Field calibration and performance evaluation of low-cost sensors for monitoring airborne PM in the occupational mining environment

Abstract

Surface mining operations emit particulate matter (PM) with varying sizes and compositions, adversely affecting the health of mine workers and adjoining communities. Measurements for regulatory compliance at present are obtained by bulky instruments, which are costly and provide data averaged over a period of time. Recently, low-cost PM sensors (LCPMS) have been increasingly used in indoor and outdoor environments for monitoring real-time PM concentrations with high spatial and temporal resolution. No comprehensive study has been undertaken to evaluate the performance of LCPMS when exposed to uncontrolled high dust concentrations prevailing in the occupational mining environment. The study aimed to achieve this by deploying three sensors (OPC N3, SPS30, and SDS011) in a highly mechanized surface coal mine. A comparative assessment of these sensors has been conducted by calibrating them against a research-grade reference instrument in uncontrolled indoor, outdoor, and mining environments. The performance of OPC N3 and SPS30 sensors was found to be good, with high linearity (R² = 0.90 – 0.99) and precision (CV = 2 – 6%). Incorporating the monitored local meteorological conditions and PM proportions improved the performance of the applied calibration models. The decision tree-based regression model performed better (R² = 0.95 – 99; RMSE = 0.8 – 11.4; MAE = 0.4 – 5.3) compared to the multiple linear model in accurately predicting reference equivalent PM mass concentration measurements. The comprehensive performance evaluation of LCPMS in this study highlights its potential applications in the different occupational environments particularly for monitoring the personal exposure of industry workers.