假冒伪劣润滑油对机械设备的正常运转具有严重的影响,这里提出基于不同基底的拉曼光谱对润滑油掺假定量鉴别方法.研究结果表明:采用拉曼光谱对润滑油掺假样品进行校正模型建立,模型的校正相关系数均大于0.948,能很好地对验证集进行预测.结合间隔偏最小二乘(iPLS)和遗传算法(GA)对建模变量进行筛选,最终通过iPLS和GA分别选出了1 148～1 484cm-1波段和185个波数点分别作为校正模型的输入变量.采用昆仑天歌掺假美孚润滑油作为特征波段筛选验证,建模结果达到很好的预期效果,大大缩减了建模的计算量,能为润滑油掺假判别分析仪器开发方面提供一定的理论指导. Fake and shoddy lubricants have a serious impact on the normal operation of machinery and equipment, here based on Raman spectroscopy of different substrates adulterated quantitative identification method.The results show that: the use of Raman spectroscopy adulterated samples for correction model The correlation coefficients of the model and the model were all greater than 0.948, which could be used to predict the validation set.Mathematical variables were screened by interval partial least squares (iPLS) and genetic algorithm (GA) 1 148 ~ 1 484cm-1 band and 185 wave number points were taken as the input variables of the calibration model respectively.Using Kunlun Tiange adulteress Mobil lube oil as the characteristic band screening validation, the modeling results achieved very good expected results, greatly reducing The computational load of modeling can provide some theoretical guidance for the development of adulteration discriminant analysis equipment.