Abstract

This article focuses on the modeling of nonlinear interactions between the design and operational variables of a system and the multivariate outside environment in predicting the system’s performance. We propose a Sparse Partitioned-Regression (SPR) model that automatically searches for a partition of the environmental variables and fits a sparse regression within each subdivision of the partition, in order to fulfill an optimal criterion. Two optimal criteria are proposed, a penalized and a held-out criterion. We study the theoretical properties of SPR by deriving oracle inequalities to quantify the risks of the penalized and held-out criteria in both prediction and classification problems. An efficient recursive partition algorithm is developed for model estimation. Extensive simulation experiments are conducted to demonstrate the better performance of SPR compared with competing methods. Finally, we present an application of using building design and operational variables, outdoor environmental variables, and their interactions to predict energy consumption based on the Department of Energy’s Energy Plus data sets. SPR produces a high level of prediction accuracy. The result of the application also provides insights into the design, operation, and management of energy-efficient buildings.

Original languageEnglish (US)
Pages (from-to)814-826
Number of pages13
JournalIISE Transactions
Volume49
Issue number8
DOIs
StatePublished - 2017

Fingerprint

Nonlinear systems
Energy utilization
Experiments

Keywords

  • Building energy management
  • Classification
  • Regression
  • Sparse model

ASJC Scopus subject areas

  • Industrial and Manufacturing Engineering

Cite this

A sparse partitioned-regression model for nonlinear system-environment interactions. / Ning, Shuluo; Byon, Eunshin; Wu, Teresa; Li, Jing.

In: IISE Transactions, Vol. 49, No. 8, 2017, p. 814-826.

Research output: Contribution to journalArticle

@article{de48990b2e284dd9b3f37dc55a68ab84,
title = "A sparse partitioned-regression model for nonlinear system-environment interactions",
abstract = "This article focuses on the modeling of nonlinear interactions between the design and operational variables of a system and the multivariate outside environment in predicting the system’s performance. We propose a Sparse Partitioned-Regression (SPR) model that automatically searches for a partition of the environmental variables and fits a sparse regression within each subdivision of the partition, in order to fulfill an optimal criterion. Two optimal criteria are proposed, a penalized and a held-out criterion. We study the theoretical properties of SPR by deriving oracle inequalities to quantify the risks of the penalized and held-out criteria in both prediction and classification problems. An efficient recursive partition algorithm is developed for model estimation. Extensive simulation experiments are conducted to demonstrate the better performance of SPR compared with competing methods. Finally, we present an application of using building design and operational variables, outdoor environmental variables, and their interactions to predict energy consumption based on the Department of Energy’s Energy Plus data sets. SPR produces a high level of prediction accuracy. The result of the application also provides insights into the design, operation, and management of energy-efficient buildings.",
keywords = "Building energy management, Classification, Regression, Sparse model",
author = "Shuluo Ning and Eunshin Byon and Teresa Wu and Jing Li",
year = "2017",
doi = "10.1080/24725854.2017.1299955",
language = "English (US)",
volume = "49",
pages = "814--826",
journal = "IISE Transactions",
issn = "2472-5854",
publisher = "Taylor and Francis Ltd.",
number = "8",

}

TY - JOUR

T1 - A sparse partitioned-regression model for nonlinear system-environment interactions

AU - Ning, Shuluo

AU - Byon, Eunshin

AU - Wu, Teresa

AU - Li, Jing

PY - 2017

Y1 - 2017

N2 - This article focuses on the modeling of nonlinear interactions between the design and operational variables of a system and the multivariate outside environment in predicting the system’s performance. We propose a Sparse Partitioned-Regression (SPR) model that automatically searches for a partition of the environmental variables and fits a sparse regression within each subdivision of the partition, in order to fulfill an optimal criterion. Two optimal criteria are proposed, a penalized and a held-out criterion. We study the theoretical properties of SPR by deriving oracle inequalities to quantify the risks of the penalized and held-out criteria in both prediction and classification problems. An efficient recursive partition algorithm is developed for model estimation. Extensive simulation experiments are conducted to demonstrate the better performance of SPR compared with competing methods. Finally, we present an application of using building design and operational variables, outdoor environmental variables, and their interactions to predict energy consumption based on the Department of Energy’s Energy Plus data sets. SPR produces a high level of prediction accuracy. The result of the application also provides insights into the design, operation, and management of energy-efficient buildings.

AB - This article focuses on the modeling of nonlinear interactions between the design and operational variables of a system and the multivariate outside environment in predicting the system’s performance. We propose a Sparse Partitioned-Regression (SPR) model that automatically searches for a partition of the environmental variables and fits a sparse regression within each subdivision of the partition, in order to fulfill an optimal criterion. Two optimal criteria are proposed, a penalized and a held-out criterion. We study the theoretical properties of SPR by deriving oracle inequalities to quantify the risks of the penalized and held-out criteria in both prediction and classification problems. An efficient recursive partition algorithm is developed for model estimation. Extensive simulation experiments are conducted to demonstrate the better performance of SPR compared with competing methods. Finally, we present an application of using building design and operational variables, outdoor environmental variables, and their interactions to predict energy consumption based on the Department of Energy’s Energy Plus data sets. SPR produces a high level of prediction accuracy. The result of the application also provides insights into the design, operation, and management of energy-efficient buildings.

KW - Building energy management

KW - Classification

KW - Regression

KW - Sparse model

UR - http://www.scopus.com/inward/record.url?scp=85025159307&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85025159307&partnerID=8YFLogxK

U2 - 10.1080/24725854.2017.1299955

DO - 10.1080/24725854.2017.1299955

M3 - Article

AN - SCOPUS:85025159307

VL - 49

SP - 814

EP - 826

JO - IISE Transactions

JF - IISE Transactions

SN - 2472-5854

IS - 8

ER -