电子鼻技术及其应用研究进展  

Advancements in Electronic Nose Technology and Its Applications

作　　者：穆申玲 沈文锋[2,3] 吕大伍 宋伟杰 谭瑞琴[1] MU Shenling;SHEN Wenfeng;LYU Dawu;SONG Weijie;TAN Ruiqin(Faculty of Electrical Engineering and Computer Science,Ningbo University,Ningbo 315211,Zhejiang,China;Ningbo Institute of Material Technology&Engineering,Chinese Academy of Sciences,Ningbo 315201,Zhejiang,China;University of Chinese Academy of Sciences,Beijing 100049,China)

机构地区：[1]宁波大学信息科学与工程学院,浙江宁波315211 [2]中国科学院宁波材料技术与工程研究所,浙江宁波315201 [3]中国科学院大学,北京100049

出　　处：《材料导报》2024年第14期52-65,共14页Materials Reports

基　　金：浙江省自然科学基金项目(LGF22F010008);浙江省基础公益研究计划项目(LGG21F040001);宁波市重点研发计划项目(2023Z021)。

摘　　要：电子鼻(Electronic nose,E-nose)技术作为一种有效的嗅觉模拟与气体识别的方法被广泛应用。电子鼻系统由气体传感器阵列组成,利用其交叉敏感性对气体进行检测。电子鼻系统利用机器学习算法,对气体进行定性定量分析。传统的机器学习算法在电子鼻系统中的应用已经成熟,如今深度学习算法也慢慢在电子鼻系统中应用。电子鼻系统具有选择性高、精密度好、反应快速、稳定性和延展性好的特点,被应用于包括有毒气体检测、空气质量管理、食品新鲜度和质量预测等方面。本文从气体传感器阵列的组成、信号采集与处理单元、模式识别算法的分类以及电子鼻系统在实际中的应用等方面综述了电子鼻系统气体识别的最新研究进展,最后对电子鼻系统气体识别目前所存在的问题以及发展前景进行了总结和展望。Electronic nose(E-nose)technology has been widely used as an effective method of olfactory simulation and odor recognition.The E-nose system is composed of an array of gas sensors that utilize their cross-reactivity for qualitative and quantitative gas analysis.Machine learning algorithms are commonly used in the E-nose system,with artificial neural networks being particularly prevalent.In addition to traditional machine learning algorithms,deep learning algorithms are also gradually being applied in E-nose systems.The E-nose system features high selectivity,precision,rapid response,stability,and scalability,and has been applied in various fields,such as toxic gas detection,air quality management,and food freshness and quality prediction.This paper reviews the latest research progress in the gas identification of E-nose systems,including the composition of gas sensor arrays based on MEMS technology,signal acquisition and processing units,and the classification of pattern recognition algorithms.Finally,the current problems and development prospects of E-nose system gas identification are summarized and discussed.

关 键 词：传感器阵列 信号采集与处理 模式识别算法 电子鼻技术 气体识别 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]