论文部分内容阅读
随机选取烤烟建模集样品(150个)和检验集样品(35个),利用傅里叶变换近红外光谱仪测定烤烟样品的近红外光谱,并用常规化学分析法测定烤烟样品的含钾量。采用偏最小二乘法(PLS)把测得的烤烟样品的光谱值与烤烟钾含量的数值拟合建立定标模型,经分析得出:预测模型分析烤烟钾含量的决定系数(R2)为0.909,预测标准差(RMSEP)为0.119%。近红外法测定结果与常规化学分析方法的结果具有较好的相关性,能够应用于烤烟钾含量的快速诊断。
Flue-cured tobacco samples were selected randomly (150 samples) and test samples (35 samples). The FTIR spectra were used to determine the near-infrared spectra of flue-cured tobacco samples. The contents of potassium in flue-cured tobacco samples were determined by conventional chemical analysis. The PLS was used to fit the measured spectra of flue-cured tobacco samples with the potassium content in flue-cured tobacco to establish a calibration model. The results showed that the determination coefficient (R2) of potassium content in flue-cured tobacco was 0.909, The standard deviation of prediction (RMSEP) was 0.119%. The results of near infrared spectroscopy have a good correlation with the results of routine chemical analysis and can be applied to the rapid diagnosis of potassium content in flue-cured tobacco.