用近红外光谱技术测定汽油辛烷值，在高精度分光光度计上测得１２个汽油标准样品和４个未知样品的近红外区（７００～２５００ｎｍ）吸收光谱，建立多元统计分析模型，用逐步回归法和偏最小二乘法对模型进行校准，并将其用于未知样品的预估分析，辛烷值的分析精度达到≤±１．０。 Near-infrared spectroscopy was used to determine the octane number of gasoline. The absorption spectra of 12 gasoline standard samples and 4 unknown samples in the near infrared region (700-2500 nm) were measured on a high-precision spectrophotometer. Multivariate statistical analysis models were established, Regression method and partial least squares method were used to calibrate the model and use it in the predictive analysis of unknown samples. The octane analysis accuracy was ≤ ± 1.0.