卢月明,王亮,仇阿根,赵阳阳,张用川

• 1:中国测绘科学研究院
• 2:辽宁工程技术大学
• 3:武汉大学资源与环境科学学院

摘要(Abstract)：

针对传统空间插值方法对影响PM2.5的插值因素考虑不全面和局部加权线性回归模型中近邻个数选择困难等问题,该文基于局部加权线性回归模型提出了一种引入正则化项的空间插值方法。以北京市3个月的PM2.5数据为例,选取SO_2、NO_2、O_3、CO作为观测指标,通过正则化进行权重系数修正、L曲线法确定正则化系数,提高了该插值模型的稳定性与自适应性。交叉验证结果显示,本方法相对于普通克里金法,3个月的平均绝对误差(MAE)与均方根误差(RMSE)分别降低28.44%、26.25%;相对于反距离加权插值法的MAE、RMSE分别降低18.07%、17.02%。研究结果表明,基于局部加权线性回归模型的PM2.5空间插值相对于传统方法有一定提升。

关键词(KeyWords)： PM2.5;空间插值;局部加权线性回归;克里金法;反距离加权插值法

Abstract:

Keywords:

基金项目(Foundation): 中国测绘科学研究院基本科研业务费项目(7771614);; 测绘新技术系统开发与示范应用项目(2016KJ0104)

