基于光谱技术鉴别机油品种的新方法  

Discrimination of Oil Varieties by Using Near Infrared Spectral Technology

作　　者：周子立[1,2] 蒋璐璐[3] 谈黎虹[3] 何勇[2] 李晓丽[2] 邵咏妮[2] 

机构地区：[1]浙江机电职业技术学院,浙江杭州310053 [2]浙江大学生物工程与食品科学学院,浙江杭州310029 [3]浙江经济职业技术学院,浙江杭州310018

出　　处：《光学学报》2009年第8期2203-2207,共5页Acta Optica Sinica

基　　金：十一五国家科技支撑项目(2006BAD10A0902);浙江省重大科技攻关项目(2005C12029);浙江省教育厅资助项目(20071275);浙江机电职业技术学院孵化基金项目(04090331)资助课题

摘　　要：提出了一种用可见近红外透射光谱技术快速鉴别机油品种的新方法,应用可见-近红外光谱仪测定三种机油的光谱曲线,然后用主成分分析法对不同品种的机油样本进行聚类分析,并获取机油可见-近红外光谱的特征信息,再结合多类判别分析技术建立机油品种鉴别的模型,对经过预处理的光谱数据进行主成分分析。结果表明,以样本在第一主成分和第二主成分上的得分做出的二维散点图,对不同种类机油具有很好的聚类,能定性区分不同种类机油;经过主成分分析得到的前8个主成分的累积可信度已达95.38%,说明这8个变量能够代表绝大部分原始光谱的信息。从180个样本中随机抽取150个样本用于建立多类判别分析品种鉴别模型,余下的30个样本用于验证。对未知的30个样本进行品种预测,准确率为100%。证明本方法具有明显的分类和鉴别作用,为不同品种的机油鉴别提供了一种新方法。A new method for discrimination of oil types by means of near infrared spectroscopy （NIRS） was developed. First, the characteristic spectrums of oil were got through principal component analysis （PCA）. The result of the analysis suggests that the reliabilities of first 8 principal components are more than 95.38%. The 2- dimensional plot was drawn with first and second principal components, which indicates that it is a good clustering analysis for classification of oil varieties. Several variables compressed by PCA were used as inputs of multiple discriminant analysis （MDA）. 150 samples from three varieties were selected randomly, then they were used to build discrimination model. 30 unknown samples were predicted by this model, and the recognition rate is 100%. This model is reliable and practicable. It could offer a new approach to the fast discrimination of oil types.

关 键 词：光谱学 机油 近红外光谱 主成分分析 多类判别分析 鉴别 

分 类 号：O657.3[理学—分析化学] TE626.32[理学—化学]