[1] Al Hanai A H, Antkiewicz D S, Hemming J D C et al. Seasonal variations in the oxidative stress and inflammatory potential of PM2.5 in Tehran using an alveolar macrophage model; the role of chemical composition and sources[J]. Environment International, 123, 417-427(2019).

[2] Laden F, Schwartz J, Speizer F E et al. Reduction in fine particulate air pollution and mortality: Extended follow-up of the Harvard six cities study[J]. American Journal of Respiratory and Critical Care Medicine, 173, 667-672(2006).

[3] Evans J, van Donkelaar A, Martin R V et al. Estimates of global mortality attributable to particulate air pollution using satellite imagery[J]. Environmental Research, 120, 33-42(2013).

[4] Rojas-Rueda D, de Nazelle A, Teixidó O et al. Health impact assessment of increasing public transport and cycling use in Barcelona: A morbidity and burden of disease approach[J]. Preventive Medicine, 57, 573-579(2013).

[5] Chai J, Peng C. The causes of haze formation and its effects on human health[J]. World Abstract of The Latest Medical Information, 17, 128-129(2017).

[6] Zhou Y, Tan O, Zhang H et al. Research progress of fine particulate matter PM2.5 in the atmospheric environment[J]. China Resources Comprehensive Utilization, 39, 90-93(2021).

[7] Yang C L. A study on the relationship between urban economy and PM2.5[J]. Science and Technology Wind, 133(2019).

[8] Zhao X, Hou L L, Wang X L et al. Simulation of spatial divergence of PM2.5 and PM10 concentrations in Beijing in 2019 based on LUR model[J]. Journal of Environmental Science, 40, 4060-4069(2020).

[9] Wang D D, Qin C. Regional PM2.5 spatio-temporal regression modeling and prediction[J]. China Environmental Monitoring, 35, 107-113(2019).

[10] Shi Y S. Application of support vector machine in PM2.5 prediction research[J]. Cooperative Economics and Technology, 48-50(2022).

[11] Wojciech Z, Ilya S, Oriol V. Recurrent neural network regularization[Z/OL]. https://arxiv.org/abs/1409.2329v5

[12] Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural Computation, 9, 1735-1780(1997).

[13] Chung J, Gulcehre C, Cho K H et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[J]. Eprint Arxiv(2014).

[14] Huang J, Zhang F, Du Z H et al. PM2.5 hourly concentration prediction based on RNN-CNN integrated deep learning model[J]. Journal of Zhejiang University (Science Edition), 46, 370-379(2019).

[15] Duan D G, Zhao Z D, Liang S H et al. LSTM-based PM2.5 concentration prediction model[J]. Computer Measurement and Control, 27, 215-219(2019).

[16] Zhang Z, Zhang S Q, Zhao X M et al. Temporal difference-based graph transformer networks for air quality PM2.5 prediction: A case study in China[J]. Frontiers in Environmental Science, 10, 924986(2022).

[17] Sutskever I, Vinyals O, Le Q V. Sequence to sequence learning with neural networks[J]. Advances In neural Information Processing Systems, 3104-3112(2014).

[18] Taylor S J, Letham B. Forecasting at scale[J]. The American Statistician, 72, 37-45(2018).