A data-centric bottom-up model for generation of stochastic internal load profiles based on space-use type

R. M. Ward, R. Choudhary, Yeonsook Heo, J. A.D. Aston

Research output: Contribution to journalArticle

Abstract

There is currently no established methodology for the generation of synthetic stochastic internal load profiles for input into building energy simulation. In this paper, a Functional Data Analysis approach is used to propose a new data-centric bottom-up model of plug loads based on hourly data monitored at a high spatial resolution and by space-use type for a case-study building. The model comprises a set of fundamental Principal Components (PCs) that describe the structure of all data samples in terms of amplitude and phase. Scores (or weightings) for each daily demand profile express the contribution of each PC to the demand. Together the principal components and the scores constitute a structure-based model potentially applicable beyond the building considered. The results show good agreement between samples generated using the model and monitored data for key parameters of interest including the timing of the daily peak demand.

Original languageEnglish
JournalJournal of Building Performance Simulation
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Bottom-up
Principal Components
Internal
Functional Data Analysis
Spatial Resolution
Model
Weighting
Timing
Express
High Resolution
Profile
Methodology
Energy
Demand
Simulation

Keywords

  • Functional Data Analysis
  • plug loads
  • Principal Components
  • stochastic

ASJC Scopus subject areas

  • Architecture
  • Building and Construction
  • Modelling and Simulation
  • Computer Science Applications

Cite this

A data-centric bottom-up model for generation of stochastic internal load profiles based on space-use type. / Ward, R. M.; Choudhary, R.; Heo, Yeonsook; Aston, J. A.D.

In: Journal of Building Performance Simulation, 01.01.2019.

Research output: Contribution to journalArticle

@article{84483c1fbdb44cf2b3be72faecb43ac5,
title = "A data-centric bottom-up model for generation of stochastic internal load profiles based on space-use type",
abstract = "There is currently no established methodology for the generation of synthetic stochastic internal load profiles for input into building energy simulation. In this paper, a Functional Data Analysis approach is used to propose a new data-centric bottom-up model of plug loads based on hourly data monitored at a high spatial resolution and by space-use type for a case-study building. The model comprises a set of fundamental Principal Components (PCs) that describe the structure of all data samples in terms of amplitude and phase. Scores (or weightings) for each daily demand profile express the contribution of each PC to the demand. Together the principal components and the scores constitute a structure-based model potentially applicable beyond the building considered. The results show good agreement between samples generated using the model and monitored data for key parameters of interest including the timing of the daily peak demand.",
keywords = "Functional Data Analysis, plug loads, Principal Components, stochastic",
author = "Ward, {R. M.} and R. Choudhary and Yeonsook Heo and Aston, {J. A.D.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/19401493.2019.1583287",
language = "English",
journal = "Journal of Building Performance Simulation",
issn = "1940-1493",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - A data-centric bottom-up model for generation of stochastic internal load profiles based on space-use type

AU - Ward, R. M.

AU - Choudhary, R.

AU - Heo, Yeonsook

AU - Aston, J. A.D.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - There is currently no established methodology for the generation of synthetic stochastic internal load profiles for input into building energy simulation. In this paper, a Functional Data Analysis approach is used to propose a new data-centric bottom-up model of plug loads based on hourly data monitored at a high spatial resolution and by space-use type for a case-study building. The model comprises a set of fundamental Principal Components (PCs) that describe the structure of all data samples in terms of amplitude and phase. Scores (or weightings) for each daily demand profile express the contribution of each PC to the demand. Together the principal components and the scores constitute a structure-based model potentially applicable beyond the building considered. The results show good agreement between samples generated using the model and monitored data for key parameters of interest including the timing of the daily peak demand.

AB - There is currently no established methodology for the generation of synthetic stochastic internal load profiles for input into building energy simulation. In this paper, a Functional Data Analysis approach is used to propose a new data-centric bottom-up model of plug loads based on hourly data monitored at a high spatial resolution and by space-use type for a case-study building. The model comprises a set of fundamental Principal Components (PCs) that describe the structure of all data samples in terms of amplitude and phase. Scores (or weightings) for each daily demand profile express the contribution of each PC to the demand. Together the principal components and the scores constitute a structure-based model potentially applicable beyond the building considered. The results show good agreement between samples generated using the model and monitored data for key parameters of interest including the timing of the daily peak demand.

KW - Functional Data Analysis

KW - plug loads

KW - Principal Components

KW - stochastic

UR - http://www.scopus.com/inward/record.url?scp=85062470986&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062470986&partnerID=8YFLogxK

U2 - 10.1080/19401493.2019.1583287

DO - 10.1080/19401493.2019.1583287

M3 - Article

AN - SCOPUS:85062470986

JO - Journal of Building Performance Simulation

JF - Journal of Building Performance Simulation

SN - 1940-1493

ER -