Application of low-cost UAV integrated with air quality monitoring sensors and artificial neural networks to simulate air pollution at a stone quarry

Abstract

This paper evaluates the application of low-cost unmanned aerial vehicles (UAVs) with integrated air quality monitoring sensors UMS-AM in environmental monitoring and assessment in construction-stone mining areas and predicting the density of components using multi-layer artificial neural networks (MLP Neural Nets). To achieve these goals, DJI Inspire 2 drones equipped with PM1.0, PM2.5 and PM10 fine dust sensors were used for experimental measurements at two stone quarries Tan My and Thuong Tan in Binh Duong province. In addition, MLP Neural Nets were used to predict the density of air components in the stone quarries when the production process is expanded with deep mining modules and larger modules in the future. Finally, 3D simulation models of PM1.0, PM2.5 and PM10 indexes were built. The results of the study show that low-cost dust monitoring in quarry areas using UAVs is completely feasible and should be considered for application in other open-pit mines, contributing positively and effectively to the control of greenhouse gas emissions in open-pit mining areas.