论文部分内容阅读
目的:建立枇杷叶药材中水分含量的近红外定量模型,快速测定枇杷叶中水分含量。方法:用烘干法测定173批枇杷叶中水分含量,采用一阶导数法预处理近红外光谱,结合偏最小二乘法建立建立枇杷叶中水分含量的定量模型。结果:所建立的校正模型,内部交叉验证决定系数达到0.965 86,校正均方差为0.082 5,预测均方差为0.137,内部交叉验证均方差为0.152 63。结论:该近红外定量模型稳定,准确可靠,可快速对枇杷叶中水分含量进行测定。
Objective: To establish a near-infrared quantitative model of moisture content in loquat leaf medicine, and to quickly determine the water content of loquat leaf. Methods: The moisture content in 173 batches of loquat leaves was determined by the drying method. The first derivative method was used to pretreat the near infrared spectrum and the partial least squares method was used to establish a quantitative model of water content in loquat leaves. Results: The calibration model was established, the internal cross validation coefficient was 0.965 86, the mean square error correction was 0.082 5, the prediction mean square error was 0.137, and the internal cross validation mean square error was 0.152 63. Conclusion: The near-infrared quantitative model is stable, accurate and reliable, and can quickly determine the moisture content of loquat leaves.