基于BP神经网络对城市空气污染数据真实性的判别  

Discrimination of Authenticity of Urban Air Pollution Data Based on BP Neural Network

作　　者：陈璐璐[1] 陈富媛 杨鹏辉[2] 黄艳红[2] 陈梦倩[2] 

机构地区：[1]安徽财经大学金融学院,安徽蚌埠233030 [2]安徽财经大学统计与应用数学学院,安徽蚌埠233030

出　　处：《河北北方学院学报（自然科学版）》2017年第7期38-44,共7页Journal of Hebei North University:Natural Science Edition

基　　金：国家自然科学基金项目:"3-流猜想;Fulkerson-覆盖及相关问题"(11601001)

摘　　要：目的针对城市空气污染数据的真实性,收集并处理京津冀、长三角和珠三角地区的空气质量数据,判别数据的真实性,并给出提高空气污染数据真实性的建议。方法使用因子分析、多元统计等方法构建BP神经网络和多元线性回归模型,利用MATLAB、SPSS和excel等软件,分析空气质量数据的波动性,得到各个城市空气污染物数据的真实性结果。结果天津、沧州、上海、东莞等城市的空气污染数据较为真实,而保定、邢台、镇江、扬州、舟山,珠海等城市空气污染数据的不真实性严重。结论应当注重基层环境统计工作,加强基层和媒体对统计数据的关注,完善软件技术设施。Objective For the truthfulness of urban air pollution data collected and processed in Beijing-Tianjin-Hebei,Yangtze river delta and Pearl river delta,the authenticity of the data was judged,and some suggestions were given for the improvement of air pollution data authenticity.Method The factor analysis and multivariate statistical methods were used to construct the BP neural network and multivariate linear regression model.MATLAB,SPSS and EXCEL softwares were used to analyze the volatility of air quality data,obtaining the authenticity of the urban air pollution data results.Results The data of Tianjin,Cangzhou,Shanghai,Dongguan and other cities air pollution were real,but the air pollution data of Baoding,Xingtai,Zhenjiang,Yangzhou,Zhoushan,Zhuhai and other cities had serious inauthenticity.Conclusion Attention should be paid to the basic environment statistical work,the grass-roots and the media focus on statistics,and software technology facilities should be improved.

关 键 词：空气质量 BP神经网络 MATLAB 

分 类 号：Q948.116[生物学—植物学]