Skip to main content
SpringerLink
Log in

A real-time hourly ozone prediction system using deep convolutional neural network

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

This study uses a deep learning approach to forecast ozone concentrations over Seoul, South Korea for 2017. We use a deep convolutional neural network (CNN). We apply this method to predict the hourly ozone concentration on each day for the entire year using several predictors from the previous day, including the wind fields, temperature, relative humidity, pressure, and precipitation, along with in situ ozone and NO2 concentrations. We refer to a history of all observed parameters between 2014 and 2016 for training the predictive models. Model-measurement comparisons for the 25 monitoring sites for the year 2017 report average indices of agreement (IOA) of 0.84–0.89 and a Pearson correlation coefficient of 0.74–0.81, indicating reasonable performance for the CNN forecasting model. Although the CNN model successfully captures daily trends as well as yearly high and low variations of the ozone concentrations, it notably underpredicts high ozone peaks during the summer. The forecasting results are generally more accurate for the stations located in the southern regions of the Han River, the result of more stable topographical and meteorological conditions. Furthermore, through two separate daytime and nighttime forecasts, we find that the monthly IOA of the CNN model is 0.05–0.30 higher during the daytime, resulting from the unavailability of some of the input parameters during the nighttime. While the CNN model can predict the next 24 h of ozone concentrations within less than a minute, we identify several limitations of deep learning models for real-time air quality forecasting for further improvement.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  1. Jenkin ME, Clemitshaw KC (2000) Ozone and other secondary photochemical pollutants: chemical processes governing their formation in the planetary boundary layer. Atmos Environ 34(16):2499–2527

    Google Scholar 

  2. Jerrett M, Burnett RT, Pope CA III, Ito K, Thurston G, Krewski D et al (2009) Long-term ozone exposure and mortality. N Engl J Med 360(11):1085–1095

    Google Scholar 

  3. McKeen S, Wilczak J, Grell G, Djalalova I, Peckham S, Hsie EY et al (2005) Assessment of an ensemble of seven real-time ozone forecasts over eastern North America during the summer of 2004. J Geophys Res 110(D21):D21307

    Google Scholar 

  4. McHenry JN, Ryan WF, Seaman NL, Coats CJ Jr, Pudykiewicz J, Arunachalam S, Vukovich JM (2004) A real-time Eulerian photochemical model forecast system: overview and initial ozone forecast performance in the northeast US corridor. B Am Meteorol Soc 85(4):525–548

    Google Scholar 

  5. Zhang Y, Bocquet M, Mallet V, Seigneur C, Baklanov A (2012) Real-time air quality forecasting, part I: history, techniques, and current status. Atmos Environ 60:632–655

    Google Scholar 

  6. Zhang Y, Bocquet M, Mallet V, Seigneur C, Baklanov A (2012) Real-time air quality forecasting, part II: state of the science, current research needs, and future prospects. Atmos Environ 60:656–676

    Google Scholar 

  7. Byun D, Schere KL (2006) Review of the governing equations, computational algorithms, and other components of the models-3 community multiscale air quality (CMAQ) modeling system. Appl Mech Rev 59:51–77

    Google Scholar 

  8. Slini T, Karatzas K, Moussiopoulos N (2001) Statistical analysis of environmental data as the basis of forecasting: an air quality application. Sci Total Environ 288:227–237

    Google Scholar 

  9. Choi Y (2014) The impact of satellite-adjusted NOx emissions on simulated NOx and O3 discrepancies in the urban and outflow areas of the Pacific and Lower Middle US. Atmos Chem Phys 14(2):675–690

    Google Scholar 

  10. Pan S, Choi Y, Roy A, Jeon W (2017) Allocating emissions to 4 km and 1 km horizontal spatial resolutions and its impact on simulated NOx and O3 in Houston, TX. Atmos Environ 164:398–415

    Google Scholar 

  11. Pan S, Choi Y, Jeon W, Roy A, Westenbarger DA, Kim HC (2017) Impact of high-resolution sea surface temperature, emission spikes and wind on simulated surface ozone in Houston, Texas during a high ozone episode. Atmos Environ 152:362–376

    Google Scholar 

  12. Durão RM, Mendes MT, Pereira MJ (2016) Forecasting O3 levels in industrial area surroundings up to 24 h in advance, combining classification trees and MLP models. Atmos Pollut Res 7:961–970

    Google Scholar 

  13. Sun W, Palazoglu A, Singh A, Zhang H, Wang Q, Zhao Z, Cao D (2015) Prediction of surface ozone episodes using clusters based generalized linear mixed effects models in Houston–Galveston–Brazoria area, Texas. Atmos Pollut Res 6:245–253

