基于低场核磁共振的乳粉中水分及酸度的快速预测

Rapid Prediction of Moisture and Acidity in Milk Powder Based on Low Field Nuclear Magnetic Resonance

为探求一种能快速无损测定乳粉中水分含量及酸度的方法,采用低场核磁共振仪对乳粉样品进行测量获得横向弛豫光谱数据。用直接干燥法和酚酞指示剂法对360个乳粉样品中的水分和酸度进行测定,获得参考值。运用偏最小二乘法(PLS)和误差反向传播人工神经网络法(BP-ANN),结合光谱数据和参考值,对乳粉中的水分和酸度进行建模及预测。PLS和BP-ANN在乳粉水分含量的测定中预测集的预测值与参考值之间的线性相关系数分别为0.8113、0.9791,均方根偏差分别为0.1066、0.0321。PLS与BP-ANN在乳粉酸度的测定中预测集的预测值与参考值之间的线性相关系数分别为0.9808、0.9993,均方根偏差分别为0.1766、0.0316。研究表明,采用低场核磁共振技术结合多元校正的方法可以用于对乳粉水分含量和酸度的快速无损测定。

In order to explore a rapid and non-destructive method for the determination of moisture content and acidity in milk powder,low field nuclear magnetic resonance(LF-NMR)was used to measure the milk powder samples and obtain the transverse relaxation spectrum data.The moisture and acidity of 360 milk powder samples were determined by direct drying method and phenolphthalein indicator method.Partial least squares(PLS)and back propagation artificial neural network(BP-ANN)were used to model and predict the moisture and acidity in milk powder with spectral data and reference values.In the determination of moisture content of milk powder,the linear correlation coefficients of the prediction model were 0.8113 for PLS and 0.9791 for BP-ANN,respectively;the root mean standard errors of prediction were 0.1066 for PLS and 0.0321 for BP-ANN,respectively.In the determination of acidity,the linear correlation coefficients of the prediction model were 0.9808 for PLS and 0.9993 for BP-ANN,respectively;the root mean standard errors of prediction were 0.1766 for PLS and 0.0316 for BP-ANN,respectively.The results showed that low field nuclear magnetic resonance combined with multivariate calibration methods can be used for rapid and non-destructive determination of moisture content and acidity of milk powder.