多成分有机气体的近红外光谱定量检测方法

Application of NIR Spectroscopy to Multiple Gas Components Identification

在环境检测领域,运用近红外光谱分析技术对多成分挥发性有机物进行连续的在线监测具有重要意义。本文分析了三种挥发性有机物——丙烷、丙烯和甲苯的近红外光谱特征和丙烯浓度与吸光度的线性关系,采用线性回归建模方法——偏最小二乘法对丙烷、丙烯和甲苯混合气体存在特征吸收的近红外光谱（1620-1750 nm）进行了建模分析,基于该模型预测了验证集样品中三种气体的含量,并对模型进行了评价。实验结果表明,近红外光谱技术可以方便、准确的定量分析有机性挥发气体的多组分含量,可以应用于多成分挥发性有机物的连续在线监测。

Multiple gas components monitoring in situ by near infrared（NIR） spectroscopy is of great significance in the environmental monitoring.In the present paper,the apectral characteristics of the three types of the volatile organic compounds（VOCs）-propane,propylene and methylbenzene were analyzed,the linear relationship of propylene between concentration and absorbance was considered,and the NIR spectra from 1 620 to 1 750 nm including the characteristic absorption of the three VOCs were acquired,A linear regression model of chemical metrology was created by partial least-squares method and it predicted the concentration of propane and isobutene in the validation set.The results of the experiment indicated that NIR spectroscopy could easily,accurately and quantitatively determine the content of the multiple gas components,and can be used for monitoring the multiple components of the VOCs in situ.