多元线性回归方法在空气质量指数AQI分析中的应用
|
摘要
针对徐州雾霾情况,通过搜集徐州市2017年365天的日空气质量指数AQI数据,其9个相关影响变量数据(包括风力,机动车保有量,火电厂、炼钢厂、炼焦厂平均各排口每小时各主要污染物的排放量),在MATALB中采用多元线性回归方法建立了模型、参数估计和模型检验,并在已得模型的基础上剔除不显著的变量和样本异常值,经过两次改进,由九元线性模型简化为四元线性模型。通过拟合优度检验、显著性检验、多重共线性诊断和异常值残差诊断后,绘制出拟合对比图,验证了所得四元线性回归模型的准确性和实用性。
In view of the fog and haze situation in Xuzhou, through collecting the daily air quality index AQI data of 365 days in 2017 in Xuzhou City, nine relevant influential variables(including wind power, vehicle ownership, average discharge of major pollutants per hour at each outlet of thermal power plants, steelmaking plants and coking plants) have been collected, and multiple linear regression method has been used in MATALB to build models, estimate parameters and establish model tests, and insignificant variables and sample outliers have been eliminated on the basis of the obtained models. After two improvements, the 9-element linear model has been simplified to 4-element linear model. After goodness-of-fit test, significance test, multi-collinearity diagnosis and residual difference constant diagnosis, the fitting contrast chart has been drawn to verify the accuracy and practicability of the 4-element linear regression model.
In view of the fog and haze situation in Xuzhou, through collecting the daily air quality index AQI data of 365 days in 2017 in Xuzhou City, nine relevant influential variables(including wind power, vehicle ownership, average discharge of major pollutants per hour at each outlet of thermal power plants, steelmaking plants and coking plants) have been collected, and multiple linear regression method has been used in MATALB to build models, estimate parameters and establish model tests, and insignificant variables and sample outliers have been eliminated on the basis of the obtained models. After two improvements, the 9-element linear model has been simplified to 4-element linear model. After goodness-of-fit test, significance test, multi-collinearity diagnosis and residual difference constant diagnosis, the fitting contrast chart has been drawn to verify the accuracy and practicability of the 4-element linear regression model.
引文
[1] 顾为东.中国雾霾特殊形成机理研究[J].宏观经济研究,2014(6):3-7.
[2] 谢中华.MATLAB统计分析与应用:40个案例分析[M].北京:北京航空航天大学出版社,2015.
[3] 赵猛.基于数据挖掘技术的大气环境预测研究[D].北京:北京交通大学,2017.
[4] GANESH S S,MODALI S H,PALREDDY S R,et al.Forecasting air quality index using regression models:a case study on Delhi and Houston[J].2017 International Conference on Trends in Electronics and Informatics (ICEI),Tirunelveli,2017:248-254.
[5] 金江强,张怀相.改进多元回归分析在空气质量监测的应用[J].杭州电子科技大学学报(自然科学版),2016,36(1):41-45.
[6] 张明辉,赵铨铣,王妍.基于线性回归模型研究汽车消费与空气污染的关系[J].中国传媒大学学报(自然科学版),2015,22(4):16-20.
[7] 杨凌霄.济南市大气PM2.5污染特征、来源解析及其对能见度的影响[D].济南:山东大学,2008.
[8] 黄思,唐晓,徐文帅,等.利用多模式集合和多元线性回归改进北京PM10预报[J].环境科学学报,2015,35(1):56-64.
[9] VLACHOGIANNI A,KASSOMENOS P,KARPPINEN A,et al.Evaluation of a multiple regression model for the forecasting of the concentrations of NOx and PM10 in Athens and Helsinki[J].Science of the Total Environment,2011,409(8):1559-1571.
[10] 胡玉筱,段显明.基于高斯烟羽和多元线性回归模型的PM2.5扩散和预测研究[J].干旱区资源与环境,2015,29(6):86-92.
[11] 张燕杰,郑煜,姜波,等.哈尔滨市AQI与空气污染物的相关分析研究[J].环境科学与管理,2017,42(2):74-76.
[12] ELBAYOUMI M,RAMLI N A,FAIZAH F M Y N.Development and comparison of regression models and feedforward backpropagation neural network models to predict seasonal indoor PM2.5–10 and PM2.5 concentrations in naturally ventilated schools[J].Atmospheric Pollution Research,2015,6(6):1013-1023.
[13] 王惠文,孙晓丹.时序立体数据多元线性回归建模方法[J].系统工程,2009,27(11):87-90.
[14] AGIRRE-BASUKK E,IBARRA-BERAITEGI G,MADARIAGA I.Regression and multilayer perceptron-based models to forecast hourly O 3 and NO 2 levels in the Bilbao area[J].Environmental Modelling and Software,2004,21(4):430-446.
[2] 谢中华.MATLAB统计分析与应用:40个案例分析[M].北京:北京航空航天大学出版社,2015.
[3] 赵猛.基于数据挖掘技术的大气环境预测研究[D].北京:北京交通大学,2017.
[4] GANESH S S,MODALI S H,PALREDDY S R,et al.Forecasting air quality index using regression models:a case study on Delhi and Houston[J].2017 International Conference on Trends in Electronics and Informatics (ICEI),Tirunelveli,2017:248-254.
[5] 金江强,张怀相.改进多元回归分析在空气质量监测的应用[J].杭州电子科技大学学报(自然科学版),2016,36(1):41-45.
[6] 张明辉,赵铨铣,王妍.基于线性回归模型研究汽车消费与空气污染的关系[J].中国传媒大学学报(自然科学版),2015,22(4):16-20.
[7] 杨凌霄.济南市大气PM2.5污染特征、来源解析及其对能见度的影响[D].济南:山东大学,2008.
[8] 黄思,唐晓,徐文帅,等.利用多模式集合和多元线性回归改进北京PM10预报[J].环境科学学报,2015,35(1):56-64.
[9] VLACHOGIANNI A,KASSOMENOS P,KARPPINEN A,et al.Evaluation of a multiple regression model for the forecasting of the concentrations of NOx and PM10 in Athens and Helsinki[J].Science of the Total Environment,2011,409(8):1559-1571.
[10] 胡玉筱,段显明.基于高斯烟羽和多元线性回归模型的PM2.5扩散和预测研究[J].干旱区资源与环境,2015,29(6):86-92.
[11] 张燕杰,郑煜,姜波,等.哈尔滨市AQI与空气污染物的相关分析研究[J].环境科学与管理,2017,42(2):74-76.
[12] ELBAYOUMI M,RAMLI N A,FAIZAH F M Y N.Development and comparison of regression models and feedforward backpropagation neural network models to predict seasonal indoor PM2.5–10 and PM2.5 concentrations in naturally ventilated schools[J].Atmospheric Pollution Research,2015,6(6):1013-1023.
[13] 王惠文,孙晓丹.时序立体数据多元线性回归建模方法[J].系统工程,2009,27(11):87-90.
[14] AGIRRE-BASUKK E,IBARRA-BERAITEGI G,MADARIAGA I.Regression and multilayer perceptron-based models to forecast hourly O 3 and NO 2 levels in the Bilbao area[J].Environmental Modelling and Software,2004,21(4):430-446.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |