论文部分内容阅读
为比较空气和纯水背景的模型效果,扫描5个(2、3、5、8和10 mm)不同光程的红茶汤近红外光谱,应用排列组合原理和偏最小二乘法(PLS)建立2种背景下单光程和多光程组合茶多酚近红外预测模型,并用t-test检验模型预测效果。结果表明:单光程模型中,纯水背景与空气背景模型预测结果间无显著性差异;多光程组合模型中,空气背景模型要全部优于纯水背景模型,并且最佳组合模型((5+8)mm)具有比单光程模型更好的预测精度。空气背景茶多酚模型要优于纯水背景茶多酚模型。
In order to compare the effects of air and pure water on the background of the model, five (2, 3, 5, 8 and 10 mm) optical pathlengths of black tea were scanned near infrared spectroscopy. The arrangement principle and partial least squares (PLS) Under the background of single light path and multi-path combination of tea polyphenol prediction model near infrared, and t-test test model to predict the effect. The results show that there is no significant difference between the pure water background and the air background model in the single-path model. In the multi-path combination model, the air background model is better than the pure water background model, and the best combination model (( 5 + 8) mm) has better prediction accuracy than the single-path model. The air background polyphenols model is better than the pure water background polyphenols model.