Calibrating numerical model by neural networks: A case study for the simulation of the indoor temperature of a building

Xiaoshu Lü; Tao Lu; Martti Viljanen

doi:10.1016/j.egypro.2015.07.215

Calibrating numerical model by neural networks: A case study for the simulation of the indoor temperature of a building

Xiaoshu Lü, Tao Lu, Martti Viljanen

Department of Civil Engineering

Research output: Contribution to journal › Article › Scientific › peer-review

8 Citations (Scopus)

153 Downloads (Pure)

Abstract

This paper proposes a method using neural networks to calibrate numerical models. The approach passes the output of numerical model to a neural network for calibration. An experimental study was conducted using a simulation of unheated and uncooled indoor temperature of a sports hall. The proposed neural network-based model improves the results and produces more accurate calibrated indoor temperature. Furthermore, the developed calibration method requires only measurements of indoor temperatures as the necessary inputs, thus significantly simplifying the calibration procedure needed to model the building performances.

Original language	English
Pages (from-to)	1366-1372
Journal	Energy Procedia
Volume	75
Issue number	August
DOIs	https://doi.org/10.1016/j.egypro.2015.07.215
Publication status	Published - 2015
MoE publication type	A1 Journal article-refereed
Event	International Conference on Applied Energy - Abu Dhabi, United Arab Emirates Duration: 28 Mar 2015 → 31 Mar 2015 Conference number: 7

Keywords

Numerical model
Neural networks
Model calibration
Generalization
Unheated and uncooled indoor temperature simulation

Access to Document

10.1016/j.egypro.2015.07.215

1-s2.0-S1876610215009832-mainFinal published version, 605 KBLicence: CC BY-NC-ND

Cite this

@article{d6b986631c4044a8b0ab2cbff4fe759f,

title = "Calibrating numerical model by neural networks: A case study for the simulation of the indoor temperature of a building",

abstract = "This paper proposes a method using neural networks to calibrate numerical models. The approach passes the output of numerical model to a neural network for calibration. An experimental study was conducted using a simulation of unheated and uncooled indoor temperature of a sports hall. The proposed neural network-based model improves the results and produces more accurate calibrated indoor temperature. Furthermore, the developed calibration method requires only measurements of indoor temperatures as the necessary inputs, thus significantly simplifying the calibration procedure needed to model the building performances.",

keywords = "Numerical model, Neural networks, Model calibration, Generalization, Unheated and uncooled indoor temperature simulation ",

author = "Xiaoshu L{\"u} and Tao Lu and Martti Viljanen",

year = "2015",

doi = "10.1016/j.egypro.2015.07.215",

language = "English",

volume = "75",

pages = "1366--1372",

journal = "Energy Procedia",

issn = "1876-6102",

publisher = "Elsevier",

number = "August",

note = "International Conference on Applied Energy , ICAE ; Conference date: 28-03-2015 Through 31-03-2015",

}

TY - JOUR

T1 - Calibrating numerical model by neural networks: A case study for the simulation of the indoor temperature of a building

AU - Lü, Xiaoshu

AU - Lu, Tao

AU - Viljanen, Martti

N1 - Conference code: 7

PY - 2015

Y1 - 2015

N2 - This paper proposes a method using neural networks to calibrate numerical models. The approach passes the output of numerical model to a neural network for calibration. An experimental study was conducted using a simulation of unheated and uncooled indoor temperature of a sports hall. The proposed neural network-based model improves the results and produces more accurate calibrated indoor temperature. Furthermore, the developed calibration method requires only measurements of indoor temperatures as the necessary inputs, thus significantly simplifying the calibration procedure needed to model the building performances.

AB - This paper proposes a method using neural networks to calibrate numerical models. The approach passes the output of numerical model to a neural network for calibration. An experimental study was conducted using a simulation of unheated and uncooled indoor temperature of a sports hall. The proposed neural network-based model improves the results and produces more accurate calibrated indoor temperature. Furthermore, the developed calibration method requires only measurements of indoor temperatures as the necessary inputs, thus significantly simplifying the calibration procedure needed to model the building performances.

KW - Numerical model

KW - Neural networks

KW - Model calibration

KW - Generalization

KW - Unheated and uncooled indoor temperature simulation

UR - http://dx.doi.org/10.1016/j.egypro.2015.07.215

U2 - 10.1016/j.egypro.2015.07.215

DO - 10.1016/j.egypro.2015.07.215

M3 - Article

SN - 1876-6102

VL - 75

SP - 1366

EP - 1372

JO - Energy Procedia

JF - Energy Procedia

IS - August

T2 - International Conference on Applied Energy

Y2 - 28 March 2015 through 31 March 2015

ER -

Calibrating numerical model by neural networks: A case study for the simulation of the indoor temperature of a building

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this