Username   Password       Forgot your password?  Forgot your username? 

 

Novel Convolution and LSTM Model for Forecasting PM2.5 Concentration

Volume 15, Number 6, June 2019, pp. 1528-1537
DOI: 10.23940/ijpe.19.06.p4.15281537

Wenfang Zhaoa,b, Yong Zhouc, and Wei Tangc

aInstitute of Urban Meteorology, China Meteorological Administration, Beijing, 100089, China
bBeijing Meteorological Information Center, Beijing Meteorological Bureau, Beijing, 100089, China
cDevelopment and Research Center, China Meteorological Administration, Beijing, 100081, China

(Submitted on March 20, 2019; Revised on April 14, 2019; Accepted on June 13, 2019)

Abstract:

Higher levels of PM2.5 concentration are becoming the leading cause of hazy days in China. However, studies have shown that the variations of PM2.5 involve complicated physical and chemical processes, which make their accurate predictions challenging. Meanwhile, the forecast results from numerical models frequently deviate from observation values. The deep learning method is a good substitute for the prediction of mass time series data in the field of meteorology. In the present study, a framework for PM2.5 concentration prediction is presented based on a three-dimensional convolutional neural network (3DCNN) and long short term memory neural network (LSTM). Using preprocessing, correlation analysis, feature extraction, and transformation, spatiotemporal sequence data was generated. In the spatiotemporal feature extraction phase, 3DCNN was used to extract high-level spatial features, and LSTM was used to extract temporal features. In the prediction phase, full connect (FC) was used to combine spatial and temporal features. To examine the efficacy of the proposed model, the PM2.5 concentration data, meteorological observation data, and grid dataset collected at ten observation stations in the Beijing Meteorological Bureau (BMB) were used. After the performance evaluation was compared with several methods including this proposed model, support vector machine (SVM), and the existing PM2.5 forecast system in BMB, root mean square errors (RMSE) and mean absolute errors (MAE) were chosen as evaluation indicators. The experimental results showed that the proposed model performed the best, the minimum MAE value was 3.24μg/m3, and the minimum RMSE value was 13.56μg/m3 over the ten stations. In addition, the proposed model overcame the underestimation produced by the existing PM2.5 forecast system in BMB and demonstrated superior performance for different time lengths over a 24-hour period. The results also confirmed the effectiveness of the deep learning method in the prediction of PM2.5 concentration.

 

