基于单片机及机器学习的电子舌系统设计与研究
发布时间：2019-06-11 17:58
【摘要】：电子舌是由传感器阵列,配以模式识别方式及多元统计方法的定性分析液体的电子检测仪器。可以用来检测不同液体食品的味道,并将其不同的味道信息“可视化”。因此电子舌是目前食品检测领域中研究的新热点。 本研究的主要目的是为了探索设计较为廉价和稳定的电子舌系统软硬件,同时通过一系列相关实验和数据分析,实现对该系统的可行性验证。该课题对于食品领域的快速无损检验的研究与发展,有着十分重要的意义。 本课题自主开发了一套电子舌系统,主要由①离子选择电极传感器阵列,②信号调理模块,③数据采集模块,④数据分析模块,共四部分组成。并开发了相应的软件,可以对系统进行控制,并对采集的数据进行统计分析及模式识别。主要的研究内容主要有以下几方面： (1)介绍了常用的电子舌传感器的原理,以及一般的电子舌结构。 (2)设计了针对离子选择电极的多通道高阻抗小信号放大电路,把电子舌传感器的信号进行滤波和放大,并以C51MCU为基础设计了一块模数信号转换板卡,可以采集传感器信号至PC机。 (3)利用C语言与VC++平台,编写了下位机和上位机程序,实现数据传输,人机交互,以及仪器控制功能。 (4)在MATLAB7.11环境下,以机器学习理论为指导,编写了一整套数据处理程序,包括PCA, LDA,以及BP, PNN, RBF神经网络及其参数优化工具,基于Lib-SVM工具箱的模式识别软件,基于二维搜索的SVM参数优化工具和基于GAOT遗传算法工具箱的SVM参数优化工具,基于LS-SVM工具箱的模式识别软件。基于C语言编写了电子舌数据BP神经网络处理程序。最后整合为原创的ET GUI TEST电子舌数据处理软件。 应用该电子舌系统对4种不同品牌的矿泉水和5种不同的苹果汁,对数据进行了分析和模式识别,结果证明,该电子舌系统对矿泉水和苹果汁的种类可以有效区分。 由对矿泉水和果汁的检测试验,证明该电子舌原型是有效和可行的,本文在总结与展望中对未来电子舌硬件设计的发展做出了展望和更先进的设计。
[Abstract]:The electronic tongue is an electronic detection instrument for qualitative analysis of liquid by a sensor array, a pattern recognition mode and a multivariate statistical method. Can be used to detect the taste of different liquid foods and to "visualization" different taste information. Therefore, the electronic tongue is a new hot spot in the field of food detection. The main purpose of this study is to explore the software and hardware of the electronic tongue system with low cost and stability, and to realize the feasibility of the system by a series of related experiments and data analysis. The subject is of great importance to the research and development of the rapid and non-destructive inspection in the food field. This topic has developed a set of electronic tongue system, which mainly consists of four-part ion-selective electrode sensor array, two-phase signal conditioning module, one-time data acquisition module and one-time data analysis module. The software is developed, and the system can be controlled, and the collected data can be analyzed and used. The main research contents are as follows: Aspects: (1) The principle of the commonly used electronic tongue sensor and the general principle are introduced. The electronic tongue structure is designed. (2) The multi-channel high-impedance small-signal amplifying circuit for the ion-selective electrode is designed, the signal of the electronic tongue sensor is filtered and amplified, an analog-to-digital signal conversion board is designed based on the C51MCU, and the sensor can be collected and sensed. and (3) using the C language and the VC ++ platform, the program of the lower computer and the upper computer is written, the data transmission and the human-machine interaction are realized, and the control function of the instrument. (4) In the environment of MATLAB7.11, a complete set of data processing programs, including PCA, LDA, and BP, PNN, RBF neural network and its parameter optimization tools, are written based on the theory of machine learning, based on the Lib-SV The pattern recognition software of the M-kit, the SVM parameter optimization tool based on the two-dimensional search and the SVM parameter optimization tool based on the GAOT genetic algorithm toolbox, based on the LS-SV The pattern recognition software of the M-kit. The number of electronic tongues is written based on C language. BP neural network processing program. Finally, it is the original ET GUI. The electronic tongue system is used to analyze and pattern the data of four different brands of mineral water and five different kinds of apple juice. The results show that the electronic tongue system is a kind of mineral water. Based on the test of mineral water and fruit juice, it is proved that the prototype of the electronic tongue is effective and feasible. In this paper, the hardware design of the electronic tongue in the future is summarized and prospected.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP368.1;TP181;TP212.9
