Machine learning technique for energy efficiency analysis of HVAC systems
Energy usage in air-conditioning system accounts for approximately 50% of the total energy consumption in commercial building sector in Singapore, presenting a significant opportunity for minimizing the energy consumption improving the energy efficiency of its operation. Machine learning techniques...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
2014
|
Subjects: | |
Online Access: | http://hdl.handle.net/10356/61491 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-61491 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-614912023-07-07T17:30:20Z Machine learning technique for energy efficiency analysis of HVAC systems Ma, Biao Soh Yeng Chai School of Electrical and Electronic Engineering DRNTU::Engineering Energy usage in air-conditioning system accounts for approximately 50% of the total energy consumption in commercial building sector in Singapore, presenting a significant opportunity for minimizing the energy consumption improving the energy efficiency of its operation. Machine learning techniques have been introduced to facilitate the modeling and optimizing of HVAC system as they are able to handle nonlinear dynamic and complex data of HVAC system. Machine learning techniques such as Support Vector Machine (SVM) and Back Propagation (BP) have been developed and grown for decades. They have proven their power in nonlinear and complex systems through successful and effective applications in business, medical and industrial fields. Extreme Learning Machine (ELM) is a new member of machine learning technique community but it has been gaining increasing attention for its outstanding performance in learning speed and adaptability. This project explores the use of machine learning techniques of ELM and SVM to derive at a better analysis of energy consumption and efficiency of large air-conditioning systems. A HVAC system simulation software, DesignBuilder was being used to set up the layout and conditions for simulation as the desired experimental lab located at Lecture Theatre 24 was not ready for use. The data collected from DesignBuilder was analyzed by ELM and SVM to examine the various factors like air temperature and CO2 level and occupancy level’s effect on energy consumption. The most influencing factor on energy consumption would then be identified. In addition, the performances of ELM and SVM in terms of testing Root Mean Square Error (RMSE) and testing time were compared to find out which machine learning technique is more suitable in handling HVAC data. Discoveries made using the machine learning techniques can be captured and organized in a format for later recall and communication to others. Bachelor of Engineering 2014-06-10T09:09:39Z 2014-06-10T09:09:39Z 2014 2014 Final Year Project (FYP) http://hdl.handle.net/10356/61491 en Nanyang Technological University 58 p. application/pdf |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
DRNTU::Engineering |
spellingShingle |
DRNTU::Engineering Ma, Biao Machine learning technique for energy efficiency analysis of HVAC systems |
description |
Energy usage in air-conditioning system accounts for approximately 50% of the total energy consumption in commercial building sector in Singapore, presenting a significant opportunity for minimizing the energy consumption improving the energy efficiency of its operation. Machine learning techniques have been introduced to facilitate the modeling and optimizing of HVAC system as they are able to handle nonlinear dynamic and complex data of HVAC system.
Machine learning techniques such as Support Vector Machine (SVM) and Back Propagation (BP) have been developed and grown for decades. They have proven their power in nonlinear and complex systems through successful and effective applications in business, medical and industrial fields. Extreme Learning Machine (ELM) is a new member of machine learning technique community but it has been gaining increasing attention for its outstanding performance in learning speed and adaptability.
This project explores the use of machine learning techniques of ELM and SVM to derive at a better analysis of energy consumption and efficiency of large air-conditioning systems. A HVAC system simulation software, DesignBuilder was being used to set up the layout and conditions for simulation as the desired experimental lab located at Lecture Theatre 24 was not ready for use. The data collected from DesignBuilder was analyzed by ELM and SVM to examine the various factors like air temperature and CO2 level and occupancy level’s effect on energy consumption. The most influencing factor on energy consumption would then be identified.
In addition, the performances of ELM and SVM in terms of testing Root Mean Square Error (RMSE) and testing time were compared to find out which machine learning technique is more suitable in handling HVAC data. Discoveries made using the machine learning techniques can be captured and organized in a format for later recall and communication to others. |
author2 |
Soh Yeng Chai |
author_facet |
Soh Yeng Chai Ma, Biao |
format |
Final Year Project |
author |
Ma, Biao |
author_sort |
Ma, Biao |
title |
Machine learning technique for energy efficiency analysis of HVAC systems |
title_short |
Machine learning technique for energy efficiency analysis of HVAC systems |
title_full |
Machine learning technique for energy efficiency analysis of HVAC systems |
title_fullStr |
Machine learning technique for energy efficiency analysis of HVAC systems |
title_full_unstemmed |
Machine learning technique for energy efficiency analysis of HVAC systems |
title_sort |
machine learning technique for energy efficiency analysis of hvac systems |
publishDate |
2014 |
url |
http://hdl.handle.net/10356/61491 |
_version_ |
1772825969387634688 |