    Google Scholar 

  14. Siwek K, Osowski S (2016) Data mining methods for prediction of air pollution. Int J Appl Math Comput Sci 26:467–478

    MathSciNet  MATH  Google Scholar 

  15. Wang D, Wei S, Luo H, Yue C, Grunder O (2017) A novel hybrid model for air quality index forecasting based on two-phase decomposition technique and modified extreme learning machine. Sci Total Environ 580:719–733

    Google Scholar 

  16. Pandey G, Zhang B, Jian L (2013) Predicting submicron air pollution indicators: a machine learning approach. Environ Sci Process Impacts 15(5):996–1005

    Google Scholar 

  17. Williams DRGHR, Hinton GE (1986) Learning representations by back-propagating errors. Nature 323:533–538

    MATH  Google Scholar 

  18. Kuremoto T, Kimura S, Kobayashi K, Obayashi M (2014) Time series forecasting using a deep belief network with restricted Boltzmann machines. Neurocomputing 137:47–56

    Google Scholar 

  19. Kumar A, Goyal P (2013) Forecasting of air quality index in Delhi using neural network based on principal component analysis. Pure Appl Geophys 170:711–722

    Google Scholar 

  20. Feng X, Li Q, Zhu Y, Hou J, Jin L, Wang J (2015) Artificial neural networks forecasting of PM25 pollution using air mass trajectory based geographic model and wavelet transformation. Atmos Environ 107:118–128

    Google Scholar 

  21. Peng H, Lima AR, Teakles A, Jin J, Cannon AJ, Hsieh WW (2016) Evaluating hourly air quality forecasting in Canada with nonlinear updatable machine learning methods. Air Qual Atmos Health 10:195–211

    Google Scholar 

  22. Hecht-Nielsen R (1988) Theory of the backpropagation neural network. Neural Netw 1:445–448

    Google Scholar 

  23. Panchal G, Ganatra A, Kosta YP, Panchal D (2011) Behaviour analysis of multilayer perceptrons with multiple hidden neurons and hidden layers. Int J Comput Sci Eng 3:333–337

    Google Scholar 

  24. Bengio Y, Lamblin P, Popovici D, Larochelle H (2007) Greedy layer-wise training of deep networks. Adv Neural Inform Process Syst 19:153–160

    Google Scholar 

  25. Hinton GE, Osindero S, Teh YW (2006) A fast learning algorithm for deep belief nets. Neural Comput 18:1527–1554

    MathSciNet  MATH  Google Scholar 

  26. Hinton GE, Salakhutdinov RR (2006) Reducing the dimensionality of data with neural networks. Science 313:504–507

    MathSciNet  MATH  Google Scholar 

  27. Bengio Y (2009) Learning deep architectures for AI. Found Trend Mach Learn 2:1–127

    MATH  Google Scholar 

  28. Zhang CY, Chen CP, Chen D, Ng KT (2016) MapReduce based distributed learning algorithm for restricted Boltzmann machine. Neurocomputing 198:4–11

    Google Scholar 

  29. Zhang CY, Chen CP, Gan M, Chen L (2015) Predictive deep Boltzmann machine for multiperiod wind speed forecasting. IEEE Trans Sustain Energy 6:1416–1425

    Google Scholar 

  30. Deng L, Yu D (2014) Deep learning: methods and applications. Found Trend Signal Process 7:197–387

    MathSciNet  MATH  Google Scholar 

  31. Schmidhuber J (2015) Deep learning in neural networks: an overview. Neural Netw 61:85–117

    Google Scholar 

  32. LeCun Y, Bengio Y, Hinton GE (2015) Deep learning. Nature 521:436–444

    Google Scholar 

  33. Hrasko R, Pacheco AG, Krohling RA (2015) Time series prediction using restricted boltzmann machines and backpropagation. Procedia Comput Sci 55:990–999

    Google Scholar 

  34. Cai X, Hu S, Lin X (2012) Feature extraction using restricted Boltzmann machine for stock price prediction. In: Proceedings of IEEE international conference computer science and automation engineering (CSAE), Zhangjiajie, China, vol 3, pp 80–83

  35. Li H, Hou J, Adhikari B, Lyu Q, Cheng J (2017) Deep learning methods for protein torsion angle prediction. BMC Bioinform 18:417

    Google Scholar 

  36. Li X, Peng L, Hu Y, Shao J, Chi T (2016) Deep learning architecture for air quality predictions. Environ Sci Pollut Res 23:22408–22417

    Google Scholar 

  37. Li X, Peng L, Yao X, Cui S, Hu Y, You C, Chi T (2017) Long short-term memory neural network for air pollutant concentration predictions: method development and evaluation. Environ Pollut 231:997–1004

    Google Scholar 

  38. Li T, Shen H, Yuan Q, Zhang X, Zhang L (2017) Estimating ground-level PM25 by fusing satellite and station observations: a geo-intelligent deep learning approach. Geophys Res Lett 44(23):11985–11993

