近红外光谱法预测浆料中胶黏物含量的研究

Study on the Rapid Prediction of Stickies Content in Pulp by NIR Spectroscopy

探讨了近红外光谱法快速预测浆料中胶黏物含量的可行性。基于实验室自制的75个胶黏物含量不同的浆料样品,用近红外光谱仪漫反射方式在12500～4000cm-1波数范围内采集相应样品的光谱,利用化学计量学软件建立样品胶黏物含量和光谱数据之间的相关性模型。结果表明,对原始光谱进行最大-最小归一法预处理后,选择12493.4～7498.4cm-1和6102.1～5349.9cm-1波数区间,用偏最小二乘法(PLS)和完全交互验证方式建立的校正模型和外部验证预测模型的相关系数R2分别为0.918和0.935,校正标准差SEC值为0.211,预测标准差SEP为0.211。该模型预测浆料胶黏物含量的重现性相对标准偏差和准确性相对平均偏差分别小于15%和10%,可以达到工业分析的要求。

A rapid measurement method to predict the stickies content in pulp by using near-infrared spectroscopy was developed in this paper. Seventy five pulp samples containing different stickies contents were prepared in the laboratory with tetrahydrofuran extractives of deinked pulp as stickies materials. The spectra of those pulp samples were collected by NIR spectroscopy in range of 12500 -4000 cm^-1, and the relationship between the stickles content and the spectra of the pulp sample was established by chemometrics software. After min-max normalization pretreatment of the original spectra, the stickies contents of the pulp samples were quantified using partial least squares （PLS） statistical analysis and full cross validation in the range of 12493.4 -7498.4 cm^-1 and 6102. 1 - 5349. 9cm^-1. The correlation coefficients of calibration model and prediction model were 0. 918 and 0. 935, respectively. The standard error of calibration （SEC） and standard error of prediction （SEP） were 0. 211 and 0. 211, respectively. The repeatability and the precision of prediction model of the stickies content in pulp were less than 15% and 10% , respectively, which could meet the industry analysis requirement.