AI aided IoT sensor for green building application

• Main Author: Mylsamy Boojha Rukmi
• Other Authors: Chau Yuen
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; Artificial intelligence; Sustainability; Internet of things; Machine learning
• Online Access: https://hdl.handle.net/10356/181707
• Institution: Nanyang Technological University

Recently Singapore has been focusing a lot more on being a green city and being sustainable. Green buildings have emerged as a pivotal solution to reduce energy consumption and minimise environmental impact. The integration of Artificial Intelligence (AI) and Internet of Things (IoT) sensors in building management systems have been an integral part in increasing comfort and reducing energy usage, specifically in heating, ventilation and air conditioning (HVAC) systems. The primary objective of this project is to determine which Machine Learning (ML) model and AI guided system would be the best to predict the internal temperature in a room and in the future adjusting air conditioner settings in real time. An all-in-one environmental sensor is deployed to collect data such as motion, Carbon Dioxide (CO2) levels, humidity and lux (light intensity). This data is then used to train various machine learning models to determine the best model to predict the ideal internal temperature of the room which is also the ideal air conditioning setting. This AI system aligns with NTU’s goal of achieving carbon neutrality by 2035 as well as the goal to reduce waste generation and energy consumption by 50% by March 2026. This also aligns with Singapore’s Green Plan 2030, which aims to reduce carbon emissions, improve energy efficiency and promote sustainability. Green buildings play a vital role in realising these sustainability goals. This project showcases the potential of AI and IoT technologies in green buildings that can reduce operational costs and environmental impact. Future work would involve refining the machine learning algorithms for greater accuracy and expanding to other building management functions such as ventilation and lighting.