    Google Scholar 

  39. Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Proc Adv Neural Inf Process Syst 25:1090–1098

    Google Scholar 

  40. Hu W, Huang Y, Wei L, Zhang F, Li H (2015) Deep convolutional neural networks for hyperspectral image classification. J Sens 2015:1–12

    Google Scholar 

  41. Jeon WB, Lee SH, Lee H, Park C, Kim DH, Park SY (2014) A study on high ozone formation mechanism associated with change of NOx/VOCs ratio at a rural area in the Korean Peninsula. Atmos Environ 89:10–21

    Google Scholar 

  42. Jeon WB, Lee SH, Lee HW, Kim HG (2012) Process analysis of the impact of atmospheric recirculation on consecutive high-O3 episodes over the Seoul Metropolitan Area in the Korean Peninsula. Atmos Environ 63:213–222

    Google Scholar 

  43. Bengio Y, LeCun Y (2007) Scaling learning algorithms towards AI. In: Bottou L, Chapelle O, DeCoste D, Weston J (eds) Large scale kernel machines. MIT Press, Cambridge, pp 321–360

    Google Scholar 

  44. Bengio Y, Simard P, Frasconi P (1994) Learning long-term dependencies with gradient descent is difficult. IEEE Trans Neural Netw 5:157–166

    Google Scholar 

  45. Gu J, Wang Z, Kuen J, Ma L, Shahroudy A, Shuai B et al (2018) Recent advances in convolutional neural networks. Pattern Recogn 77:354–377

    Google Scholar 

  46. Nogueira K, Penatti OA, dos Santos JA (2017) Towards better exploiting convolutional neural networks for remote sensing scene classification. Pattern Recogn 61:539–556

    Google Scholar 

  47. Kim Y (2014) Convolutional neural networks for sentence classification. arXiv preprint. arXiv:14085882

  48. Lawrence S, Giles CL, Tsoi AC, Back AD (1997) Face recognition: a convolutional neural-network approach. IEEE Trans Neural Netw 8(1):98–113

    Google Scholar 

  49. Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S et al (2015) Imagenet large scale visual recognition challenge. Int J Comput Vision 115(3):211–252

    MathSciNet  Google Scholar 

  50. Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:14091556

  51. Sun Yi, Wang X, Tang X (2013) Deep convolutional network cascade for facial point detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 3476–3483

  52. Belagiannis V, Rupprecht C, Carneiro G, Navab N (2015) Robust optimization for deep regression. In Proceedings of the IEEE international conference on computer vision, pp 2830–2838

  53. LeCun Y, Bengio Y (1995) Convolutional networks for images, speech, and time series. Handb Brain Theory Neural Netw 3361(10):1

    Google Scholar 

  54. Hinton GE (2014) What’s wrong with convolutional nets? MIT brain and cognitive sciences—fall colloquium series. URL http://www.techtvmitedu/collections/bcs/videos/30698-what-s-wrong-with-convolutional-nets. Accessed 10 Feb 2017

  55. Chollet F (2015) Keras. https://www.kerasio

  56. Abadi M, Barham P, Chen J, Chen Z, Davis A, Dean J, Devin M et al (2016) Tensor flow: a system for large-scale machine learning. OSDI 16:265–283

    Google Scholar 

  57. Mazumder R, Hastie T, Tibshirani R (2010) Spectral regularization algorithms for learning large incomplete matrices. J Mach Learn Res 11:2287–2322

    MathSciNet  MATH  Google Scholar 

  58. Willmott CJ (1981) On the validation of models. Phys Geogr 2(2):184–194

    Google Scholar 

  59. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint. arXiv:1412.6980

  60. Chen XW, Lin X (2014) Big data deep learning: challenges and perspectives. IEEE Access 2:514–525

    Google Scholar 

  61. Bengio Y, Courville A, Vincent P (2013) Representation learning: a review and new perspectives. IEEE Trans Pattern Anal Mach Intell 35:1798–1828

    Google Scholar 

Download references

Acknowledgements

This study was supported by funding from the Department of Earth and Atmospheric Science (EAS Research Grant) of the University of Houston and the National Institute of Environmental Research (NIER).

Author information

Authors and Affiliations

  1. Department of Earth and Atmospheric Sciences, University of Houston, Houston, TX, 77004, USA

    Ebrahim Eslami, Yunsoo Choi, Yannic Lops & Alqamah Sayeed

Authors
  1. Ebrahim Eslami
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yunsoo Choi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Yannic Lops
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Alqamah Sayeed
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Yunsoo Choi.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (DOCX 1925 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Eslami, E., Choi, Y., Lops, Y. et al. A real-time hourly ozone prediction system using deep convolutional neural network. Neural Comput & Applic 32, 8783–8797 (2020). https://doi.org/10.1007/s00521-019-04282-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-019-04282-x

Keywords

Search

Navigation