基于近红外光谱建立纸张表面施胶量测定模型

采集不同施胶量纸张的近红外光谱,利用偏最小二乘法建立测定纸张表面施胶量基于近红外光谱的校正模型。得到校正模型的交叉验证均方差（RMSECV）和外部验证均方差（RMSEP）分别为0.0928和0.1460,校正数据集和独立的检验数据集的预测值与实际测定值之间的相关系数分别为0.9609和0.9294,表明所建立的校正模型具有较高的预测精度和较好的推广性,为纸张无损伤、无预处理的快速、简便、准确的检测提供了新的途径,并且可望实现纸机上的在线检测。

Determination model for paper surface sizing weight based on near infrared spectroscopy

Near infrared spectroscopy（NIR） of paper samples with different surface sizing weight was collected and partial least squares（PLS） was applied to establish calibration model of paper surface sizing weight based on NIR.RMSECV and RMSEP of the calibration model in cross-validation were gained which are 0.0928 and 0.1460 respectively.The correlation coefficients of calibration model and prediction model were 0.9609 and 0.9294 respectively.It was proved that models built with NIRS and PLSR are an accurate,simple and non-destructive,no-pretreatment approach for rapid and easy paper determination technology,which may realize an on-line detection for surface sizing weight in paper machines.