Abstract—In order to better grasp the change rule of PM2.5 concentration, this paper presents a prediction model of PM2.5 concentration based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN)- Permutation Entropy (PE)-Long Short-Term Memory (LSTM). The PM2.5 concentration time series is decomposed into several sub-sequences with obvious complexity differences by CEEMDAN-PE. Then, the LSTM prediction model is built by adding meteorological parameters to each different sub-sequence. The final results are got by adding the prediction results. The data of four monitoring stations in Tangshan City, Hebei Province is used to implement simulation experiment. Experiment results confirm that the proposed prediction model compared with other combined and single forecasting methods, and shows a high prediction precision, and good universality, which provided effective technical support for pre-control of air pollution.
Index Terms—PM2.5, concentration prediction, ensemble empirical mode decomposition, permutation entropy, time series.
Tao Liang and Gaofeng Xie are with the Hebei University of Technology, college of Artificial Intelligence, Tianjin, 300130, China (Corresponding author; e-mail: LiangTao@hebut.edu.cn, email@example.com).
Dabin Mi, Wen Jiang, and Guilin Xu are with the Jointo Energy Investment Co., Ltd. Hebei, Shijiazhuang, 050000, China (e-mail: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org).
Cite: Tao Liang, Gaofeng Xie, Dabin Mi, Wen Jiang, and Guilin Xu, "PM2.5 Concentration Forecasting Based on Data Preprocessing Strategy and LSTM Neural Network," International Journal of Machine Learning and Computing vol. 10, no. 6, pp. 729-734, 2020.Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).