近红外光谱技术用于白砂糖质量的实时监测研究

采用国产近红外在线分析仪,结合漫反射光谱分析技术,对广西某糖厂的白砂糖质量进行了在线应用研究.讨论了影响建模的因素,并提出一些解决问题的方案.采用SupNIR-4000型长波近红外在线分析仪,直接对产品传送带上的白砂糖进行扫描获得光谱数据,运用偏最小二乘法对实验数据进行多元统计分析建立数学模型,并进行了预测.样品色值的预测值和实测值间的相关系数为0.885,标准校正偏差为4.300,色值预测偏差小于糖厂要求的±10;样品浊度预测值和实测值间的相关系数为0.909,标准校正偏差为2.308,浊度预测偏差小于糖厂要求的±4.结果显示,模型预测结果符合糖厂要求,表明在线近红外技术在白砂糖质量监控中具有较好的应用前景.

Real-time Monitoring the Quality of Granulated Sugar by Near Infrared Spectroscopic Technique

Real-time monitoring of the quality of granulated sugar produced in Guangxi sugar factory was carried out using a domestic on-line near-infrared analyzer(SupNIR-4000) combined with diffuse reflectance spectrometry.The spectral data were obtained by scanning the granulated sugar on the belt conveyer with NIR analyzer.The proposed calibration models for the chroma and turbidity of sugar were based on the multivariate statistical analysis of their experimental data by the partial least square method.The factors,e.g.the thickness and graininess of granulated sugar,the fixing height of the diffuse reflectance probe and the temperature of granulated sugar and environment that affected the performances of calibration models were discussed in detail.Some methods were suggested to solve the problems.The chroma and turbidity values of granulated sugar were predicted by the optimized models.The results showed that the correlation coefficients between the predicted and experimental values were 0.885 for chroma with a standard deviation of 4.300,and 0.909 for turbidity with a standard deviation of 2.914.Since the prediction errors required by the sugar factory are ±10 for chroma and ±4 for turbidity,the predicted results could meet the requirement of the factory.Thus the proposed method is feasible for the real-time monitoring of the quality of granulated sugar in sugar industry.