References: 33

  1. Y. J. Kaufman, T. Didier, and B. Olivier, “A Satellite View of Aerosols in the Climate System,” Nature, Vol. 419, pp. 215-233, 2002
  2. C. K. Chan and X. H. Yao, “Air Pollution in Mega Cities in China,” Atmospheric Environment, Vol. 42, No. 11, pp. 1-42, 2008
  3. X. Y. Zhang and H. B. Hu, “Risk Assessment of Exposure to PM2.5 in Beijing using Multi-Source Data,” Acta Scientiarum Naturalium Universitatis Pekinensis, Vol. 54, No. 1, pp. 1103-1113, 2018
  4. Q. Jin, X. Fang, B. Wen, and A. Shan, “Spatiotemporal Variations of PM2.5 Emission in China from 2005 to 2014,” Chemosphere, Vol. 183, pp. 429-436, 2014
  5. A. V. Donkelaar, R. V. Martin, M. Brauer, and B. L. Boys, “Use of Satellite Observations for Long-Term Exposure Assessment of Local Concentrations of Fine Particulate,” Environmental Health Perspectives, Vol. 123, No. 2, pp. 135-143, 2015
  6. C. L. Fang, Z. B. Wang, and G. Xu, “Spatial-Temporal Characteristics of PM2.5 in China: A City Level Perspective Analysis,” Journal of Geographical Sciences, Vol. 26, pp. 1519-1532, 2016
  7. X. Su, W. Gough, and Q. Shen, “Correlation of PM2.5 and Meteorological Variables in Ontario Cities: Statistical Downscaling Method Coupled with Artificial Neural Network,” in Proceedings of the 24th International Conference on Modeling , Monitoring and Management of Air Pollution, pp. 215-226, 2016
  8. R. Chen, X. Wang, X. Meng, J. Hua, Z. J. Zhou, B. H. Chen, et al., “Communicating Air Pollution-Related Health Risks to the Public: An Application of the Air Quality Health Index in Shanghai, China,” Environment International, Vol. 1, No. 5, pp. 168-173, 2013
  9. J. Chen, J. Lu, J. C. Avise, J. A. DaMass, M. J. Kleeman, and A. P. Kaduwela, “Seasonal Modeling of PM2.5 in California's San Joaquin Valley,” Atmospheric Environment, Vol. 92, pp. 182-190, 2014
  10. Q. Z. Wu, W. S. Xu, A. Shi, Y. Li, X. J. Zhao, Z. F. Wang, et al., “Air Quality Forecast of PMl0 in Beijing with Community Multi-Scale Air Quality Modeling (CMAQ) System: Emission and Improvement,” Geoscience Model, Vol. 12, No. 7, pp. 2243-2259, 2014
  11. L. Chen, D. M. Wu, and Q. Chen, “Prediction of Air Pollution based on Wavelet Analysis and Support Vector Machine,” Journal of XI'AN University of Science and Technology, Vol. 30, No. 6, pp. 726-730, 2010
  12. G. -Q. Zhou, Y. Xie, J. -B. Wu, Z. -Q. Yu, L. -Y. Chang, and W. Gao, “WRF-Chem based PM2.5 Forecast and Bias Analysis over the East China Region,” China Environment Science, Vol. 36, No. 8, pp. 2251-2259, 2016
  13. G. Yi and M. G. Zhang, “Numerical Simulation of a Heavy Fog–Haze Episode over the North China Plain in January 2013,” Climatic and Environmental Research, Vol. 19, No. 2, pp. 40-152, 2014
  14. H. Z. De, H. X. Yun, and H. X. Yong, “Haze Forecast based on Time Series Analysis and Kalman Filtering,” Journal of Computer Applications, Vol. 37, No. 11, pp. 3311-3316, 2017
  15. J. Z. Xiu, X. Jing, and Z. Z. Yin, “Beijing Regional Environmental Meteorology Prediction System and its Performance Test of PM2.5 Concentration,” Journal of Application Meteorological Science, Vol. 27, No. 2, pp. 160-172, 2016
  16. T. J. Wang, F. Jiang, J. J. Deng, Y. Shen, Q. Y. Fu, Q. Wang, et al., “Urban Air Quality and Regional Haze Weather Forecast for Yangtze River Delta,” Atmospheric Environment, Vol. 58, No. 15, pp. 70-83, 2012
  17. J. Liu, P. Yang, W. S. Lv, A. Liu, and J. X. Liu, “Prediction Model of PM2. 5 Mass Concentrations based on Fuzzy Time Series and Support Vector Machine,” Journal of University of Science and Technology Beijing, Vol. 36, No. 12, pp. 1694-1703, 2014
  18. L. Li, L. Ma, J. F. He, D. G. Shao, S. L. Yi, Y. Xiang, et al., “PM2.5 Concentration Prediction Model of Least Squares Support Vector Machine based on Feature Vector,” Journal of Computer Applications, Vol. 34, No. 8, pp. 2212-2216, 2014
