1
|
Zhou Z, Qiu C, Zhang Y. A comparative analysis of linear regression, neural networks and random forest regression for predicting air ozone employing soft sensor models. Sci Rep 2023; 13:22420. [PMID: 38104205 PMCID: PMC10725498 DOI: 10.1038/s41598-023-49899-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/30/2023] [Accepted: 12/13/2023] [Indexed: 12/19/2023] Open
Abstract
The proposed methodology presents a comprehensive analysis of soft sensor modeling techniques for air ozone prediction. We compare the performance of three different modeling techniques: LR (linear regression), NN (neural networks), and RFR (random forest regression). Additionally, we evaluate the impact of different variable sets on prediction performance. Our findings indicate that neural network models, particularly the RNN (recurrent neural networks), outperform the other modeling techniques in terms of prediction accuracy. The proposed methodology evaluates the impact of different variable sets on prediction performance, finding that variable set E demonstrates exceptional performance and achieves the highest average prediction accuracy among various software sensor models. In comparing variable set E and A, B, C, D, it is observed that the inclusion of an additional input feature, PM10, in the latter sets does not improve overall performance, potentially due to multicollinearity between PM10 and PM2.5 variables. The proposed methodology provides valuable insights into soft sensor modeling for air ozone prediction.Among the 72 sensors, sensor NNR[Y]C outperforms all other evaluated sensors, demonstrating exceptional predictive performance with an impressive R2 of 0.8902, low RMSE of 24.91, and remarkable MAE of 19.16. With a prediction accuracy of 81.44%, sensor NNR[Y]C is reliable and suitable for various technological applications.
Collapse
Affiliation(s)
- Zheng Zhou
- Department of Material and Environmental Engineering, Chengdu Technological University, Chengdu, China
| | - Cheng Qiu
- Department of Material and Environmental Engineering, Chengdu Technological University, Chengdu, China.
| | - Yufan Zhang
- Department of Material and Environmental Engineering, Chengdu Technological University, Chengdu, China
| |
Collapse
|
2
|
Kuerban M, Waili Y, Fan F, Liu Y, Qin W, Dore AJ, Peng J, Xu W, Zhang F. Spatio-temporal patterns of air pollution in China from 2015 to 2018 and implications for health risks. ENVIRONMENTAL POLLUTION (BARKING, ESSEX : 1987) 2020; 258:113659. [PMID: 31806463 DOI: 10.1016/j.envpol.2019.113659] [Citation(s) in RCA: 69] [Impact Index Per Article: 17.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 05/25/2019] [Revised: 11/19/2019] [Accepted: 11/20/2019] [Indexed: 06/10/2023]
Abstract
China has been seriously affected by particulate matter (PM) and gaseous pollutants in the atmosphere. In this study, we systematically analyse the spatio-temporal patterns of PM2.5, PM10, SO2, CO, NO2, and O3 and the associated health risks, using data collected from 1498 national air quality monitoring sites. An analysis of the averaged data from all the sites indicated that, from 2015 to 2018, annual mean concentrations of PM2.5, PM10, SO2 and CO declined by 3.2 μg m-3, 3.7 μg m-3, 3.9 μg m-3, and 0.1 mg m-3, respectively. In contrast, those of NO2 and O3 increased at rates of 0.4 and 3.1 μg m-3, respectively. Except for O3, the annual mean concentrations of all pollutants were generally the highest in North China and lowest in the Tibetan Plateau. The concentrations were generally higher in the north of the country than in the south. In all regions of China, the pollutant concentrations were the highest in winter and lowest in summer, except for O3, which showed an opposite seasonal pattern. Overall, the seasonal mean concentrations of all the pollutants (except for O3) significantly decreased between the same seasons in 2018 and 2015, whereas the seasonal mean O3 concentrations generally significantly increased, and/or remained at stable levels in all four seasons except for winter. Diurnal variations of all pollutants (except for O3) exhibited a bimodal pattern with peaks between 8:00 and 11:00 a.m. and 9:00 and 12:00 p.m., whereas O3 exhibited a unimodal pattern with maximum values between 5:00 and 7:00 p.m. No significant differences in the daily mean concentrations of all pollutants were found between weekdays and weekends in all regions, except for PM2.5 and PM10 in Northeast China. In Northwest China and Southeast China, PM2.5 showed stronger correlations with NO2 relative to SO2, suggesting that NOx emission control may be more effective than SO2 emission control for alleviating PM2.5 formation. Compared with 2015, the total PM2.5-attributable mortality, number of respiratory and cardiovascular diseases, and incidence of chronic bronchitis decreased overall by 23.4%-26.9% in 2018. In contrast, for O3-attributable deaths, there was an increase of 18.9%. Our study not only improves the understanding of the spatial and temporal patterns of air pollutants in China, but also highlights that synchronous control of PM2.5 and O3 pollution should be implemented to achieve dual benefits in protecting human health.
Collapse
Affiliation(s)
- Mireadili Kuerban
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China
| | - Yizaitiguli Waili
- College of Resources and Environmental Science, Xinjiang University, Urumqi, 830046, China
| | - Fan Fan
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China
| | - Ye Liu
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China
| | - Wei Qin
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China
| | - Anthony J Dore
- Centre for Ecology and Hydrology, Edinburgh, Bush Estate, Penicuik, Midlothian, EH26 0QB, UK
| | - Jingjing Peng
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China
| | - Wen Xu
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China.
| | - Fusuo Zhang
- College of Resources and Environmental Sciences, National Academy of Agriculture Green Development, Key Laboratory of Plant-Soil Interactions of MOE, China Agricultural University, Beijing, 100193, China
| |
Collapse
|