作者(Author): 卢月明,王亮,仇阿根,赵阳阳,张用川

DOI: 10.16251/j.cnki.1009-2307.2018.11.013

参考文献(References)：

• [1] FANN N,LAMSON A D,ANENBERG S,et al.Estimating the national public health burden associated with exposure to ambient PM2.5and ozone[J].Risk Analysis,2012,32(1):81-95.
• [2]刘杰,杨鹏,吕文生.北京大气颗粒物污染特征及空间分布插值分析[J].北京科技大学学报,2014,36(9):1269-1279.(LIU Jie,YANG Peng,LV Wensheng.Pollution characteristics of particulate matter and interpolation analysis of its spatial distribution in the Beijing area[J].Journal of University of Science and Technology Beijing,2014,36(9):1269-1279.)
• [3]李杰,翟亮,桑会勇,等.PM2.5浓度插值中不同空间插值方法对比[J].测绘科学,2016,41(4):50-54,101.(LI Jie,ZHAI Liang,SANG Huiyong,et al.Comparison of different spatial interpolation methods for PM2.5[J].Science of Surveying and Mapping,2016,41(4):50-54,101.)
• [4]樊子德,李佳霖,邓敏.顾及多因素影响的自适应反距离加权插值方法[J].武汉大学学报(信息科学版),2016,41(6):842-847.(FAN Zide,LI Jialin,DENG Min.An adaptive inverse-distance weighting spatial interpolation method with the consideration of multiple factors[J].Geomatics and Information Science of Wuhan University,2016,41(6):842-847.)
• [5]徐武平,邱峰,徐爱萍.空间数据插值的自动化方法研究[J].武汉大学学报(信息科学版),2016,41(4):498-502.(XU Wuping,QIU Feng,XU Aiping.Automatic method of Kriging interpolation of spatial data[J].Geomatics and Information Science of Wuhan University,2016,41(4):498-502.)
• [6]段平,盛业华,张思阳,等.顾及异向性的局部径向基函数三维空间插值[J].武汉大学学报(信息科学版),2015,40(5):632-637.(DUAN Ping,SHENG Yehua,ZHANG Siyang,et al.A 3Dlocal RBF spatial interpolation considering anisotropy[J].Geomatics and Information Science of Wuhan University,2015,40(5):632-637.)
• [7]吕海洋,盛业华,段平,等.局部最优形态参数的RBF分块地形插值方法与实验[J].地球信息科学学报,2015,17(3):260-267.(LV Haiyang,SHENG Yehua,DUAN Ping,et al.Interpolation method and experiment of RBF block topography with local optimal morphological[J].Journal of Geo-information Science,2015,17(3):260-267.)
• [8]徐成东,孔云峰,仝文伟.线性加权回归模型的高原山地区域降水空间插值研究[J].地球信息科学,2008,10(1):14-19.(XU Chengdong,KONG Yunfeng,TONG Wenwei.A weighted linear regression model for precipitation spatial interpolation in altiplano and mountain area[J].Journal of Geo-information Science,2008,10(1):14-19.)
• [9]王占华,刘振波,朱迪,等.基于局部加权线性回归模型的凝结水流量在线预警方法[J].计算机与应用化学,2016,33(8):861-865.(WANG Zhanhua,LIU Zhenbo,ZHU Di,et al.An online alarming method for flow of condensate pump based on locally weighted linear regression model[J].Computers and Applied Chemistry,2016,33(8):861-865.)
• [10]梁丹.重庆市PM2.5污染分布特征及影响因素分析[D].北京:北京林业大学,2016:19-23.(LIANG Dan.Analysis on spatial and temporal distribution and influencing factor of PM2.5 pollution in Chongqing[D]. Beijing:Beijing Forestry University,2016:19-23.)
• [11]张怡文,胡静宜,王冉.基于神经网络的PM2.5预测模型[J].江苏师范大学学报(自然科学版),2015,33(1):63-65.(ZHANG Yiwen,HU Jingyi,WANG Ran.PM2.5 forecasting model based on neural network[J].Journal of Jiangsu Normal University(Natural Science Edition),2015,33(1):63-65.)
• [12]王琪.基于北京市PM2.5污染数据的特征提取及相关分析研究[D].北京:北京化工大学,2015:25-57.(WANG Qi.Feature extraction and the research of relatedanalysis based on the data of PM2.5pollution in Beijing[D].Beijing:Beijing University of Chemical Technology,2015:25-57.)
• [13]EFRON B,GONG G.A Leisurely look at the bootstrap,the jackknife,and cross-validation[J].The American Statistician,1983,37(1):36-48.
• [14]任杰,刘家瑛,白蔚,等.基于隐式分段自回归模型的图像插值算法[J].软件学报,2012,23(5):1248-1259.(REN Jie,LIU Jiaying,BAI Wei,et al.Image interpolation algorithm based on implicit piecewise autoregressive model[J].Journal of Software,2012,23(5):1248-1259.)
• [15]张国峰,杨立荣,翟明凯,等.基于地理加权回归克里格的日平均气温插值[J].应用生态学报,2015,26(5):1531-1536.(ZHANG Guofeng,YANG Lirong,ZHAI Mingkai,et al.Interpolation of daily mean temperature by using geographically weighted regression-Kriging[J].Chinese Journal of Applied Ecology,2015,26(5):1531-1536.)
• [16]TIKHONOV A N.Regularization of incorrectly posed problems[J].Soviet Mathematical Doklady,1963,4(3):1624-1627.
• [17]孙梅兰,朱功勤,谢进.构造有理插值函数的一种参数法[J].计算机工程与应用,2014,50(19):47-52.(SUN Meilan,ZHU Gongqin,XIE Jin.Parameter method of constructing rational interpolating function[J].Computer Engineering and Applications,2014,50(19):47-52.)
• [18]肖雪梦,张应应.三种回归方法在消除多重共线性及预测结果的比较[J].统计与决策,2015(24):75-78.(XIAO Xuemeng,ZHANG Yingying. Three regression methods in eliminating multiple collinearity and predictive results[J].Statistics and Decision,2015(24):75-78.)
• [19]方建刚,毛明策,程肖侠.陕西降水的正态分布特征分析[J].西北大学学报(自然科学版),2009,39(1):131-136.(FANG Jiangang,MAO Mingce,CHENG Xiaoxia.The analysis of statistics distribution characteristics on precipitation in Shaanxi[J].Journal of Northwest University(Natural Science Edition),2009,39(1):131-136.
• [20]KEARNS M,RON D.Algorithmic stability and sanitycheck bounds for leave-one-out cross-validation[EB/OL].[2017-02-13].http://www.seas.upenn.edu/~mkearns/papers/multi.pdf.
• [21]姜晓剑,刘小军,黄芬,等.逐日气象要素空间插值方法的比较[J].应用生态学报,2010,21(3):624-630.(JIANG Xiaojian,LIU Xiaojun,HUANG Fen,et al.Comparison of spatial interpolation methods of daily meteorological elements[J]. Chinese Journal of Applied Ecology,2010,21(3):624-630.)