Design, Comparison and Application of Artificial Intelligence Predictive Models Based on Experimental Data for Estimating Carbon Dioxide Concentration Inside a Building

Abstract

This paper outlines the development of an affordable microclimate station built with Wi-Fi-enabled microcontrollers similar to Arduino, designed to monitor environmental factors such as temperature, humidity, human presence, atmospheric pressure, and carbon dioxide levels. Installed in a primary school classroom in Bialystok, Poland, the station gathers data on a minute-by-minute basis. This data is then utilized to create and assess predictive models for estimating indoor CO2 levels, even in the absence of a direct CO2 sensor. Among the various models tested, the random forest method, which relied solely on temperature, humidity, and human presence measurements, produced the most accurate results, achieving an R-squared value of 0.89. The use of temperature, humidity, and presence sensors, which are more affordable than CO2 sensors, highlights the cost-effectiveness of predictive modeling in environmental monitoring.

Keywords: Predictive models, Artificial Intelligence, CO2 estimation, Microcontroller, IAQ, public buildings.