基于近红外光谱的黄酒风格判别方法.PDFVIP

2014-01-03 基于近红外光谱的黄酒风格判别方法 1 1,2 2 3 1* 薛 磊 ，吕 进 ，施 秧 ，屠海云 ，刘辉军 (1 中国计量学院，计量测试工程学院，杭州 310018 ；2 浙江科技学院，自动化电气工程学院，杭州 310023 ； 3 杭州市质量技术监督检测院，杭州 301110 ) 摘 要： 本文提出了利用近红外透射光谱进行黄酒风格快速鉴别的方法。采用了标准化、标准正态变量变换、一阶导数等 多种预处理方法，建立了传统型、清爽型和特型等3种风格的绍兴黄酒偏最小二乘判别分析(PLS-DA)判别模型。结果表明传 统型、清爽型和特型校正集和预测集的识别率均为100%，成功地实现了黄酒风格的判别。本研究为实现不同风格黄酒快速、 无损定性判别提供了新方法。 关键词：近红外光谱；黄酒；PLS-DA ；风格判别 中图分类号： O657.3 文献标志码：A Identification of Rice Wine Styles Based On Near-infrared Spectroscopy 1 1,2 2 3 1* XUE Lei ，LV Jin ， SHI Yang ，TU Hai-yun ， LIU Hui-jun （ 1. College of Metrological Technology and Engineering, China Jiliang University,Hangzhou 310018 , China; 2. College of Automation and Electrical Engineering,Zhe jiang University of Science and Technology,Hangzhou 310023,china; 3.Hangzhou Institute of Test and Calibration for Quality and Technology Supervision，Hangzhou 301110，China) Abstract: A method of rapid identification of rice wine styles via near-infrared transmission spectroscopy was proposed in this paper. Standardization, standard normal variable transformation and first derivative et al. were selected for pretreatments, then the models of three types Shaoxing rice wine, including traditional style, soft flavor style, and special style were established by PLS-DA (Partial Least Squares Discriminant Analysis). The results showed that identification accuracies of calibration and validation sets of three styles were 100%. This study provided a new method of non-destructive identification of different rice wine styles rapidly. Keywords: near-infrared spect