基于贝叶斯的大数据模型在大气污染成因分析中的应用

大气与环境光学学报 ›› 2023, Vol. 18 ›› Issue (3): 227-234.

基于贝叶斯的大数据模型在大气污染成因分析中的应用

王莉君 *, 周玉 , 万丽娟 , 程亮亮

合肥师范学院电子信息与电气工程学院

收稿日期:2022-02-08 修回日期:2022-04-17 出版日期:2023-05-28 发布日期:2023-05-28
通讯作者: E-mail: lijwang@qq.com E-mail:120639919@qq.com
作者简介:王莉君 (1976– ), 女, 安徽亳州人, 硕士, 讲师, 主要从事信息处理、传感技术、大数据技术研究与应用。E-mail: lijwang@qq.com
基金资助:
安徽省教育厅省级教学示范课程建设项目 (2020SJJXSFK1991), 安徽省质量工程项目 (2020xsxxkc380), 电子信息仿真设计安徽省重点实验室项目 (2020ZDSYSYB05)

Application of the Bayesian-based big data model in the analysis of the source of air pollution

WANG Lijun *, ZHOU Yu , WAN Lijuan , CHENG Liangliang

School of Electronic Information and Electrical Engineering, Hefei Normal University

Received:2022-02-08 Revised:2022-04-17 Published:2023-05-28 Online:2023-05-28

摘要/Abstract

摘要： 提出了一种基于多维高斯贝叶斯分类算法的复杂系统影响因素的分析方法，并利用大数据方法建立了不同 PM2.5范围的分类模型，结合马氏距离开展了影响因素的关键性分析。以合肥市2013―2018 年间的天气与空气质量数据为基础，筛选了PM10、SO2、NO2、CO、O3等8 个大气污染的关键影响因素，采用散点矩阵对PM2.5与这些影响因素的相关性进行了分析。利用高斯贝叶斯分类器建立了基于8 个主要参量的PM2.5大气污染分析模型，研究发现， PM2.5与CO 浓度具有较强的正相关性，对NO2具有选择性，与O3具有负相关性，而CO与SO2对PM2.5的产生存在某种竞争机制。

关键词: 大数据技术, 高斯贝叶斯模型, 影响因素, PM2.5, 相关分析

Abstract: A method for analyzing influencing factors of complex systems based on multi-dimensional Gaussian Bayesian classification algorithm is proposed, classification models for diverse range of PM2.5 are established, and then the analysis of the key influencing factors on complex systems is carried out in combination with Mahalanobis distance. Based on the weather and air quality data of Hefei City from 2013 to 2018, 8 main influencing factors for PM2.5, such as PM10, SO2, NO2, CO, O3 and so on, are screened out, and then the correlation between PM2.5 and the influencing factors is analyzed using scatter matrix. The PM2.5 analysis model based on Gaussian Bayesian classifier is established with these data. It is found that PM2.5 has a strong positive correlation with CO concentration, is selective to NO2, and has a negative correlation with O3. As for CO and SO2, a certain competitive mechanism between the two factors in the production of PM2.5 is observed.

Key words: big data technology, Gaussian Bayesian model, influencing factors, PM2.5, correlation analysis

中图分类号:

TP391.4
X511

王莉君, 周玉, 万丽娟, 程亮亮. 基于贝叶斯的大数据模型在大气污染成因分析中的应用[J]. 大气与环境光学学报, 2023, 18(3): 227-234.

WANG Lijun , ZHOU Yu , WAN Lijuan , CHENG Liangliang. Application of the Bayesian-based big data model in the analysis of the source of air pollution[J]. Journal of Atmospheric and Environmental Optics, 2023, 18(3): 227-234.

