IJEP 45(1): 16-30 : Vol. 45 Issue. 1 (January 2025)
Ram Kishor Singh, Bindhu Lal* and Navin Prasad
Birla Institute of Technology, Department of Civil and Environmental Engineering, Ranchi – 835 215, Jharkhand, India
Abstract
Vehicular pollution is a major contributor to poor urban air quality. This study aims to estimate the concentrations of primary pollutants emitted from vehicles in high-elevation areas, considering meteorological conditions. Pollution due to vehicular emissions is increasing daily due to changes in people’s economic conditions and lifestyles. Ranchi, the capital of the newly developed state of Jharkhand, has seen a lot of development, causing environmental pollution, especially vehicular pollution. The primary pollutants that contribute to air pollution from vehicular emissions are SOx, NOx and PM10. Air pollution forecasting is an essential step in managing air quality pollution. Various predictive models, such as neural network, Gaussian-support vector machine (SVM), decision tree, ensemble bagged tree, SVM kernel, linear regression, bilayered neural network, XGboost and convolution neural network (CNN) were studied and compared. This research attempts to estimate the concentration of the above pollutants using neural networks, regression tree analysis and deep learning models. For the SOx model, feed forward neural network (FFNN) has an R2 of 0.71 and root mean square error (RMSE) of 2.21; classification and regression tree (CART) has R2 of 0.61 and RMSE of 1.53 and CNN has R2 of 0.99 and RMSE of 4.33. Similar performances have been seen for NOx and PM10 models. The CNN model based on deep learning has performed better than FFNN and CART.
Keywords
Air pollution modelling, Feed forward neural network, SOx, NOx, PM10, Classification of regression tree, Convolution neural network
References
- Zhou, C., S. Li and S. Wang. 2018. Examining the impacts of urban form on air pollution in developing countries: A case study of China’s megacities. Int. J. Env. Res. Public Health. 15(8): 1565. DOI: 10.3 390/ijerph15081565.
- Ahmed, M., et al. 2021. Source apportionment of volatile organic compounds, CO, SO2and trace metals in a complex urban atmosphere. Env. Adv., 5: 100098. DOI: 10.1016/j.envadv.2021.100127.
- Tsai, J. H., M.Y. Lee and H.L. Chiang. 2021. Effectiveness of SOx, NOx and particulate matter control strategies in the improvement of ambient PM concentration in Taiwan. Atmos., 12(4): 460. DOI: 10.3390/ATMOS12040460.
- Ghosh, K. and B. Maitra. 2020. Vulnerability assessment of urban intersections apropos of incident impact on road network and identification of critical intersections. Transportation Res. Record J. Transportation Res. Board. 2674: 8. doi: 10.117 7/0361198120919400.
- Winkler, S.L., et al. 2018. Vehicle criteria pollutant (PM, NOx, CO, HCs) emissions: how low should we go? Climate Atmos. Sci., 1: 26. doi: 10.1038/s41612-018-0037-5.
- Hamilton, R., et al. 2009. The effect of air pollution on cultural heritage. springer verlag, NewYork, USA. DOI: 10.1007/978-0-387-84893-8.
- Chetana, K. and K. Sharda. 2014. A review of vehicular pollution in urban India and its effects on human health. J. Adv. Laboratory Res. Biol., 5(3): 54-61.
- Kheirbek, I., et. al. 2016. The contribution of motor vehicle emissions to ambient fine particulate matter public health impacts in New York city: a health burden assessment. Env. Health.15: 89. DOI: 10.1186/s12940-016-0172-6.
- Goulier, L., et al. 2020. Modelling of urban air pollutant concentrations with artificial neural networks using novel input variables. Int. j. env. res. public health. 17(6): 2025. DOI: 10.3390/ijerph17062 025.
- Gupta, N. and S. Ram. 2023. An approach for modelling vehicular pollution using artificial neural networks. National conference on recent advances in traffic engineering. 377: 19-33.
- Azman, A., et al. 2014. Prediction of the level of air pollution using principal component analysis (PCA) and artificial neural network technique: A case study in Malaysia. Water air soil poll., 225: 2063. DOI: 10.1007/11270-014-2063-I.
