漫反射红外光谱法结合PLS测定稻谷脂肪酸值研究

为了快速、简便、无污染地测定稻谷中脂肪酸含量,提出基于漫反射傅里叶变换红外光谱法(DRIFTS)和近红外漫反射光谱法(NIDRS)测定稻谷脂肪酸值快速检测方法。同时探讨了反向区间偏最小二乘法(BiPLS)对中红外光谱区域的优化效果。傅里叶变换原始光谱模型经7点平滑预处理及BiPLS优化,得到的最佳模型的R~2、RMSECV、RMSEP分别为0.998、3.65、3.69。近红外光谱经一阶导数和多元散射校正预处理后建立最佳模型,其R~2为0.97,RMSECV为3.43。在验证实验中,傅里叶变换红外光谱预测值与国标测定值的相对标准偏差为1.16%,近红外光谱预测值与国标测定值的相对标准偏差为1.70%。结果表明,DRIFTS能够更加准确检测脂肪酸值。

Determination of fatty acid value of paddy by diffuse reflectance infrared spectroscopy and PLS

In order to determine the content of fatty acids in paddy quickly, easily and without pollution, the rapid detection method based on diffuse reflectance Fourier transform infrared spectroscopy (DRIFTS) and near infrared diffuse reflectance spectroscopy (NIDRS) for the determination of fatty acid value in paddy was proposed. Meanwhile, the reverse interval partial least squares (BiPLS) is applied to optimize the mid infrared spectral region. The Fourier transform original spectral model was optimized by 7 point smoothing pretreatment and BiPLS. The R2, RMSECV and RMSEP of the best models were 0.998, 3.65 and 3.69, respectively. NIR spectra were first processed by first derivative and multivariate scatter correction, and the best R2 was 0.97 and RMSECV was 3.43. In the test, the relative error between the predicted value of Fourier transform infrared spectroscopy and the national standard is 1.16%, and the relative error between the predicted value of the near infrared spectrum and the national standard is 1.70%. The results showed that DRIFTS could detect fatty acid value more accurately.