Assessment of Indoor Air Quality in Academic Buildings Using IoT and Deep Learning

Abstract

Humans spend most of their lifetime indoors; thus, it is important to keep indoor air quality within acceptable levels. As a result, many initiatives have been developed by multiple research centers or through academic studies to address the harmful effects of increased indoor pollutants on public health. This research introduces a system for monitoring different air parameters to evaluate the indoor air quality (IAQ) and to provide real-time readings. The proposed system aims to enhance planning and controlling measures and increase both safety and occupants’ comfort. The system combines microcontrollers and electronic sensors to form an Internet of Things (IoT) solution that collects different indoor readings. The readings are then compared with outdoor readings for the same experiment period and prepared for further processing using artificial intelligence (AI) models. The results showed the high effectiveness of the IoT device in transferring data via Wi-Fi with minimum disruptions and missing data. The average readings for temperature, humidity, air pressure, CO2, CO, and PM2.5 in the presented case study are 30 °C, 42%, 100,422 pa, 460 ppm, 2.2 ppm, and 15.3 µ/m3, respectively. The developed model was able to predict multiple air parameters with acceptable accuracy. It can be concluded that the proposed system proved itself as a powerful forecasting and management tool for monitoring and controlling IAQ.