Artificial Intelligence for Evaluation of PM2.5 and PM10 Levels in Urban Streets

Abstract

Atmospheric pollution, especially due to fine particles (PM2.5) and larger particles (PM10), constitutes a major risk to public health and the long-term viability of urban areas. This research focuses on measuring the presence of PM2.5 and PM10 across selected city streets, investigating the variables that affect their concentration, and identifying possible solutions to limit their impact. The study utilizes continuous air quality monitoring, data interpretation through statistical tools, and a review of urban planning measures to offer a thorough understanding of particulate contamination in metropolitan settings. Findings reveal significant patterns in how these pollutants spread, their links to vehicular traffic and weather conditions, and suggest practical approaches to lower human exposure. The contribution of artificial intelligence starts in collecting raw data, converting in Matlab’s matrices and help the operator identify correlations between the taken measurements and the the influence of environmental parameters, which allow particulate matter to vary daily even without following consolidated trends in the same time period.