Tools
Tools
Naik, K. 
ORCID: 0000-0002-8787-0732,
2023.
Cloud based development of novel data-driven algorithms for heat demand prediction to improve control of heat generation.
PhD, Nottingham Trent University.
![[img]](/style/images/fileicons/text.png) |
Text
Kevin Naik 2023 restric3rdpartycopyright.pdf - Published version Full-text access embargoed until 22 November 2024. Download (59MB) |
Abstract
This thesis explores the potential of machine learning algorithms to improve heat demand prediction in district heating systems. The study compares the performance of various machine learning methods on datasets generated for SEMS (EU H2020 SHARINGCITIES project) and real-life data collected from EU H2020 REMOURBAN project in Nottingham. The thesis demonstrates the importance of selecting the appropriate algorithm for the specific dataset and highlights the potential of machine learning in district heating. Further research in this area could lead to more efficient and sustainable energy use by optimising heat demand prediction in district heating systems.
As proof of concept, an Internet of Things (IoT) framework is implemented in this thesis. The thesis also introduces a novel human-centric approach to utilise real-life data to create an individual heat profile for district heating users and generate realistic individual heat demand of the heating system to set up an optimum heat generation mode.
One of the outcomes of the thesis is that real-time data-driven heat demand prediction is undertaken using more than one machine learning algorithms simultaneously. Additionally, a simulation tool has been developed for data analysis and can be used for further data exploration.
The thesis further investigates the theoretical approach to calculate the heat prediction of the individual homes based on customer behaviour, which can be used to make critical decisions about the energy performance of buildings. This approach is human-centric, as customer behaviour is driven by socio-economic conditions. Finally, the thesis illustrates the cost of optimisation of district heating systems by developing a test rig of a smart radiator using the IoT framework with a novel control strategy.
| Item Type: | Thesis | ||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Description: | Abridged version. | ||||||||||||||||
| Creators: | Naik, K. | ||||||||||||||||
| Contributors: |
|
||||||||||||||||
| Date: | October 2023 | ||||||||||||||||
| Rights: | The author holds the copyright for this work. You are permitted to copy up to 5% of this work for private study or personal, non-commercial research. If you wish to reuse any information contained within this document, please ensure that you fully reference the author, title, university, degree level, and pagination. If you have any queries or requests for any other use, or if you require a more substantial copy, please contact the author directly. | ||||||||||||||||
| Divisions: | Schools > School of Architecture, Design and the Built Environment | ||||||||||||||||
| Record created by: | Linda Sullivan | ||||||||||||||||
| Date Added: | 12 Dec 2023 14:51 | ||||||||||||||||
| Last Modified: | 12 Dec 2023 15:13 | ||||||||||||||||
| URI: | https://irep.ntu.ac.uk/id/eprint/50526 |
Actions (login required)
 |
Edit View |
Views
Views per month over past year
Downloads
Downloads per month over past year