本文编号：2497370
[Abstract]:The electronic tongue is an electronic detection instrument for qualitative analysis of liquid by a sensor array, a pattern recognition mode and a multivariate statistical method. Can be used to detect the taste of different liquid foods and to "visualization" different taste information. Therefore, the electronic tongue is a new hot spot in the field of food detection. The main purpose of this study is to explore the software and hardware of the electronic tongue system with low cost and stability, and to realize the feasibility of the system by a series of related experiments and data analysis. The subject is of great importance to the research and development of the rapid and non-destructive inspection in the food field. This topic has developed a set of electronic tongue system, which mainly consists of four-part ion-selective electrode sensor array, two-phase signal conditioning module, one-time data acquisition module and one-time data analysis module. The software is developed, and the system can be controlled, and the collected data can be analyzed and used. The main research contents are as follows: Aspects: (1) The principle of the commonly used electronic tongue sensor and the general principle are introduced. The electronic tongue structure is designed. (2) The multi-channel high-impedance small-signal amplifying circuit for the ion-selective electrode is designed, the signal of the electronic tongue sensor is filtered and amplified, an analog-to-digital signal conversion board is designed based on the C51MCU, and the sensor can be collected and sensed. and (3) using the C language and the VC ++ platform, the program of the lower computer and the upper computer is written, the data transmission and the human-machine interaction are realized, and the control function of the instrument. (4) In the environment of MATLAB7.11, a complete set of data processing programs, including PCA, LDA, and BP, PNN, RBF neural network and its parameter optimization tools, are written based on the theory of machine learning, based on the Lib-SV The pattern recognition software of the M-kit, the SVM parameter optimization tool based on the two-dimensional search and the SVM parameter optimization tool based on the GAOT genetic algorithm toolbox, based on the LS-SV The pattern recognition software of the M-kit. The number of electronic tongues is written based on C language. BP neural network processing program. Finally, it is the original ET GUI. The electronic tongue system is used to analyze and pattern the data of four different brands of mineral water and five different kinds of apple juice. The results show that the electronic tongue system is a kind of mineral water. Based on the test of mineral water and fruit juice, it is proved that the prototype of the electronic tongue is effective and feasible. In this paper, the hardware design of the electronic tongue in the future is summarized and prospected.
【学位授予单位】：浙江大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP368.1;TP181;TP212.9
【参考文献】
相关期刊论文 前10条
1 张夏宾;王晓萍;;基于调幅脉冲扫描法的电子舌及其在酒类识别中的应用[J];传感技术学报;2007年03期
2 席自强;单纯形-模拟退火算法[J];湖北工学院学报;2000年01期
3 滕炯华,王磊,袁朝辉;基于电子舌技术的果汁饮料识别[J];测控技术;2004年11期
4 蒋行国;侯洪彬;陈真诚;;基于主元分析的电子舌味觉检测[J];科技导报;2011年23期
5 许建华,张学工,李衍达;支持向量机的新发展[J];控制与决策;2004年05期
6 田师一;邓少平;;多频脉冲电子舌对酒类品种区分与辨识[J];酿酒科技;2006年11期
7 王俊,胡桂仙,于勇,周亦斌;电子鼻与电子舌在食品检测中的应用研究进展[J];农业工程学报;2004年02期
8 杜锋,雷鸣;味觉识别及其应用[J];中国调味品;2003年01期
9 唐慧敏;任麒;沈慧凤;;苦味评价方法的国内外研究进展[J];中国新药杂志;2009年02期
10 姜叔平;离子电极分析仪中的高阻抗转换电路[J];中国医疗器械杂志;1997年03期
相关硕士学位论文 前4条
1 胡洁;人工味觉系统——电子舌的研究[D];浙江大学;2002年
2 雷勇杰;基于虚拟仪器的电子舌及其应用研究[D];武汉理工大学;2006年
3 王伟广;基于虚拟仪器的电子舌系统研究[D];东北电力大学;2009年
4 马钰华;金属离子选择性电极的研究及应用[D];西南大学;2010年
，本文编号：2497370
本文链接：https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/2497370.html
