近红外光谱法对甲醇柴油中甲醇含量测定

发布时间：2018-06-11 22:35

本文选题：近红外光谱法 + 甲醇柴油　；参考：《光谱学与光谱分析》2017年04期

【摘要】：应用近红外光谱结合化学计量学方法实现了对甲醇柴油中的甲醇含量的定量分析。以实验室配制的32种不同浓度[浓度范围为2%~25.8%(φ)]的甲醇柴油溶液为研究对象,在4 500~7 000cm~(-1)光谱范围内,建立偏最小二乘(PLS)、支持向量机(SVM)和最小二乘支持向量机(LS-SVM)三种定量分析模型。在建立SVM模型时,经过比较分析,径向基函数(radial basis function,RBF)作为SVM模型的核函数时可以获得更高的预测精度。最终获得甲醇含量的PLS,SVM和LS-SVM三种模型的预测相关系数RP分别为0.985 9,0.990 3,0.998 9,预测均方根误差RMSEP分别为0.405 2,0.356 3,0.062 4,可以看出,三种预测模型都可以达到很好的效果,最优的预测模型是使用LS-SVM建模。研究结果表明,利用近红外光谱法结合化学计量学方法对甲醇柴油中甲醇含量的检测具有可行性,并可以达到很好的效果。采用近红外光谱技术结合化学计量方法对甲醇柴油中甲醇含量进行定量分析,也为近红外光谱技术快速无损检测甲醇柴油甲醇含量提供参考和应用价值。
[Abstract]:The quantitative analysis of methanol content in methanol diesel oil was carried out by means of near infrared spectroscopy (NIR) and chemometrics. Three quantitative analysis models of partial least squares (PLS), support vector machine (SVM) and least squares support vector machine (LS-SVM) were established in the range of 4 500 ~ 7 000 cm ~ (-1) for methanol diesel solution with different concentrations [concentration range 2: 25. 8 (蠁)]. When the SVM model is established, the radial basis function (RBF) can be used as the kernel function of the SVM model to obtain higher prediction accuracy. Finally, the predicted correlation coefficient RP of the three models of methanol content of PLSN SVM and LS-SVM is 0.985 ~ 0.990 ~ 3 ~ 0.998 9, and the RMSEP of root mean square error is 0.405 ~ 2 ~ 0.356 ~ 30.062 ~ 4, respectively. It can be seen that the three prediction models can achieve good results. The optimal prediction model is based on LS-SVM. The results show that the near infrared spectroscopy combined with chemometrics is feasible for the determination of methanol content in methanol diesel oil, and a good effect can be achieved. The quantitative analysis of methanol content in methanol diesel oil by means of near infrared spectroscopy combined with stoichiometry also provides a reference and application value for fast nondestructive detection of methanol content in methanol diesel oil by near infrared spectroscopy.
【作者单位】：华东交通大学机电工程学院光机电技术及应用研究所;
【基金】：国家自然科学基金项目(61178036,31160250) 江西省光电检测工程技术研究中心项目(赣科发财字[2012]155号)资助
【分类号】：TQ517;O657.33

【相似文献】