Journals / Atmospheric Pollution Research / 2019 / Cilt: 10 - Sayı: 1

Applying machine learning methods in managing urban concentrations oftraffic-related particulate matter (PM10and PM2.5)

Pages
134–144
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

In conclusion,well trained ANN and BRT models can be successfully applied in predictions of roadside PM10and PM2.5con-centrations.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:This study presents a new method for evaluating the effectiveness of roadside PM10and PM2.5reduction sce-narios using Machine Learning (ML) based models.
📊
Key Findings & Data:Traffic, meteorological and pollutant datacollected at nineteen Air Quality Monitoring (AQM) sites in London for a period between 2007 and 2012 wasused.
💡
Conclusion & Impact:In conclusion,well trained ANN and BRT models can be successfully applied in predictions of roadside PM10and PM2.5con-centrations.
🔬Design:Modeling & Simulation
📍Scope:2007 Verileri
⚙️Variables:Domain Parameters

Özet

This study presents a new method for evaluating the effectiveness of roadside PM10and PM2.5reduction sce-narios using Machine Learning (ML) based models. The ML methods include Artificial Neural Networks (ANN),Boosted Regression Trees (BRT) and Support Vector Machines (SVM). Traffic, meteorological and pollutant datacollected at nineteen Air Quality Monitoring (AQM) sites in London for a period between 2007 and 2012 wasused. The ML models performed very well in predicting the concentrations of PM10and PM2.5with around 95%of their predictions falling within the factor of two of the observed concentrations at the roadsides. The pre-diction errors observed were very small as indicated by the average normalised mean gross errors of 0.2. Also,the predictions of the models correlated well with the observed concentrations as shown by the average values ofR (0.8) and index of agreement (0.74). Additionally, when some PM10and PM2.5reduction scenarios weremodelled, the ML models predicted various degree of reductions in the roadside concentrations. In conclusion,well trained ANN and BRT models can be successfully applied in predictions of roadside PM10and PM2.5con-centrations. Moreover, they can be applied in measuring the effectiveness of roadside particle reduction sce-narios.

🔖

Cite This Article

SULEİMAN, A., TİGHT, M., QUİNN, A. (2019). Applying machine learning methods in managing urban concentrations oftraffic-related particulate matter (PM10and PM2.5). Atmospheric Pollution Research, Cilt: 10 - Sayı: 1, 134–144. https://ajindex.com/articles/198409

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (10)

  1. Aurelie, C., Harrison, R.M., 2005. Comparison between SMPS, nano-SMPS andEpiphaniometer data at an urban Background site (Bloomsbury) and a Roadside site(Marylebone road).http://uk-air.defra.gov.uk/reports/cat05/0506061412_Comparison_between_SMPS-Nano-SMPS.pdf. Balsamà, A.P., De Biase, L., Janssens-Maenhout, G., Pagliari, V., 2014. Near-term pro-jection of anthropogenic emission trends using neural networks. Atmos. Environ. 89,581–592.
  2. Berkowicz, R., 2000. OSPM - a parameterised street pollution model. Environ. Monit.Assess. 65, 323–331.
  3. Buuren, S., Groothuis-Oudshoorn, K., 2011. mice: multivariate imputation by chainedequations in R. J. Stat. Software 45.Chave, J., Levin, S., 2003. Scale and scaling in ecological and economic systems. Environ.Resour. Econ. 26, 527–557.
  4. Cherkassky, V., Ma, Y., 2004. Practical selection of SVM parameters and noise estimationfor SVM regression. Neural Networks 17, 113–126.
  5. Chunming, W., Xinbiao, L., Xiaofeng, C., Yalong, M., Chen, C., 2010. Application ofsupport vector regression to predict metallogenic favourability degree. Int. J. Phys.Sci. 5, 5. De Gennaro, G., Trizio, L., Di Gilio, A., Pey, J., Pérez, N., Cusack, M., Alastuey, A., Querol,X., 2013. Neural network model for the prediction of PM10daily concentrations intwo sites in the Western Mediterranean. Sci. Total Environ. 463, 875–883.
  6. Debry, E., Mallet, V., 2014. Ensemble forecasting with machine learning algorithms forozone, nitrogen dioxide and PM10 on the Prev'Air platform. Atmos. Environ. 91,71–84. DEFRA, 2015.
  7. DEFRA, Emissions Factors Toolkit [Online]. Available:http://laqm.defra.gov.uk/review-and-assessment/tools/emissions-factors-toolkit.html, Accessed date: 6April 2015 2016. DFT, 2014. Traffic Counts [Online]. Department for Transport. Available:http://www.dft.gov.uk/traffic-counts/, Accessed date: 15 February 2014.
  8. Ding, S., Jia, W., Su, C., Zhang, L., Liu, L., 2011a. Research of neural network algorithmbased on factor analysis and cluster analysis. Neural Comput. Appl. 20, 297–302.
  9. Ding, S., Li, H., Su, C., Yu, J., Jin, F., 2011b. Evolutionary artificial neural networks: areview. Artif. Intell. Rev. 39, 251–260.
  10. Elangasinghe, M.A., Singhal, N., Dirks, K.N., Salmond, J.A., 2014a. Development of anANN-based air pollution forecasting system with explicit knowledge through sensi-tivity analysis. Atmos. Pollut. Res. 5. Elangasinghe, M.A., Singhal, N., Dirks, K.N., Salmond, J.A., Samarasinghe, S., 2014b.Complex time series analysis of PM10 and PM2.5 for a coastal