论文部分内容阅读
目的:采用近红外光谱技术结合PLS建立一种木香药材中水分含量快速测定的方法。方法:运用近红外漫反射光谱技术采集木香的近红外漫反射光谱,以甲苯法测定的含量为参考值,结合偏最小二乘法(PLS)建立木香药材中水分含量的定量分析模型,并用未知样品验证该模型。结果:所建水分定量模型的校正集内部交叉验证相关系数(R2)、校正均方差(RMSEC)和预测均方差(RMSEP)分别为0.982 9,0.178,0.196;验证集NIR预测值与甲苯法参考值的t检验值为-0.615,双侧P>0.05,差异无统计学意义。结论:该方法操作简便,测定快速,结果准确,可用于木香药材中水分含量的快速测定。
Objective: To establish a method for rapid determination of water content in woody herbs by near-infrared spectroscopy and PLS. Methods: Near-infrared diffuse reflectance spectroscopy was used to collect the near-infrared diffuse reflectance spectra of woody wood. The content of toluene in wood was taken as the reference value. The quantitative analysis model of water content in woody medicinal herbs was established by using partial least squares (PLS) Unknown samples validate the model. Results: Correlation coefficient (R2), RMSEC and RMSEP of calibration set were 0.982 9,0.178,0.196 respectively. The validation set of NIR predictive value was similar to that of toluene method The value of t test was -0.615, bilateral P> 0.05, the difference was not statistically significant. Conclusion: The method is simple, rapid determination, accurate results, can be used for rapid determination of the moisture content of woody herbs.