基于近红外光谱结合化学计量学的正天丸素丸中水分的检测研究

Study on the Detection of Moisture in SuWan of Zhengtian Pills Based on Near Infrared Spectroscopy Combined with Chemometrics

摘要: 目的：本研究旨在利用近红外光谱法建立一种快速检测正天丸素丸中水分含量的方法。方法：通过近红外光谱仪对正天丸素丸样品进行扫描，对光谱数据进行预处理和波段选择，并结合偏最小二乘法(partial least squares, PLS)建立了水分含量的快速无损检测模型。结果：研究显示，所建立的模型具有良好的性能，决定系数R达到0.97543，交叉验证均方根差值为0.64069。通过对验证集样品进行预测并进行统计分析，发现预测值与真实值之间无显著差异(P > 0.05)。结论：所建立的模型具有高准确度，适用于正天丸素丸中水分的快速检测，为相关领域提供了一种可行的分析方法。

Abstract: Objective: The aim of this study was to establish a rapid method for the determination of moisture content in zhengtian pills using near-infrared spectroscopy. Methods: The samples of Zhengtianwan Suwan were scanned by near-infrared spectroscopy, the spectral data were preprocessed and the waveband selection was performed, and a model for the rapid nondestructive detection of moisture content was established by combining with the partial least squares (PLS) method. Results: The study showed that the established model has good performance with a coefficient of determination R of 0.97543 and a cross-validated root mean square difference value of 0.64069. By predicting the samples from the validation set and statistically analysing the results, it was found that there was no significant difference between the predicted values and the true values (P>0.05). Conclusion: The established model has high accuracy and is suitable for the rapid detection of moisture in zhengtian pills, which provides a feasible analytical method for related fields.