基于近红外光谱技术的日照绿茶茶鲜叶模型的优化

Optimizing the Model for Rapid Determination of Tea Polyphenol in Fresh Tea Shoots Using Near-infrared Spectroscopy

建立一种茶鲜叶中茶多酚含量的快速测定方法。采取日照市2个主要茶叶产区60个茶园的茶鲜叶,利用近红外光谱技术结合化学计量学方法建立优化模型进行预测。茶多酚的参考值采用国家标准方法进行测定。结果表明,采用Savitzky-Golay平滑和一阶导数处理光谱,能够有效消除光谱中的噪音及非目标因素的影响。遗传算法(Genetic Algorithm,GA)选择了针对目标组分茶多酚的200个信息变量,简化了模型。同时,采用主成分回归(PCR)和偏最小二乘(PLS)两种建模方法进行预测,其预测均方根误差(RMSEP)及剩余预测偏差(RPD)分别为0.6158、0.6743和4.7238、4.3141;决定系数(R2)分别为0.9514和0.9451。

In order to build a method for rapid determination of tea polyphenol（TP） in fresh tea shoots, near-infrared spectroscopy and ehemometries method were applied. Fresh tea shoot samples were from 60 tea gardens of main tea growing districts in Rizhao. TP contents of the samples were measured with the national standard. Principal component regression（PCR） and partial least squares（PLS） regression were used for building the calibration model, and the effects of spectral preproeessing and genetic algorithm （GA） on the model were investigated for optimization of the model. The results show that Savitzky- Golay smooth can significantly improve the model, and first-order derivative has slight influence on the model. 200 informative variables were selected by GA. To validate the optimal model, the TP contents in an independent sample set were predicted. The residual predictive deviation（RPD） values are as high as 4. 7238 and 4. 3141 for PCR and PLS models. The squared correlation coefficients （R2 ） between the predicted values and the contents measured with the national standard are 0. 9514 and 0. 9451 for polyphenol in fresh tea shoots, respectively. The method is important for the tea quality safety, standardized production and improving the tea quality.