摘要
基于聚类分析法利用数据之间距离系数进行分类的原理,建立空气监测点位聚类分析优化模型,结合阜新市地形、气象及历史监测数据,进行阜新市空气监测点位布设优化应用,优化结果表明:距离相似水平取d=0.3时,环保局(B点)与人民公园(C点)监测点空气污染物浓度分布相似性最高,合并为1点,增设气象台监测点位作为清洁背景点,4个点位构成阜新市空气监测新网络;利用CALPUFF模型模拟对优化后监测点位进行相关性检验。检验结果表明:监测点位优化后SO2浓度与实测值相关系数为0.984,PM10相关系数为0.968,NO2相关系数为0.973。CALPUFF模型模拟值与实际监测值之间相关系数均大于0.75,表明优化后的阜新市空气监测点位具有客观环境代表性;监测点位优化与检验方法具有一定应用价值。
with the unceasing change of Fuxin City,China,former air monitoring can not fully reflect the status of air environmental quality,so the optimization of air monitoring sites is imperative.Based on the classification of distance coefficient of data,optimization model for air monitoring sites was established.Clustering analysis of the optimization model was also discussed.Results show that the SO2 concentration in the correlation coefficient is 0.984,PM10 correlation coefficient is 0.968,and NO2 correlation coefficient is 0.973.CALPUFF model correlation coefficient between simulation value and the actual monitoring is higher than 0.75,indicating that the optimized Fuxin city air monitoring points is objective representative.
出处
《地球与环境》
CAS
CSCD
2015年第3期350-355,共6页
Earth and Environment
基金
辽宁省教育厅科研项目(L2012121)
