Articles online

Long-term field Evaluation of Low-cost Particulate Matter Sensors in Nanjing

Category: Aerosol Physics and Instrumentation

Accepted Manuscripts
DOI: 10.4209/aaqr.2018.11.0424

Export Citation:  RIS | BibTeX

Lu Bai1, Lin Huang1, Zhenglu Wang2, Qi Ying 1,2, Jun Zheng1, Xiaowen Shi1, Jianlin Hu 1

  • 1 Jiangsu Key Laboratory of Atmospheric Environment Monitoring and Pollution Control, Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology, Nanjing University of Information Science and Technology, Nanjing 210044, China
  • 2 Zachry Department of Civil Engineering, Texas A&M University, College Station, TX 77843, USA


  • 18-month field evaluation on low-cost aerosol monitors was performed.
  • Linear, power-law, and ANN techniques were used for data calibration.
  • Low-cost monitors are more accurate for concentrations over 35 µg m–3.
  • High humidity could cause larger errors for low-cost monitors.
  • A clear sensor deterioration trend is found over the 18-month field calibration.


Low-cost particulate matter (PM) sensors can be widely deployed to measure aerosol concentrations at higher spatial and temporal resolutions than traditional instruments, but they need to be carefully calibrated under ambient conditions. In this study, a long-term field study was conducted from December 2015 to May 2017 at a site in Nanjing to evaluate the capability of some in-house built low-cost PM sensors using the Shinyei PPD42NS sensor for ambient PM2.5 monitoring. A BAM-1020 particulate monitor is co-located with the low-cost sensors to provide a reference reading. Least-square regressions with linear and power-law functions, and an artificial neural network (ANN) technique were used to convert instrument electrical readings to ambient aerosol concentrations. The ANN method leads to the best estimation of the hourly PM2.5 when compared with the BAM-1020 (R2 = 0.84; Mean Normalized Bias = 12.7% and Mean Normalized Error (MNE) = 29.7%). The low-cost PM sensors have relatively better performance for high aerosol concentrations but have larger errors measuring concentrations under 35 μg m-3. High humidity (RH > 75%) could cause larger MNE for the low-cost PM sensors, but the impact of temperature on the low-cost sensors is negligible in this study. A clear sensor deterioration trend is found over the 18–month field calibration. High correlations are found between different low-cost PM sensors and BAM-1020 when calibrated individually, but the correlations are moderate if applying relationship established with one sensor to other sensors, possibly due to internal variations among the sensors. The results suggest that the low-cost PM sensors can measure ambient PM2.5 concentrations with acceptable accuracy, but each sensor should be individually calibrated to improve the accuracy of these sensors. Special attention should be paid to the accuracy of the low-cost PM sensors after long-term application and when applying to high humidity environment.


PPD42NS Field calibration Long-term Artificial neural network

Related Article