近红外光谱分析技术在石脑油分析中的应用

先用由20个石脑油试样组成的样品集和相关软件建立石脑油的近红外光谱离线数学模型,后用建立的数学模型预测石脑油的族组成、馏程、密度等性能参数,还考察了模型的准确度和精密度.可采用甲苯验证波长的漂移情况,漂移范围应为±1 cm-1.仪器分辨率应为4 cm-1.推荐扫描次数为32次.用建立的数学模型预测某石脑油试样的族组成,20次扫描预测结果显示,数学模型的精密度能满足ASTM D 5443要求.用建立的数学模型预测20个石脑油盲样的族组成并与用PIONA多维色谱仪测得的结果进行对比,结果显示数学模型的准确度可满足ASTM D 5443,1319要求.

Application of near - infrared spectroscopy in naphtha analysis

The off- line mathematical model ofnear- infrared spectroscopy for naphtha was estab- lished with the naphtha sample set consisting of 20 samples and the related softwares,and then the per- formance parameters of naphtha,such as group com- position, boiling range and density were predicted by the established model. The accura the model were investigated. Tolu to examine the wave length shift, cy and precision of ene could be used and the shin rangeshould be ±1 cm^-1.The strument needed to ded scanning times be 4 was resolution ratio -1 cm ,and the of the in- recommen- 32. When the model wasused to predict the group comp,3sition of a naphtha sample ,the results of scanning 20 times showed that the precision could meet the need of ASTM D 5443.20 blind naphtha samples were predicted ingroup composition by the model, and the results were compared with those measured by PIONA multidi- mensional chromatography analyzer, the results indi- cated that the accuracy of the model could satisfy the demand of ASTM D 5443 and 1319