Article Details

Abstract

Based on the air quality data of 368 cities in China from 2014 to 2021, Deep Learning models (including SCINet, Informer, LSTM, GRU and RNN) are used to construct air quality prediction models. The platform is developed by using Qt Designer and Python, and combined with technologies such as ETL and data visualization to achieve the visualization of air quality prediction. The platform supports accurate prediction of the air quality index and analysis of the impact of pollution sources. Users can select their own regions and models for prediction and visualize the prediction performance. The platform is suitable for environmental protection, meteorological, scientific research institutions and the public, providing efficient and accurate air quality monitoring and prediction tools.

Keywords

PySide6; SCINet; Deep Learning

References

1. [1] Qin Yaochen, Xie Zhixiang, Li Yang Research progress on the impact of air pollution on residents' health [J]. Environmental Science, 2019, 40 (3): 1512-1520
2. [2] HUANG W，LI T，LIU J，et al. An Overview of Air Quality Analysis by Big Data Techniques: Monitoring, Forecasting, and Traceability [J].Information Fusion，2021，75：28-40.
3. [3] AWASTHI A，PATTNAYAK K C，DHIMAN G，et al. Artificial Intelligence for Air Quality Monitoring and Prediction [M].Boca Raton：CRC Press，2024.
4. [4] ZHANG Z，ZHANG S Q，CHEN C C，et al. A Systematic Survey of Air Quality Prediction Based on Deep Learning [J].Alexandria Engineering Journal，2024，93：128-141.
5. [5] LU Y F，WANG J Y，WANG D S，et al. Incorporating Temporal Multi-Head Self-Attention Convolutional Networks and LightGBM for Indoor Air Quality Prediction [J/OL].Applied Soft Computing，2024，157：111569（2024-04-03）.https://doi.org/10.1016/j.asoc.2024.111569.
6. [6] ZHANG M J，ZHANG Y J，ZHANG J G H，et al. Time Series Analysis of the Impact of Meteorological Conditions and Air Quality on the Number of Medical Visits for Hypertension in Haikou City, China [J/OL].Atmosphere，2024，15（3）：370（2024-03-18）.https://doi.org/10.3390/atmos15030370.
7. [7] LIU M，ZENG A，CHEN M，et al. Scinet: Time Series Modeling and Forecasting with Sample Convolution and Interaction [J].Advances in Neural Information Processing Systems，2022，35：5816-5828.
8. [8] Tianyuan, Sun Mengjue, Zhou Zhigao, etc A LSTM time series data prediction model based on information entropy [J]. Science and Technology Innovation and Application, 2024, 14 (7): 28-34
9. [9] Li Yi Research on RMB exchange rate prediction based on LSTM neural network [J]. China Price, 2021 (12): 20-22
10. [10] Lu Jixiang, Zhang Qipei, Yang Zhihong, etc Short term load forecasting method based on CNN-LSTM hybrid neural network model [J]. Power System Automation, 2019, 43 (8): 131-137
11. [11] Liang Hongtao, Liu Shuo, Du Junwei, etc A Review of the Application of Deep Learning in Time Series Prediction Research [J]. Computer Science and Exploration, 2023, 17 (6): 1285-1300
12. [12] Yi Lirong, Wang Shaoyu, Yin Lili, etc Time series data prediction of industrial sensors based on multivariate LSTM [J]. Intelligent Computers and Applications, 2018, 8 (5): 13-16
13. [13] Yao Jiajia Research on the Theory and Practice of Promoting Deep Learning among College Students through Peer Dialogue Feedback Strategies [D]. Hangzhou: Zhejiang University, 2021
14. [14] Fu En, Zhang Yinong, Yang Fan, etc Time series long-term prediction model based on frequency decomposition Transformer [J]. Manufacturing Automation, 2022, 44 (11): 177-181

Funding

National College Students' Innovation and Entrepreneurship Plan Project (S202310554004)

About the authors

Zhang Bin (2003-), male, Han ethnicity, from Hengyang, Hunan, undergraduate studies, research direction: air quality prediction; Corresponding author: Qin Yemei (1979-), female, Han ethnicity, from Xiangtan, Hunan Province, associate professor, PhD, research direction: intelligent information processing, time series analysis and prediction.