中文摘要：

该研究采用拉曼光谱法对汽油中的总芳烃和总烯烃含量进行直接测定,通过随机蛙跳方法选择特征波段并采用偏最小二乘法建立模型。总芳烃模型的测试集相关系数为0.985,预测均方根误差为1.08,总烯烃模型的测试集相关系数为0.942,预测均方根误差为0.78。结果表明模型相关性较好,可满足实际应用需求,为拉曼光谱在汽油成分检测中的应用奠定基础。

英文摘要：

In this paper, the Raman spectroscopy was used to determine the contents of aromatic hydrocarbon and olefin hydrocarbon in gasoline. Random Frog method was employed to select the Raman bands and the best model was established on the basis of optimized conditions. The correlation coefficients of prediction were 0.985 and 0.942 for aromatic hydrocarbon and olefin hydrocarbon, respectively. The RMSEP（ Root Mean Square Error of Prediction） was 1.08 and0.78, respectively.The results show this model has good linear correlation and adapt to the practical application demand,lay a foundation for gasoline composition detection used raman spectroscopy.