【摘要】 目的:建立化橘红中柚皮苷近红外定量模型。方法:收集不同产地不同批号化橘红样品的近红外光谱图,运用TQ Analyst 8.0软件,经一阶导数和Norris平滑预处理,在10 000~4 000 cm-1范围内,选择9个主因子数,采用偏最小二乘法(PLS)建立化橘红中柚皮苷含量近红外光定量模型。结果:所建模型预测集相关系数r=0.9927;校正集均方根偏差(RMSEC)=0.0746;预测集均方根偏差(RMSEP)=0.282;验证集预测平均回收率为101.65%(n=9)。结论:本研究所建立的模型性能较好,对化橘红中柚皮苷含量的预测准确。近红外光谱法具有快速、简便、准确、无损的优点,可以应用于化橘红药材的质量控制及其后续产品的开发。更多还原

【Abstract】 Objective: To establish the quantitative model of naringin in Citrus Grandis Exocarpium by near infrared reflectance spectroscopy( NIRS). Methods: NIRSs of 54 Citrus Grandis Exocarpium samples were collected and pretreated by TQ Analyst 8. 0 software with first derivative + Norris filter. The wavebands in 10 000 ~ 4 000 cm-1were collected and 9 numbers of factors were used. The calibration model was built with partial least squares( PLS). Results: The quantitative calibration model had good correlation coefficients( r = 0. 9927) and low root mean square error of calibration( RMSEC = 0. 0746). Root mean square error of prediction( RMSEP) was0. 282. The average rate of recovery of validation was 101. 65%( n = 9). Conclusion: The quantitative calibration model is trustworthy and it can be used to predict naringin in Citrus Grandis Exocarpium accurately. This simple,fast and non-destructive advantages of NIRS can be used for quality control and exploitation of Citrus Grandis Exocarpium.更多还原