Gaussian process modeling for measurement and verification of building energy savings

Yeonsook Heo, Victor M. Zavala

Research output: Contribution to journalArticle

83 Citations (Scopus)

Abstract

We present a Gaussian process (GP) modeling framework to determine energy savings and uncertainty levels in measurement and verification (M&V) practices. Existing M&V guidelines provide savings calculation procedures based on linear regression techniques that are limited in their predictive and uncertainty estimation capabilities. We demonstrate that, unlike linear regression, GP models can capture complex nonlinear and multivariable interactions as well as multiresolution trends of energy behavior. In addition, because GP models are developed under a Bayesian setting, they can capture different sources of uncertainty in a more systematic way. We demonstrate that these capabilities can ultimately lead to significantly less expensive M&V practices. We illustrate the developments using simulated and real data settings.

Original languageEnglish
Pages (from-to)7-18
Number of pages12
JournalEnergy and Buildings
Volume53
DOIs
Publication statusPublished - 2012 Oct 1
Externally publishedYes

Fingerprint

Energy conservation
Linear regression
Uncertainty

Keywords

  • Gaussian process modeling
  • Measurement and verification
  • Performance-based contracts
  • Retrofit analysis
  • Uncertainty

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction
  • Mechanical Engineering
  • Electrical and Electronic Engineering

Cite this

Gaussian process modeling for measurement and verification of building energy savings. / Heo, Yeonsook; Zavala, Victor M.

In: Energy and Buildings, Vol. 53, 01.10.2012, p. 7-18.

Research output: Contribution to journalArticle

@article{58bf21dde0fc40bfafde7d20854d60b6,
title = "Gaussian process modeling for measurement and verification of building energy savings",
abstract = "We present a Gaussian process (GP) modeling framework to determine energy savings and uncertainty levels in measurement and verification (M&V) practices. Existing M&V guidelines provide savings calculation procedures based on linear regression techniques that are limited in their predictive and uncertainty estimation capabilities. We demonstrate that, unlike linear regression, GP models can capture complex nonlinear and multivariable interactions as well as multiresolution trends of energy behavior. In addition, because GP models are developed under a Bayesian setting, they can capture different sources of uncertainty in a more systematic way. We demonstrate that these capabilities can ultimately lead to significantly less expensive M&V practices. We illustrate the developments using simulated and real data settings.",
keywords = "Gaussian process modeling, Measurement and verification, Performance-based contracts, Retrofit analysis, Uncertainty",
author = "Yeonsook Heo and Zavala, {Victor M.}",
year = "2012",
month = "10",
day = "1",
doi = "10.1016/j.enbuild.2012.06.024",
language = "English",
volume = "53",
pages = "7--18",
journal = "Energy and Buildings",
issn = "0378-7788",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Gaussian process modeling for measurement and verification of building energy savings

AU - Heo, Yeonsook

AU - Zavala, Victor M.

PY - 2012/10/1

Y1 - 2012/10/1

N2 - We present a Gaussian process (GP) modeling framework to determine energy savings and uncertainty levels in measurement and verification (M&V) practices. Existing M&V guidelines provide savings calculation procedures based on linear regression techniques that are limited in their predictive and uncertainty estimation capabilities. We demonstrate that, unlike linear regression, GP models can capture complex nonlinear and multivariable interactions as well as multiresolution trends of energy behavior. In addition, because GP models are developed under a Bayesian setting, they can capture different sources of uncertainty in a more systematic way. We demonstrate that these capabilities can ultimately lead to significantly less expensive M&V practices. We illustrate the developments using simulated and real data settings.

AB - We present a Gaussian process (GP) modeling framework to determine energy savings and uncertainty levels in measurement and verification (M&V) practices. Existing M&V guidelines provide savings calculation procedures based on linear regression techniques that are limited in their predictive and uncertainty estimation capabilities. We demonstrate that, unlike linear regression, GP models can capture complex nonlinear and multivariable interactions as well as multiresolution trends of energy behavior. In addition, because GP models are developed under a Bayesian setting, they can capture different sources of uncertainty in a more systematic way. We demonstrate that these capabilities can ultimately lead to significantly less expensive M&V practices. We illustrate the developments using simulated and real data settings.

KW - Gaussian process modeling

KW - Measurement and verification

KW - Performance-based contracts

KW - Retrofit analysis

KW - Uncertainty

UR - http://www.scopus.com/inward/record.url?scp=84864212460&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864212460&partnerID=8YFLogxK

U2 - 10.1016/j.enbuild.2012.06.024

DO - 10.1016/j.enbuild.2012.06.024

M3 - Article

VL - 53

SP - 7

EP - 18

JO - Energy and Buildings

JF - Energy and Buildings

SN - 0378-7788

ER -