Forecasting of PM2.5 Concentration in Beijing Using Hybrid Deep Learning Framework Based on Attention Mechanism

被引:12
作者
Li, Dong [1 ,2 ,3 ,4 ]
Liu, Jiping [1 ,2 ]
Zhao, Yangyang [2 ]
机构
[1] Lanzhou Jiaotong Univ, Fac Geomat, Lanzhou 730070, Peoples R China
[2] Chinese Acad Surveying & Mapping, Beijing 100830, Peoples R China
[3] Natl Local Joint Engn Res Ctr Technol & Applicat, Lanzhou 730070, Peoples R China
[4] Gansu Prov Engn Lab Natl Geog State Monitoring, Lanzhou 730070, Peoples R China
来源
APPLIED SCIENCES-BASEL | 2022年 / 12卷 / 21期
关键词
air pollution; PM2; 5; concentration; deep learning; attention mechanism; SHORT-TERM-MEMORY; NEURAL-NETWORK; PREDICTION; MODEL;
D O I
10.3390/app122111155
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Air pollution has become a critical factor affecting the health of human beings. Forecasting the trend of air pollutants will be of considerable help to public health, including improving early-warning systems. The article designs a novel hybrid deep learning framework FPHFA (FPHFA is the abbreviation of the title of this paper) for PM2.5 concentration forecasting is proposed, which learns spatially correlated features and long-term dependencies of time series data related to PM2.5. Owing to the complex nonlinear dynamic and spatial features of pollutant data, the FPHFFA model combines multi-channel one-dimensional convolutional neural networks, bi-directional long short-term memory neural networks, and attention mechanisms for the first time. Multi-channel 1D CNNs are applied to capture trend features between some sites and overall spatial characteristics of PM2.5 concentration, Bi LSTMs are used to learn the temporal correlation of PM2.5 concentration, and the attention mechanism is used to focus more effective information at different moments. We carried out experimental evaluations using the Beijing dataset, and the outcomes show that our proposed model can effectively handle PM2.5 concentration prediction with satisfactory accuracy. For the prediction task from 1 to 12 h, our proposed prediction model performs well. The FPHFA also achieves satisfactory results for prediction tasks from 13 to 96 h.
引用
收藏
页数:23
相关论文
共 41 条
[1]   Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks [J].
Abdeljaber, Osama ;
Avci, Onur ;
Kiranyaz, Serkan ;
Gabbouj, Moncef ;
Inman, Daniel J. .
JOURNAL OF SOUND AND VIBRATION, 2017, 388 :154-170
[2]   Cost of economic growth: Air pollution and health expenditure [J].
Chen, Fanglin ;
Chen, Zhongfei .
SCIENCE OF THE TOTAL ENVIRONMENT, 2021, 755
[3]   A novel prediction model of PM2.5 mass concentration based on back propagation neural network algorithm [J].
Chen, Yegang ;
An, JianMei ;
Yanhan .
JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2019, 37 (03) :3175-3183
[4]  
Chung J, 2014, ARXIV
[5]   PM2.5 analog forecast and Kalman filter post-processing for the Community Multiscale Air Quality (CMAQ) model [J].
Djalalova, Irma ;
Delle Monache, Luca ;
Wilczak, James .
ATMOSPHERIC ENVIRONMENT, 2015, 108 :76-87
[6]   Deep Air Quality Forecasting Using Hybrid Deep Learning Framework [J].
Du, Shengdong ;
Li, Tianrui ;
Yang, Yan ;
Horng, Shi-Jinn .
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2021, 33 (06) :2412-2424
[7]  
[范竣翔 Fan Junxiang], 2017, [测绘科学, Science of Surveying and Mapping], V42, P76
[8]   Highly Efficient Picture-Based Prediction of PM2.5 Concentration [J].
Gu, Ke ;
Qiao, Junfei ;
Li, Xiaoli .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2019, 66 (04) :3176-3184
[9]   A Deep CNN-LSTM Model for Particulate Matter (PM2.5) Forecasting in Smart Cities [J].
Huang, Chiou-Jye ;
Kuo, Ping-Huan .
SENSORS, 2018, 18 (07)
[10]   Short-Term Electricity Prices Forecasting Using Functional Time Series Analysis [J].
Jan, Faheem ;
Shah, Ismail ;
Ali, Sajid .
ENERGIES, 2022, 15 (09)