Calibration of Low-Cost Air Quality Sensors Using Machine Learning Models
Resumo
Sensores de baixo custo para monitoramento da qualidade do ar têm se destacado como alternativa acessível às estações de referência, porém apresentam limitações de precisão devido à influência de variáveis ambientais. Este trabalho propõe uma abordagem baseada em aprendizado de máquina para a calibração de sensores de material particulado (PM2.5 e PM10), utilizando dados reais coletados em uma cidade do nordeste do Brasil. Foram avaliados os modelos Ridge Regression, Random Forest, XGBoost e Redes Neurais por meio de validação cruzada com cinco dobras. Os resultados indicam que modelos ensemble, especialmente o XGBoost, apresentam melhor desempenho, com R2 de 0,4480 para PM2.5 e 0,6320 para PM10. A abordagem proposta melhora a confiabilidade dos dados e contribui para o monitoramento ambiental de baixo custo em áreas urbanas.
Referências
Agrawal, D., Saini, K., A., Rai, A. C., and Kala, P. (2024). In chamber calibration and performance evaluation of air quality low-cost sensors. Atmospheric Pollution Research, 15:102299.
Albala, E. M., Faet, G. M., Castell, S. F., Solano, J. J. P., and Jordan, R. F. (2025). Assessment of a multisensor zphs01b-based low-cost air quality monitoring system: Case study. Electronics, 14:1531.
Badura, M., Batog, P., Osiadacz, A. D., and Modzel, P. (2018). Evaluation of low-cost sensors for ambient pm2.5 monitoring. Journal of Sensors, pages 1–16.
Bai, Y., Dong, Z., and Liu, L. (2025). Hybrid feature selection-based machine learning methods for thermal preference prediction in diverse seasons and building environments. Building and Environment, 269:112450.
CETESB (2024). Poluentes atmosféricos. Technical report, Companhia Ambiental do Estado de São Paulo. Disponível em: [link]. Acesso em: 16 jun 2025.
Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco California USA.
Ionascu, M.-E., Marcu, M., Bogdan, R., and Darie, M. (2024). Calibration of no, so2, and pm using airify: A low-cost sensor cluster for air quality monitoring. Atmospheric Environment, 339:120841.
Koziel, S., Dabrowska, A., Wojcikowski, M., and Pankiewicz, B. (2025). Efficient calibration of cost-efficient particulate matter sensors using machine learning and time-series alignment. Knowledge-Based Systems, 295:111879.
Lovric, M., Pavlovic, K., Vukovic, M., Grange, S. K., Haberl, M., and Kern, R. (2021). Understanding the true effects of the covid-19 lockdown on air pollution by means of machine learning. Environmental Pollution, 274:115900.
Mead, M. I., Popoola, O. A. M., Stewart, G. B., Landshoff, P., Calleja, M., Hayes, M., Baldovi, J. J., McLeod, M. W., Hodgson, T. F., Dicks, J., Lewis, A., Cohen, J., Baron, R., Saffell, J. R., and Jones, R. L. (2013). The use of electrochemical sensors for monitoring urban air quality in low-cost, high-density networks. Atmospheric Environment, 70:186–203.
Peng, Z., Yang, J., Sun, J., Duan, J., Chen, Z., Niu, X., Hu, T., Huang, Y., Xu, H., Cao, J., and Shen, Z. (2025). Exploring indoor pm2.5 pollution characteristics in xi’an city and its health implications using interpretable machine learning. Sustainable Horizons, 13:100131.
Rawlings, J. O., Dickey, D. A., and Pantula, S. G. (1998). Applied regression analysis: A research tool. New York: Ny Springer.
Refaeilzadeh, P., Tang, L., and Liu, H. (2009). Cross-Validation. Encyclopedia of Database Systems. Springer, US, Boston, MA.
Smith, S., Trefonides, T., Malarvizhi, A. S., LaGarde, S., Liu, J., Jia, X., Wang, Z., Cain, J., Huang, T., Pourhomayoun, M., Llewellyn, G., Phyo, W., Hasheminassab, S., Roberts, J., Marlis, K., Duffy, D. Q., and Yang, C. (2025). A systematic study of popular software packages and ai/ml models for calibrating in situ air quality data: An example with purple air sensors. Sensors, 25:1028.
Snyder, E. G., Watkins, T. H., S., A., P., Thoma, E. D., Williams, R. W., Hagler, G. S. W., Shelow, D., Hindin, D. A., Kilaru, V. J., and Preuss, P. W. (2013). The changing paradigm of air pollution monitoring. Environmental Science and Technology, 47:11369–11377.
Souza, G. R. P., Souza, T. I. A., Batista, N. S., Bertoncini, B. V., and Gomes, D. G. (2025a). Predição de poluentes atmosféricos via métodos de aprendizado de máquina. In Proceedings of the XIX Brazilian e-Science Workshop (BreSci).
Souza, T. I. A., Brito, J. I. C., Bertoncini, B. V., and Gomes, D. G. (2025b). Repoair: Repositório de dados para monitoramento em tempo real da qualidade do ar em ambientes externos. In Proceedings of the XIX Brazilian e-Science Workshop (BreSci).
Vovk, T., Kryza, M., and Werner, M. (2024). Using random forest to improve emep4pl model estimates of daily pm2.5 in poland. Atmospheric Environment, 332:120615.
WHO (2019). Vision 2050: A strategy to decarbonize the global transport sector by midcentury. Technical report, Disponível em: [link]. Acesso em: 17 jun 2025.
Xiang, F., Zhao, Y., Zhang, Z., Zuo, Y., Zou, X., and Tao, F. (2025). Ensemble learning-based stability improvement method for feature selection towards performance prediction. Journal of Manufacturing Systems, 74:55–67.
