低场核磁共振结合多元校正对休闲豆干水分含量的快速测定

目的]采用低场核磁共振仪对休闲豆干样品进行测量获取横向弛豫数据,结合多元校正方法对水分含量进行快速测定。方法]使用直接干燥法测定豆干水分含量,测得的结果作为化学值。运用偏最小二乘(PLS)和误差反向传播人工神经网络(BP-ANN)方法结合豆干样品的核磁共振数据与化学值建立多元校正模型,实现对豆干水分含量的快速测定。结果]对于PLS与BP-ANN方法,校正集样品的水分含量预测值和化学值之间的相关系数分别为0.923 5和0.917 6,校正均方根误差分别为0.027 2和0.028 1;预测集样品的水分预测值和化学值之间的相关系数分别为0.918 9和0.921 5,预测均方根误差分别为0.024 8和0.022 3。结论]2种方法都能快速而准确地对休闲豆干的水分含量进行预测。

Rapid Determination of Moisture Content of Leisure Dried Tofu by Using Low Field Nuclear Magnetic Resonance Combined with Multivariate Calibration

Objective] The samples of leisure dried tofu were measured by using low field nuclear magnetic resonance spectrometer (LF-NMR) for obtaining transverse relaxation data.Rapid determination of moisture content of leisure dried tofu was conducted by using LF-NMR combined with multivariate calibration methods.Method] Direct drying method was used to determine moisture content of dried tofu.The measured results were used as chemical values.The model was established by employing the partial least squares (PLS) and back-propagation artificial neural network (BP-ANN) methods combined with NMR data and chemical values.It realized the rapid determination of moisture content of dried tofu.Result] The calibration and prediction moisture content models respectively gave the correlation coefficients of 0.923 5 and 0.918 9 for PLS and of 0.917 6 and 0.921 5 for BP-ANN;the root mean standard errors of calibration (RMSEC) were 0.027 2 for PLS and 0.028 1 for BP-ANN,respectively;the root mean standard errors of prediction (RMSEP) were 0.024 8 for PLS and 0.022 3 for BP-ANN,respectively.Conclusion] The two methods can quickly and accurately predict the moisture content of leisure dried tofu.