Enhancing Indoor Air Quality Through Smart Home Automation System
Jha, Shardendu (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202504287912
https://urn.fi/URN:NBN:fi:amk-202504287912
Tiivistelmä
This research aims to design and develop an intelligent indoor air quality system utilizing advanced air quality sensor technology. The fundamental architecture of the model seeks to establish a comprehensive indoor air quality automation system that enhances the health and well-being of apartment occupants. This system addresses challenges such as data security and privacy, cost-effectiveness, and energy efficiency by integrating smart indoor air quality monitoring and home automation technologies. The research question focuses on addressing these challenges and gaps, as identified during the literature review.
To achieve these objectives, the study employs machine learning models to forecast key indoor air quality parameters, including temperature, humidity, CO2 (Carbon Dioxide), VOC (Volatile Organic Compounds) , and PM2.5 (Particulate Matter) levels. Data is stored on an external hard drive, and predefined thresholds for predicted temperature and humidity are used to automate systems, thereby saving energy and resources. The system includes an alert mechanism to notify users via email when thresholds are exceeded and employs a smart hub and air cooler to maintain optimal indoor air quality. Additionally, it generates comprehensive analytical insights into training and testing performance metrics. The research also incorporates energy cost calculations, making the system cost-effective and particularly beneficial for individuals with disabilities who may face challenges managing their environment.
The research also focuses on user privacy by implementing robust privacy measures, including a Decentralized Private Network, to protect the user’s network and valuable indoor data.
Furthermore, this thesis involves the development of a user-friendly web interface tailored for occupants. The system's performance is rigorously evaluated through experiments conducted in residential settings, with historical data stored locally to enable detailed analysis.
To achieve these objectives, the study employs machine learning models to forecast key indoor air quality parameters, including temperature, humidity, CO2 (Carbon Dioxide), VOC (Volatile Organic Compounds) , and PM2.5 (Particulate Matter) levels. Data is stored on an external hard drive, and predefined thresholds for predicted temperature and humidity are used to automate systems, thereby saving energy and resources. The system includes an alert mechanism to notify users via email when thresholds are exceeded and employs a smart hub and air cooler to maintain optimal indoor air quality. Additionally, it generates comprehensive analytical insights into training and testing performance metrics. The research also incorporates energy cost calculations, making the system cost-effective and particularly beneficial for individuals with disabilities who may face challenges managing their environment.
The research also focuses on user privacy by implementing robust privacy measures, including a Decentralized Private Network, to protect the user’s network and valuable indoor data.
Furthermore, this thesis involves the development of a user-friendly web interface tailored for occupants. The system's performance is rigorously evaluated through experiments conducted in residential settings, with historical data stored locally to enable detailed analysis.