Xiong, J., Li, A., Liu, Y., Wang, L., Qin, X., and Yu, J. (2022). Multi-scale nanoarchitectured fibrous networks for high-performance, self-sterilization, and recyclable face masks. Small, 18.
Albala, E. M., Faet, G. M., Castell, S. F., Solano, J. J. P., and Jordan, R. F. (2025). Assessment of a multisensor zphs01b-based low-cost air quality monitoring system: Case study. Electronics, 14:1531.
Badura, M., Batog, P., Osiadacz, A. D., and Modzel, P. (2018). Evaluation of low-cost sensors for ambient pm2.5 monitoring. Journal of Sensors, pages 1–16.
Bai, Y., Dong, Z., and Liu, L. (2025). Hybrid feature selection-based machine learning methods for thermal preference prediction in diverse seasons and building environments. Building and Environment, 269:112450.
CETESB (2024). Poluentes atmosféricos. Technical report, Companhia Ambiental do Estado de São Paulo. Disponível em: [link]. Acesso em: 16 jun 2025.
Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco California USA.
Ionascu, M.-E., Marcu, M., Bogdan, R., and Darie, M. (2024). Calibration of no, so2, and pm using airify: A low-cost sensor cluster for air quality monitoring. Atmospheric Environment, 339:120841.
Koziel, S., Dabrowska, A., Wojcikowski, M., and Pankiewicz, B. (2025). Efficient calibration of cost-efficient particulate matter sensors using machine learning and time-series alignment. Knowledge-Based Systems, 295:111879.
Lovric, M., Pavlovic, K., Vukovic, M., Grange, S. K., Haberl, M., and Kern, R. (2021). Understanding the true effects of the covid-19 lockdown on air pollution by means of machine learning. Environmental Pollution, 274:115900.
Mead, M. I., Popoola, O. A. M., Stewart, G. B., Landshoff, P., Calleja, M., Hayes, M., Baldovi, J. J., McLeod, M. W., Hodgson, T. F., Dicks, J., Lewis, A., Cohen, J., Baron, R., Saffell, J. R., and Jones, R. L. (2013). The use of electrochemical sensors for monitoring urban air quality in low-cost, high-density networks. Atmospheric Environment, 70:186–203.
Peng, Z., Yang, J., Sun, J., Duan, J., Chen, Z., Niu, X., Hu, T., Huang, Y., Xu, H., Cao, J., and Shen, Z. (2025). Exploring indoor pm2.5 pollution characteristics in xi’an city and its health implications using interpretable machine learning. Sustainable Horizons, 13:100131.
Rawlings, J. O., Dickey, D. A., and Pantula, S. G. (1998). Applied regression analysis: A research tool. New York: Ny Springer.
Refaeilzadeh, P., Tang, L., and Liu, H. (2009). Cross-Validation. Encyclopedia of Database Systems. Springer, US, Boston, MA.
Smith, S., Trefonides, T., Malarvizhi, A. S., LaGarde, S., Liu, J., Jia, X., Wang, Z., Cain, J., Huang, T., Pourhomayoun, M., Llewellyn, G., Phyo, W., Hasheminassab, S., Roberts, J., Marlis, K., Duffy, D. Q., and Yang, C. (2025). A systematic study of popular software packages and ai/ml models for calibrating in situ air quality data: An example with purple air sensors. Sensors, 25:1028.
Snyder, E. G., Watkins, T. H., S., A., P., Thoma, E. D., Williams, R. W., Hagler, G. S. W., Shelow, D., Hindin, D. A., Kilaru, V. J., and Preuss, P. W. (2013). The changing paradigm of air pollution monitoring. Environmental Science and Technology, 47:11369–11377.
Souza, G. R. P., Souza, T. I. A., Batista, N. S., Bertoncini, B. V., and Gomes, D. G. (2025a). Predição de poluentes atmosféricos via métodos de aprendizado de máquina. In Proceedings of the XIX Brazilian e-Science Workshop (BreSci).
Souza, T. I. A., Brito, J. I. C., Bertoncini, B. V., and Gomes, D. G. (2025b). Repoair: Repositório de dados para monitoramento em tempo real da qualidade do ar em ambientes externos. In Proceedings of the XIX Brazilian e-Science Workshop (BreSci).
Vovk, T., Kryza, M., and Werner, M. (2024). Using random forest to improve emep4pl model estimates of daily pm2.5 in poland. Atmospheric Environment, 332:120615.
WHO (2019). Vision 2050: A strategy to decarbonize the global transport sector by midcentury. Technical report, Disponível em: [link]. Acesso em: 17 jun 2025.
Xiang, F., Zhao, Y., Zhang, Z., Zuo, Y., Zou, X., and Tao, F. (2025). Ensemble learning-based stability improvement method for feature selection towards performance prediction. Journal of Manufacturing Systems, 74:55–67.
Xiong, J., Li, A., Liu, Y., Wang, L., Qin, X., and Yu, J. (2022). Multi-scale nanoarchitectured fibrous networks for high-performance, self-sterilization, and recyclable face masks. Small, 18.
Publicado
19/10/2026
Como Citar
CASTRO, Andrey S.; SOUZA, Thiago I. A.; BRITO, Jairo I. C.; GOMES, Danielo G.; BERTONCINI, Bruno V..
Calibration of Low-Cost Air Quality Sensors Using Machine Learning Models. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 23. , 2026, Cuiabá/MT.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 2041-2052.
ISSN 2763-9061.
DOI: https://doi.org/10.5753/eniac.2026.31497.
