Predictive Control of Hydronic Floor Heating Systems using Neural Networks and Genetic Algorithms

Kasper Vinther, Torben Green, Søren Østergaard, Jan Dimon Bendtsen

Research output: Contribution to journalConference article in JournalResearchpeer-review

2 Citations (Scopus)
97 Downloads (Pure)

Abstract

This paper presents the use a neural network and a micro genetic algorithm to
optimize future set-points in existing hydronic floor heating systems for improved energy efficiency. The neural network can be trained to predict the impact of changes in set-points on future room temperatures. Additionally, weather disturbances such as solar heat gain can be anticipated and compensated for, while taking into account the slow dynamics of the floor. Together with a genetic algorithm, they provide a way to search for optimal future set-point sequences, when convexity and continuity in the solution space is not guaranteed. Evaluation of the performance of multiple neural networks is performed, using different levels of information, and optimization results are presented on a detailed house simulation model.
Original languageEnglish
Book seriesIFAC-PapersOnLine
Volume50
Issue number1
Pages (from-to)7381-7388
Number of pages8
ISSN2405-8963
DOIs
Publication statusPublished - Jul 2017
Event2017 IFAC Congress -
Duration: 9 Jul 201714 Jul 2017
https://www.ifac2017.org/

Conference

Conference2017 IFAC Congress
Period09/07/201714/07/2017
Internet address

Fingerprint

Genetic algorithms
Neural networks
Heating
Energy efficiency
Temperature
Hot Temperature

Keywords

  • Modeling for control optimization
  • Evolutionary algorithms
  • Nonlinear predictive control

Cite this

Vinther, Kasper ; Green, Torben ; Østergaard, Søren ; Bendtsen, Jan Dimon. / Predictive Control of Hydronic Floor Heating Systems using Neural Networks and Genetic Algorithms. In: IFAC-PapersOnLine. 2017 ; Vol. 50, No. 1. pp. 7381-7388.
@inproceedings{cf788e475c2c48149e82167ad7e13261,
title = "Predictive Control of Hydronic Floor Heating Systems using Neural Networks and Genetic Algorithms",
abstract = "This paper presents the use a neural network and a micro genetic algorithm tooptimize future set-points in existing hydronic floor heating systems for improved energy efficiency. The neural network can be trained to predict the impact of changes in set-points on future room temperatures. Additionally, weather disturbances such as solar heat gain can be anticipated and compensated for, while taking into account the slow dynamics of the floor. Together with a genetic algorithm, they provide a way to search for optimal future set-point sequences, when convexity and continuity in the solution space is not guaranteed. Evaluation of the performance of multiple neural networks is performed, using different levels of information, and optimization results are presented on a detailed house simulation model.",
keywords = "Modeling for control optimization, Evolutionary algorithms, Nonlinear predictive control",
author = "Kasper Vinther and Torben Green and S{\o}ren {\O}stergaard and Bendtsen, {Jan Dimon}",
year = "2017",
month = "7",
doi = "10.1016/j.ifacol.2017.08.1477",
language = "English",
volume = "50",
pages = "7381--7388",
journal = "I F A C Workshop Series",
issn = "1474-6670",
publisher = "Elsevier",
number = "1",

}

Predictive Control of Hydronic Floor Heating Systems using Neural Networks and Genetic Algorithms. / Vinther, Kasper; Green, Torben; Østergaard, Søren; Bendtsen, Jan Dimon.

In: IFAC-PapersOnLine, Vol. 50, No. 1, 07.2017, p. 7381-7388.

Research output: Contribution to journalConference article in JournalResearchpeer-review

TY - GEN

T1 - Predictive Control of Hydronic Floor Heating Systems using Neural Networks and Genetic Algorithms

AU - Vinther, Kasper

AU - Green, Torben

AU - Østergaard, Søren

AU - Bendtsen, Jan Dimon

PY - 2017/7

Y1 - 2017/7

N2 - This paper presents the use a neural network and a micro genetic algorithm tooptimize future set-points in existing hydronic floor heating systems for improved energy efficiency. The neural network can be trained to predict the impact of changes in set-points on future room temperatures. Additionally, weather disturbances such as solar heat gain can be anticipated and compensated for, while taking into account the slow dynamics of the floor. Together with a genetic algorithm, they provide a way to search for optimal future set-point sequences, when convexity and continuity in the solution space is not guaranteed. Evaluation of the performance of multiple neural networks is performed, using different levels of information, and optimization results are presented on a detailed house simulation model.

AB - This paper presents the use a neural network and a micro genetic algorithm tooptimize future set-points in existing hydronic floor heating systems for improved energy efficiency. The neural network can be trained to predict the impact of changes in set-points on future room temperatures. Additionally, weather disturbances such as solar heat gain can be anticipated and compensated for, while taking into account the slow dynamics of the floor. Together with a genetic algorithm, they provide a way to search for optimal future set-point sequences, when convexity and continuity in the solution space is not guaranteed. Evaluation of the performance of multiple neural networks is performed, using different levels of information, and optimization results are presented on a detailed house simulation model.

KW - Modeling for control optimization

KW - Evolutionary algorithms

KW - Nonlinear predictive control

U2 - 10.1016/j.ifacol.2017.08.1477

DO - 10.1016/j.ifacol.2017.08.1477

M3 - Conference article in Journal

VL - 50

SP - 7381

EP - 7388

JO - I F A C Workshop Series

JF - I F A C Workshop Series

SN - 1474-6670

IS - 1

ER -