软件技术在近红外光谱定量分析中的应用

Application of Software Technology in Quantitative Analysis of Near Infrared Spectroscopy

为了更好地运用化学计量学方法提高近红外光谱定量分析的质量,全面研究了相关算法技术与软件工具的应用。根据近红外光谱定量分析的一般流程,依次探讨了光谱数据预处理、异常样本检测、样本集划分、波长选择、模型建立及其参数优化的常用算法,并给出了其中各种主流算法的基本步骤;然后介绍了OMNIC、OPUS、Origin、The Unscrambler X和MATLAB等软件工具的功能特点和主要用途,可为相关技术研究和分析检测工作提供参考。

To improve the quality of quantitative analysis of near infrared spectroscopy by chemometrics, the application of related algorithms and software tools were comprehensively studied. According to the general process of quantitative analysis of near infrared spectrum, the common algorithms of spectral data preprocessing, abnormal sample detection, sample set division, wavelength selection, model establishment and parameter optimization were discussed in turn, and the basic steps of various mainstream algorithms were given. Then the functions, lea-tures and main applications of OMNIC, OPUS, Origin, The Unscrambler X and MATLAB software tools were introduced to provide reference for related technology research and analysis detection work.