Monitoring of Air Quality with Low-Cost Electrochemical Sensors and the Use of Artificial Neural Networks for the Atmospheric Pollutants Concentration Levels Prediction

Abstract

This paper shows the preliminary results of the monitoring and estimation of air pollutants at a strategic point within the district of San Isidro, Lima - Peru. Low-cost, portable, wireless and geo-locatable electrochemical sensors were used to capture reliable contamination levels in real-time which could be used not only to quantify atmospheric pollution exposure but also for prevention and control, and even for legislative purposes. For the prediction of \(CO{_2}\) and \(SO{_2}\) levels, computational intelligence algorithms were applied and validated with experimental data. We proved that the use of Artificial Neural Networks (ANNs) has a high potential as a tool to use it as a forecast methodology in the area of air pollution.