Occupancy modelling using data driven models

Heating, Ventilation and Air-conditioning (HVAC) systems are typically designed using static extreme values, resulting in them being over dimensioned for most of their operating time. To reduce the energy wastage, efficient control of the usage of HVAC can be done through occupancy sensing. Effectiv...

Full description

Saved in:
Bibliographic Details
Main Author: Teng, Sherlyn Xue Qi
Other Authors: Soh Yeng Chai
Format: Final Year Project
Language:English
Published: Nanyang Technological University 2023
Subjects:
Online Access:https://hdl.handle.net/10356/167451
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Heating, Ventilation and Air-conditioning (HVAC) systems are typically designed using static extreme values, resulting in them being over dimensioned for most of their operating time. To reduce the energy wastage, efficient control of the usage of HVAC can be done through occupancy sensing. Effective feature engineering methods will be used to estimate the accuracy of the occupancy in the building. Non-intrusive aspect of the environmental sensors such as the CO2, humidity, lighting, and temperature is used to collect data. The approach of visualising the different features and analyse its usefulness in estimating the occupancy in the enclosed space will be analysed. In this discussion, various machine learning methods are used to model the occupancy and estimate the accuracy. The models with the best accuracies will be presented and further discussed.