主成分分析在大气质量监测优化布点中的应用

主成分分析法(PCA)以少数的综合变量取代原有的多维变量,在原始数据信息丢失最小的情况下,使数据结构得以简化。作者分析了合肥市新站区大气环境监测数据,采用主成分分析法建立空气质量污染特征因子与污染物之间的数学模型,再用该模型计算出各点位相对污染程度,并对监测布点进行分类。以此选出最佳监测点位,可为大气质量监测优化布点提供方法,为合肥市新站区环境质量的分区和分级治理提供理论依据。

Application of principal components analysis in optimizing sites selection of air quality monitoring

The principal component analysis could substitute for original multi-dimensional variable by a small number comprehensive variable?and simplify data structure under the condition of minimizing loss of original data information. The paper was aimed to analyze the air environment monitoring data of Xinzhan area in Hefei city, using the principal component analysis to establish a mathematical model about the characteristic factors of air quality and pollutants. The model?was made to calculate each monitoring sites, relative pollutional degree and classifying monitoring sites, thus to select the best sites to monitor the air quality. The result provided an optimized selection of air quality monitoring sites, a theoretical basis for classifying air quality of Xinzhan area of Hefei and governing by different levels.