摘要
随着信息化及智能化制造的普及,酿酒行业面临海量数据的采集和统计需求,依托于大数据系统和快速检测方法的各种模式识别方法建模在酒行业中的应用越来越广泛。总结了偏最小二乘法、主成分分析法、人工神经网络法、支持向量机及其他模式识别方法在酒行业中理化检测、质量等级分类、品牌鉴别、酒龄鉴定、产地溯源、智能摘酒等方面的应用研究成果。此外,还对不同方法间组合使用建模进行了说明,旨在为酒类酿造过程中产生的模糊数据建模、分析和应用提供思路。
With the popularization of informationization and intelligent manufacturing,the wine industry is faced with the collection and statistical demand of massive data.Various pattern recognition methods based on big data system and fast detection methods are widely used in the wine industry.The application research achievements of partial least squares method,principal component analysis method,artificial neural network method,support vector machine and other pattern recognition methods in some aspects such as physical and chemical detection,quality classification,brand identification,wine age identification,origin tracing,intelligent wine picking and so on were summarized in this paper.In addition,the modeling of the combined use of different methods was explained,aiming to provide ideas for the modeling,analysis and application of fuzzy data generated in the process of wine brewing.
作者
韩云翠
王冠霖
吕志远
刘玉涛
张梦梦
卢春玲
邱振清
汪俊卿
HAN Yuncui;WANG Guanlin;L Zhiyuan;LIU Yutao;ZHANG Mengmeng;LU Chunling;QIU Zhenqing;WANG Junqing(School of Bioengineering,Qilu University of Technology(Shandong Academy of sciences),Jinan 250353,China;Jinan Baotuquan Brewery Co.,Ltd.Jinan 250115,China;Computer Science Institute,Beijing University of Aeronautics and Astronautics,Beijing 100191,China)
出处
《齐鲁工业大学学报》
CAS
2023年第2期74-80,共7页
Journal of Qilu University of Technology
基金
山东省重点研发计划(重大科技创新工程)(2022CXGC020206)。
关键词
模式识别
偏最小二乘法
主成分分析法
人工神经网络法
支持向量机
pattern recognition
partial least square method
principal component analysis
artificial neural network
support vector machine
