摘要
研究了采用近红外光谱测定柴油中芳烃含量的方法。通过高效液相色谱分离、移动丝氢火焰检测器检测获得柴油中芳烃的基础数据。在5000~10000cm-1波长范围内,应用傅立叶变换近红外光谱仪,交互检验偏最小二乘方法对使用的光谱范围进行优化。所得校正模型对训练集样品预测的平均偏差为023%,对未知样品将近红外预测结果与HPLC的测定结果比较,平均偏差为027%。作为一种快速分析技术,近红外光谱测定柴油中的芳烃含量是简便可靠的方法。
In this paper,a method for the determination of aromatic hydrocarbons in diesel products by near infrared spectroscopy is presented.The aromatics content in the diesel samples which were used as trained set for building calibration model was determined by high performance liquid chromatography with moving wire flame ionization detector.The spectral region of 5 000~10 000 cm -1 spectral region was optimized by using Fourier transform near infrared spectrometer together with cross validated partial least squares regression.The mean deviation between the determined value by HPLC and the predicted value for trained set samples is 0.23%,and an excellent correlation is obtained ( R 2=0 906 4).The mean deviation for unknown samples determined by HPLC and NIR is 0.27%.
出处
《石油化工》
CAS
CSCD
北大核心
1999年第4期263-265,共3页
Petrochemical Technology
关键词
柴油
芳烃
含量
近红外光谱
偏最小二乘回归
diesel,aromatic hydrocarbons,near infrared spectroscopy,partial least square regression