activities and technologies in the home. One or more technologies may indicate the occurrence of certain activities. Third, data from smart meters, plug monitors and sensor data are collected. Smart meter data measuring aggregate electricity use are disaggregated and processed together with the plug monitor and sensor data to identify when and for how long different activities are occurring. Sensor data are particularly useful for activities that are not always associated with an energy-using device. Fourth, the ontologies are applied to the disaggregated data to make inferences on hourly time profiles of ten everyday activities. These include washing, doing laundry, watching TV (reliably inferred), and cleaning, socialising, working (inferred with uncertainties). Fifth, activity time diaries and structured interviews are used to validate both the ontologies and the inferred activity time profiles. Two case study homes are used to illustrate the methodology using data collected as part of a UK trial of smart home technologies. The methodology is demonstrated to produce reliable time profiles of a range of domestic activities that are meaningful to households. The methodology also emphasises the value of integrating coded interview and video ethnography data into both the development of the activity inference process.",
keywords = "information anc communication technologies, smart metering, activity patterns, energy data, households",
author = "Charlie Wilson and Lina Stankovic and Vladimir Stankovic and Jing Liao and Michael Coleman and Richard Hauxwell-Baldwin and Tom Kane and Steven Firth and Tarek Hassan",
year = "2015",
month = "6",
language = "English",
isbn = "978-91-980482-6-1",
pages = "933--946",
editor = "Lindstr{\"o}m, {Therese Laitinen}",
booktitle = "Eceee Summer Study proceedings",

}

Wilson, C, Stankovic, L, Stankovic, V, Liao, J, Coleman, M, Hauxwell-Baldwin, R, Kane, T, Firth, S & Hassan, T 2015, Identifying the time profile of everyday activities in the home using smart meter data. in TL Lindström (ed.), Eceee Summer Study proceedings: First fuel now., 5-046-15, pp. 933-946, ECEEE-2015, Toulon/Hyères, United Kingdom, 1/06/15.

Identifying the time profile of everyday activities in the home using smart meter data. / Wilson, Charlie; Stankovic, Lina; Stankovic, Vladimir; Liao, Jing; Coleman, Michael; Hauxwell-Baldwin, Richard; Kane, Tom; Firth, Steven; Hassan, Tarek.

Eceee Summer Study proceedings: First fuel now. ed. / Therese Laitinen Lindström. 2015. p. 933-946 5-046-15.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Identifying the time profile of everyday activities in the home using smart meter data

AU - Wilson, Charlie

AU - Stankovic, Lina

AU - Stankovic, Vladimir

AU - Liao, Jing

AU - Coleman, Michael

AU - Hauxwell-Baldwin, Richard

AU - Kane, Tom

AU - Firth, Steven

AU - Hassan, Tarek

PY - 2015/6

Y1 - 2015/6

N2 - Activities are a descriptive term for the common ways households spend their time. Examples include cooking, doing laundry, or socialising. Smart meter data can be used to generate time profiles of activities that are meaningful to households’ own lived experience. Activities are therefore a lens through which energy feedback to households can be made salient and understandable. This paper demonstrates a multi-step methodology for inferring hourly time profiles of ten household activities using smart meter data, supplemented by individual appliance plug monitors and environmental sensors. First, household interviews, video ethnography, and technology surveys are used to identify appliances and devices in the home, and their roles in specific activities. Second, ‘ontologies’ are developed to map out the relationships between activities and technologies in the home. One or more technologies may indicate the occurrence of certain activities. Third, data from smart meters, plug monitors and sensor data are collected. Smart meter data measuring aggregate electricity use are disaggregated and processed together with the plug monitor and sensor data to identify when and for how long different activities are occurring. Sensor data are particularly useful for activities that are not always associated with an energy-using device. Fourth, the ontologies are applied to the disaggregated data to make inferences on hourly time profiles of ten everyday activities. These include washing, doing laundry, watching TV (reliably inferred), and cleaning, socialising, working (inferred with uncertainties). Fifth, activity time diaries and structured interviews are used to validate both the ontologies and the inferred activity time profiles. Two case study homes are used to illustrate the methodology using data collected as part of a UK trial of smart home technologies. The methodology is demonstrated to produce reliable time profiles of a range of domestic activities that are meaningful to households. The methodology also emphasises the value of integrating coded interview and video ethnography data into both the development of the activity inference process.

AB - Activities are a descriptive term for the common ways households spend their time. Examples include cooking, doing laundry, or socialising. Smart meter data can be used to generate time profiles of activities that are meaningful to households’ own lived experience. Activities are therefore a lens through which energy feedback to households can be made salient and understandable. This paper demonstrates a multi-step methodology for inferring hourly time profiles of ten household activities using smart meter data, supplemented by individual appliance plug monitors and environmental sensors. First, household interviews, video ethnography, and technology surveys are used to identify appliances and devices in the home, and their roles in specific activities. Second, ‘ontologies’ are developed to map out the relationships between activities and technologies in the home. One or more technologies may indicate the occurrence of certain activities. Third, data from smart meters, plug monitors and sensor data are collected. Smart meter data measuring aggregate electricity use are disaggregated and processed together with the plug monitor and sensor data to identify when and for how long different activities are occurring. Sensor data are particularly useful for activities that are not always associated with an energy-using device. Fourth, the ontologies are applied to the disaggregated data to make inferences on hourly time profiles of ten everyday activities. These include washing, doing laundry, watching TV (reliably inferred), and cleaning, socialising, working (inferred with uncertainties). Fifth, activity time diaries and structured interviews are used to validate both the ontologies and the inferred activity time profiles. Two case study homes are used to illustrate the methodology using data collected as part of a UK trial of smart home technologies. The methodology is demonstrated to produce reliable time profiles of a range of domestic activities that are meaningful to households. The methodology also emphasises the value of integrating coded interview and video ethnography data into both the development of the activity inference process.

KW - information anc communication technologies

KW - smart metering

KW - activity patterns

KW - energy data

KW - households

UR - http://www.eceee.org/summerstudy

UR - http://proceedings.eceee.org/visabstrakt.php?event=5&doc=5-046-15

M3 - Conference contribution book

SN - 978-91-980482-6-1

SP - 933

EP - 946

BT - Eceee Summer Study proceedings

A2 - Lindström, Therese Laitinen

ER -

Wilson C, Stankovic L, Stankovic V, Liao J, Coleman M, Hauxwell-Baldwin R et al. Identifying the time profile of everyday activities in the home using smart meter data. In Lindström TL, editor, Eceee Summer Study proceedings: First fuel now. 2015. p. 933-946. 5-046-15