- Choi, W., et al. 2013. Evaluating methodological comparability in air quality studies: classification and regression trees for primary pollutants in California’s south coast air basin. Atmos. Env.,
64: 150-159. DOI: 10.1016/j.atmosenv.2012.09. 049. - Ganesh, S.S., P. Arulmozhivarma and R. Tatavarti. 2017. Forecasting air quality index using an ensemble of artificial neural networks and regression models. J. Intelligent Systems. 28(5). DOI: 10.1 515/Jisys-2017-0277.
- Sharma, M., et al. 2021. Forecasting and prediction of air pollutant concentrations using machine learning techniques: The case of India. IOP conf. series Mater. sci. eng., 1022: 012123. doi: 10.10 88/1757-899X/1022/1/012123.
- Snezhana, G., G. Ilieva and M.P. Stoimenova. 2018. PM10prediction and forecasting using CART: A case study for Pleven, Bulgaria. Int. j. env. ecol. eng., 12: 9. DOI: 10.5281/zenodo.1474475.
- Van, N.H., et al. 2023. A new model of air quality prediction using lightweight machine learning. Int. j. env. sci. Tech., 20: 2983- 2994. DOI: 10.1007/s13762-022-04185-w.
- Lozhkina, O.V. and V.N. Lozhkina. 2015. Estimation of road transport-related air pollution in Saint petersburg using european and Russian calculation models. Transp. Res., 36: 178- 189. DOI: 10.1016/j.trd.2015.02.013.
- Gao, Y., et al. 2021. Assessing neighbourhood variations in ozone and PM2.5 concentration using the decision tree method. Building Env., 188: 107479.
- Zichus, M., A.J. Greig and M. Niranjan. 2002. Comparison of four machine learning methods for predicting PM10concentrations in Helsinki, Finland. water air Soil poll., 2: 717-729. DOI: 10.1023/A:1021321820639.
- Althuwaynee, O.F., A.L. Balogui and W.A. Madhoun. 2020. Air pollution hazard assessment using decision tree algorithm and bivariate probability cluster polar function: evaluating inter-correlation clusters of PM10and other air pollutants. GISci. Remote Sensing. 57(2): 207-226. DOI:10.1080/154816 03.2020. 1712064.
- Jorge, G.G., et al. 2021. Road pollution estimation from vehicle tracking in surveillance videos by deep convolutional neural networks. Appl. soft computing. 113: 107950. DOI: 10.1016/j.asoc.2021.107 950.
- Breiman, L., et al. 1984. Classification and regression trees. Chapman and Hall/CRC, New York. DOI: 10.1201/978131513-9470.
- LeCun, Y., Y. Bengio and G. Hinton. 2015. Deep learning. Nature. 521(7553): 436-444. DOI: 10.103 8/nature14539.
- Kiran, S. 2019. Comparing artificial neural networks (ANN) and regression tree (CART) for estimating soil shear strength parameters. J. Adv. res. dynamical control system. 2: 9. DOI: 10.5373/JAR DCS/V11/20192647.
- Zhang, B., et al. 2022. Deep learning for air pollution concentration prediction: A review. Atmos. Env., 290: 119347.
- Samal, K.R.K., et al. 2021. An improved pollution forecasting model with meteorological impact using multiple inputs and fine lining approach. Sustain. cities soc., 70: 102923. DOI: 10.1016/j.scs2 021.102923.
- Sorek-Hamer, M., A.C. Just and I. Kloog. 2020. Tree-based ensemble methods for estimating PM concentrations from satellite observations. Atmos. Poll. Res., 11(8): 1327-1334. DOI: 10.1016/j.apr. 2020.04.011.
- Ivanov, A., et al. 2018. Random forest models for particulate matters: A case study. AIP Conference Proceedings. 2025(1):030001. DOI: 10.1063/1.5 064879.
- Ropkins, K. and J.E. Tate. 2016. Early observations on the impact of the Covid -19 lockdown on air quality trends across the UK. Sci. Total Env., 746: 1417-1432. DOI: 10.1016/j.scitotenv.2020. 142374.
- Forehead, H. and N. Huynh. 2018. Review of modelling air pollution from traffic at street level- The state of the science. Env. Poll., 241: 775-786. DOI: 10.1016/j.envpol.2018.06.019.
