1. School of Science,Jiangnan University,Wuxi 214122,China 2. School of Internet of Things Engineering,Jiangnan University,Wuxi 214122,China 3. College of Energy and Electrical Engineering,Hohai University,Nanjing 210098, China
摘要: 以某清香型白酒为研究对象,将三维荧光光谱技术与平行因子分析方法(parallel factor analysis,PARAFAC)、BP神经网络结合,建立清香型白酒年份鉴别模型。首先,利用FLS920全功能型荧光光谱仪测量获得不同年份白酒的三维荧光光谱数据,对激发发射三维矩阵进行三线性分解,得到四个主成分对应的浓度得分和激发-发射光谱轮廓图。将这4个浓度得分作为BP神经网络的输入,建立10,20和30年份白酒的鉴别模型。随机选取每个年份的10个样本,共30个样本组成测试集,剩余的90个白酒样本组成训练集建立训练模型。据此对未知样品进行预测,其预测正确率分别为90%,100%和100%。同时将该方法与多维偏最小二乘判别分析法(multi-way partial least squares discriminant analysis, NPLS-DA)进行了比较。研究结果表明:平行因子结合神经网络的判别模型具有更强的预测能力,该方法能够有效提取年份白酒的特征光谱信息,同时又降低了神经网络输入变量的维数,取得较好的鉴别效果。
关键词:年份鉴别;三维荧光光谱;平行因子;神经网络
Abstract:Three-dimensional fluorescence spectroscopy coupled with parallel factor analysis and neural network was applied to the year discrimination of mild aroma Chinese liquors. The excitation-emission fluorescence matrices (EEMs) of 120 samples with various years were measured by FLS920 fluorescence spectrometer. The trilinear decomposition of the data array was performed and the loading scores of and the excitation-emission profiles of four components were also obtained. The scores were employed as the inputs of the BP neural networks and the PARAFAC-BP identification model was constructed. 10 samples were collected from 10, 20 and 30 years of liquors respectively, and 30 samples were selected as the test sets. The remaining 90 samples were used as the training sets to build the training model. The year prediction of unknown samples was also carried out, and the prediction accuracy was 90%, 100% and 100%, respectively. Meanwhile, the discrimination analysis method and the multi-way partial least squares discriminant analysis were compared, namely PARAFAC-BP and NPLS-DA. The results indicated that parallel factor combined with the neural network (PARAFAC-BP) has higher prediction accuracy. The proposed method can effectively extract the spectral characteristics, and also reduce the dimension of the input variables of neural network. A good year discrimination result was finally achieved.
[1] Peinad R A, Moreno J A, Muoz D, et al. Journal of Agricultural and Food Chemistry, 2004, 52(21): 6389. [2] ZHANG Qin-yi, XIE Chang-sheng, ZHANG Shun-ping, et al. Sensors and Actuators B, 2005, 110(2): 370. [3] CHENG Pin-yan, FAN Wen-lai, XU Yan. Food Control, 2014, 35(1): 153. [4] QIN Hui, HUO Dan-qun, Zhang Liang, et al. Food Research International, 2012, 45(1): 45. [5] YANG Jian-lei, ZHU Tuo, WU Hao(杨建磊, 朱 拓, 武 浩). Journal of Optoelectronics·Laser(光电子·激光), 2009, 20(4): 495. [6] CHU Xiao-li(褚小立). Molecular Spectroscopy Analytical Technology Combined with Chemometrics and Its Applications(化学计量学方法与分子光谱技术). Beijing:Chemical Industry Press(北京:化学工业出版社),2011. 111. [7] Amigo J M, Coello J, Maspoch S. Analytical and Bioanalytical Chemistry, 2005, 382(6): 1380. [8] DU Shu-xin, LI Lin-jun(杜树新, 李林军). Chinese Journal of Luminescence(发光学报),2013, 34(4):523. [9] Bro R, Kiers H A L. Journal of Chemometrics, 2003,17(5):274. [10] Bro R. Journal of Chemometrics, 1996, 10: 47. [11] DU Shu-xin, SHEN Jin-chang, YUAN Zhi-bao(杜树新, 沈进昌, 袁之报). Laser Journal(激光杂志),2012,33(1): 36. [12] ZHANG He-sheng(张和笙). Liquor-Making Science and Technology(酿酒科技), 2003, 118(4): 58. [13] Qiao Hua, Zhang Shengwan, Wang Wei. Journal of Bioscience and Bioengineering, 2013, 115(4): 405. [14] Farmaki E G, Thomaidis N S, Efstathiou C E. International Journal of Environmental and Analytical Chemistry, 2010, 90(2): 85.