  19. J. Pan, H. Z. Wang, H. Gao, W. X. Zhao, H. X. Huo, and H. R. Dong, “Paradise Pointer: A Sightseeing Scenes Images Search Engine based on Big Data Processing,” in Proceedings of the International Conference of Young Computer Scientists, pp. 448-452, Harbin, China, January 2015
  20. D. Mishra, P. Goyal, and A. Upsfhysy, “Artificial Intelligence based Approach to Forecast PM2.5 During Haze Episodes: A Case Study of Delhi, India,” Atmospheric Environment, Vol. 120, pp. 239-248, 2015
  21. G. O. Philip, L. Gunnar, and D. Ottfried, “Relationship Between Rice Yield and Climate Variables in Southwest Nigeria using Multiple Linear Regression and Support Vector Machine Analysis,” International Journal of Biometeorology, Vol. 62, No. 3, pp. 459-469, 2018
  22. X. J. Shi, Z. R. Chen, H. Wang, D. Y. Yeung, W. K. Wong, and W. C. Woo, “Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting,” in Proceedings of the International Conference of Language Process, pp. 1-12, Los Angeles, USA, August 2015
  23. 23. J. C. Zhao, F. Deng, Y. Y. Cai, and J. Chen, “Long Short-Term Memory-Fully Connected (LSTM-FC) Neural Network for PM2.5 Concentration Prediction,” Chemosphere Volume, Vol. 220, pp. 486-492, 2019
  24. C. Vidushi, D. Anand, K. Vijayan, et al., “Time Series based LSTM Model to Predict Air Pollutant's Concentration for Prominent Cities in India,” in Proceedings of 1st International Workshop on Utility-Driven Mining, pp. 1-9, London, United Kingdom, August 2018
  25. S. A. Weber, T. Z. Insaf, E. S. Hall, T. O. Talbot, and A. K. Huff, “Assessing the Impact of Fine Particulate Matter (PM2.5) on Respiratory-Ardiovascular Chronic Diseases in the New York City Metropolitan Area using Hierarchical Bayesian Model Estimates,” Environment Research, Vol. 151, pp. 399-409, 2016
  26. J. H. Chou and H. K. Ping, “A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities,” Sensors, Vol. 18, No. 2, pp. 2220-2241, 2018
  27. C. C. Wen, S. F. Liu, X. J. Yao, L. Peng, X. Li, Y. Hu, et al., “A Novel Spatiotemporal Convolutional Long Short-Term Neural Network for Air Pollution Prediction,” Science of the Total Environment, Vol. 654, pp. 1091-1099, 2019
  28. W. S. Ping, W. C. Jia, and W. H. Jin, “Adaptive Deep Learning-based Air Quality Prediction Model using the Most Relevant Spatial-Temporal Relations,” IEEE Access, Vol. 6, No. 38, pp. 186-382, 2018
  29. D. Chu, Y. J. Kaufman, G. Zibordi, J. Chern, J. T. Mao, C. C. Li, et al., “Global Monitoring of Air Pollution Over Land from the Earth Observing System-Terra Moderate Resolution Imaging Spectro-Radiometer (MODIS),” Journal of Geophysical Research, Vol. 21, No. 108, pp. 4661-4667, 2003
  30. P. E. Saide, G. R. Carmichael, S. N. Spak, L. Gallardo, A. E. Osses, M. A. Mena-Carrasco, et al., “Forecasting Urban PM10 and PM2.5 Pollution Episodes in Very Stable Nocturnal Conditions and Complex Terrain using WRF-Chem CO Tracer Model,” Atmospheric Environment, Vol. 45, No. 16, pp. 2769-2780, 2011
  31. T. Du, L. Bourdev, R. Fergus, L. Torresani, and M. Paluri, “Learning Spatiotemporal Features with 3D Convolutional Networks,” in Proceedings of the International Conference of Computer Vision ICCV, pp. 4489-4497, Santiago, Chile, Dec. 2015
  32. Y. Chen, R. Shi, S. Shu, and W. Gao, “Ensemble and Enhanced PM10 Concentration Forecast Model based on Stepwise Regression and Wavelet Analysis,” Atmospheric Environment, Vol. 74, pp. 346-359, 2013
  33. Z. Liang, G. Z. Guan, and Y. S. Pei, “Learning Spatiotemporal Features using 3DCNN and Convolutional LSTM for Gesture Recognition,” in Proceedings of the International Conference of Computer Vision, pp. 1-9, Venice, Italy, October 2017

 

Please note : You will need Adobe Acrobat viewer to view the full articles.Get Free Adobe Reader

 
This site uses encryption for transmitting your passwords. ratmilwebsolutions.com