参考文献 20

[1]	General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Ministry of
	Environmental Protection of the People's Republic of China. Ambient air quality standard: GB 3095—2012 [S]. Beijing: China
	Environmental Science Press, 2012.
	国家质量监督检验检疫总局, 环境保护部. 环境空气质量标准: GB 3095—2012 [S]. 北京: 中国环境科学出版社, 2012.
[2]	Luo Y P, Chen T, Ding H, et al. Characteristics analysis of PM2.5 and NO2 pollution at Beijing traffic stations during the
	Spring Festival from 2014 to 2018 [J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2020, 59(1): 50-63.
	罗银萍, 陈同, 丁卉, 等. 2014―2018 年春节期间北京交通站PM2.5及NO2污染特征分析 [J]. 中山大学学报 (自然科学
版)	, 2020, 59(1): 50-63.
[3]	Zhou Y M, Zhao X Y. Correlation analysis between PM2.5 concentration and meteorological factors in Beijing area [J]. Acta
	Scientiarum Naturalium Universitatis Pekinensis, 2017, 53(1): 111-124.
	周一敏, 赵昕奕. 北京地区PM2.5浓度与气象要素的相关分析 [J]. 北京大学学报 (自然科学版), 2017, 53(1): 111-124.
[4]	Huang L K, Xu Y Y, Wang G Z, et al. Study on pollution characteristics and toxic effects of fine particulate matter in Harbin,
	China [J]. China Environmental Science, 2019, 39(12): 5326-5332.
	黄丽坤, 徐媛媛, 王广智, 等. 哈尔滨市大气细颗粒物污染特征及其毒性效应 [J]. 中国环境科学, 2019, 39(12): 5326-
	5332.
[5]	Liu K K, Zhang H, Liu G J. Elemental composition characteristics of PM2.5 and PM10, and heavy pollution analysis in Hefei
[J]	Environmental Science, 2019, 40(8): 3415-3420.
	刘可可, 张红, 刘桂建. 合肥市PM2.5和PM10中元素组成特征及重污染成因分析 [J]. 环境科学, 2019, 40(8): 3415-3420.
[6]	Zhang J Y, Tian L, Li S M, et al. Relationship between ambient PM2.5 exposure and outpatient number for respiratory
	diseases in Chaoyang District, Beijing, China: A time series analysis [J]. Journal of Environmental Hygiene, 2020, 10(4):
36	7-371.
	张金艳, 田霖, 李书明, 等. 北京市朝阳区大气PM2.5暴露与医院呼吸系统疾病门诊量关系的时间序列分析 [J]. 环境卫
	生学杂志, 2020, 10(4): 367-371.
[7]	Li W, Hu H, Xu F C, et al. Studies on big data analysis technology used in environmental sciences [J]. Environmental
	Monitoring in China, 2015, 31(3): 118-122.
	李蔚, 胡昊, 徐富春, 等. 大数据解析技术在大气环境监测中的应用研究 [J]. 中国环境监测, 2015, 31(3): 118-122.
[8]	Zhu J X, Chen X D, Lü B L, et al. Smart failure/fault diagnosis and influence analysis for mechanical equipment with
	multivariate Gaussian Bayesian method [J]. Journal of Mechanical Engineering, 2020, 56(4): 35-41.
	朱建新, 陈学东, 吕宝林, 等. 基于多维高斯贝叶斯的机械设备失效/故障智能诊断及参数影响分析 [J]. 机械工程学报,
20	20, 56(4): 35-41.
[9]	Past-weather website. Historical weather query of Hefei City [OL]. http://www.tianqihoubao.com/weather/top/hefei.html.
	天气后报网站. 合肥历史天气查询 [OL]. http://www.tianqihoubao.com/weather/top/hefei.html.
[10]	McKinney W. Python for Data Analysis [M]. Sebastopol, CA: O'Reilly, 2012.
[11]	Zhang J, Yuan M, Liu Y Y. The Bayes classifier based on the normal mixture model and its application [J]. Acta
	Mathematicae Applicatae Sinica, 2020, 43(4): 742-755.
	张婧, 袁敏, 刘妍岩. 基于正态混合模型的贝叶斯分类方法及其应用 [J]. 应用数学学报, 2020, 43(4): 742-755.
[12]	Ministry of Environmental Protection of the People's Republic of China. Technical regulation on ambient air quality index (on
	trial): HJ 633—2012 [S]. Beijing: China Environmental Science Press, 2012.
	环境保护部. 环境空气质量指数 (AQI) 技术规定: HJ 633—2012 [S]. 北京: 中国环境科学出版社, 2012.
[13]	Han F R, Lu X, Wu T Y, et al. Characteristics of air pollution in Jinhua City and its relationship with meteorological
	conditions [J]. Journal of Zhejiang Meteorology, 2021, 42(4): 13-18.
	韩芙蓉, 鹿翔, 吴天贻, 等. 金华市大气污染特征及与气象条件的关系分析 [J]. 浙江气象, 2021, 42(4): 13-18.
[14]	Liu X H. High-Dimensional Data Analysis Based on Scatter Plot Classification [D]. Tianjin: Tianjin University, 2019.
	刘训瀚. 基于散点图模式探索的高维数据分析 [D]. 天津: 天津大学, 2019.
[15]	Chen X R. Research on characteristics of air pollutant emission changes in eco-city based on big data analysis [J].
	Environmental Science and Management, 2022, 47(1): 60-63.
	陈相如. 基于大数据分析的生态城市大气污染物排放变化特征研究 [J]. 环境科学与管理, 2022, 47(1): 60-63.
[16]	Xu D, Zhang Q, Fan G Q, et al. Characteristics of particles pollution based on lidar network in Hangzhou and surrounding
	areas [J]. Journal of Atmospheric and Environmental Optics, 2019, 14(3): 171-178.
	徐达, 张全, 范广强, 等. 基于激光雷达观测网的杭州及周边地区颗粒物污染特征研究 [J]. 大气与环境光学学报,
20	19, 14(3): 171-178.