基于近红外光谱技术的老陈醋品质分析  

Analysis of Mature Vinegar Quality Based on Near Infrared Spectroscopy Technology

作　　者：陆辉山[1] 吴远远[1] 刘修林 

机构地区：[1]中北大学机械与动力工程学院,太原030051

出　　处：《中国调味品》2017年第5期103-106,共4页China Condiment

基　　金：山西省科技攻关项目(20150311023-2);山西省2015高校科技创新项目(180012-117)

摘　　要：为得到稳定、精确的老陈醋品质光谱模型,采用近红外光谱分析技术结合反向传播神经网络(BP-ANN),分别对不同醋龄的两种山西老陈醋中可溶性固形物含量(SSC)及pH值进行定量分析。对经过标准归一化(SNV)与25点平滑相结合处理后的光谱进行主成分分析,根据主成分的累计贡献率选取主成分数作为BP神经网络的输入变量建立模型,并与偏最小二乘法(PLS)模型进行比较。结果表明:BP-ANN建立的老陈醋SSC和pH值定量分析模型最优,其SSC和pH值的模型相关系数(R)分别为0.9999和0.9997,校正集均方根误差(RMSEC)分别为0.0128和0.0045,预测集均方根误差(RMSEP)分别为0.0118和0.0088。采用近红外光谱技术结合反向传播神经网络(BP-ANN)对不同醋龄、不同品牌的老陈醋品质分析建模是可行的。In order to obtain the stable and accurate spectral model of mature vinegar quality, the near infrared spectroscopy analysis technology combined with back-propagation neural network（ANN）is used for the quantitative analysis of the soluble solids content（SSC）and the pH value respectively for two kinds of Shanxi mature vinegar with different age. Principal component analysis of spectrum is made after standard normal variable （SNV）and 25 point smoothing combination treatment. The principal component is selected according to the cumulative contribution rate, which is used as input variables of the BP neural network model, and the BP neural network model is compared with the partial least squares （PLS） model. The results show that the model of mature yinegar SSC and pH value quantitative analysis constructed by BP-ANN is the best： the correlation coefficient（R）of the model of SSC and pH value is 0. 9999 and 0. 9997 respectively, the root mean square error（RMSEC）of the calibration set is 0. 0128 and 0. 0045 respectively, the root mean square error of prediction（RMSEP）set is 0. 0118 and 0. 0088 respectively. It is feasible to use near infrared spectroscopy combined with back propagation neural network （BP-ANN） to build the analysis model of mature vinegar with different age and different

关 键 词：近红外光谱技术 可溶性固形物含量(SSC) pH值 预处理 BP神经网络 偏最小二乘法(PLS) 

分 类 号：TS264.22[轻工技术与工程—发酵工程]