摘要
本文先分别运用灰色预测GM(1,1)模型和BP神经网络模型分别对我国所有一线城市和新一线城市的空气质量状况进行了一个预测,并对比预测结果,从更优的预测结果中挑出了到2030年不能达到我国十三五计划规定的空气质量标准的六座城市——北京,武汉,天津,重庆,西安,长沙。接着对上述六座城市的空气质量状况与各空气污染物进行了灰色关联度分析,找出了各城市的主要空气污染物。针对每个城市的主要污染物综合运用了主成分分析和多元线性回归等方法找出了各城市主要污染物的主要成因,对各城市的环境治理提供了重要的参考。
In this paper, a gray prediction GM (1,1) model and a BP neural network model are respectively used to predict the air quality status of all first-tier cities and new-tier cities in China. From the better predictions, six cities that cannot meet the air quality standards set by the 13th Five-Year Plan of China by 2030. They are Beijing, Wuhan, Tianjin, Chongqing, Xi’an and Changsha. Then, the air quality status and air pollutants of the above six cities were analyzed by gray relational analysis. Therefore, we identify the main air pollutants in each city. Finally, the main pollutants are integrated using principal component analysis and multiple linear regression to find out the main causes of the major pollutants in each city and provide an important reference for the en-vironmental governance in each city.
出处
《统计学与应用》
2018年第2期154-163,共10页
Statistical and Application
基金
华中师范大学2017年度大学生创新创业训练计划B类项目
上海市青锐环境数据有限公司科研立项资助。