【摘要】 在大气颗粒物源解析领域中,在一套数据中由于选择不同的拟合源和拟合组分,CMB模型会产生多个解析结果。提出了"NKCMB最优拟合法",解决现存的CMB模型在所限定的条件内不能穷举所有的解析结果以及在众多的结果中人工择优的局限等问题。基于"NKCMB最优拟合法"和美国环保总署提供的EPA-CMB8.2软件开发了NKCMB3.0软件。使用美国环保总署提供的测试数据进行的对比试验证明,"NKCMB最优拟合法"能够找到比EPA-CMB8.2的"Best Fit"更优的源解析结果。更多还原

【Abstract】 In the source apportionment field of atmospheric particulate matter,Chemical mass balance(CMB) model regularly gives out multi-results when choosing different species and sources from the same data set.The issues are that how to find all potential results automatically, and how to choose the best result automatically among those multi-results.The"NKCMB Best Fit"method was introduced to supply the gaps of current CMB models.A new CMB software_NKCMB3.0 was developed based on"NKCMB Best Fit"and USEPA-CMB8. 2.Test data provided by USEPA was employed to evaluate the performance of two models.It indicates that the"NKCMB Best Fit"approach of NKCMB30 can give better result than the"Best Fit"approach of EPA-CMB8.2.更多还原