电子鼻多组分气体检测算法与性能研究  

Algorithm and Performance Study of Electronic Nose for Multi-component Gas Detection

作　　者：李龙[1] 应滨州 李华曜[1] 刘欢[1] LI Long;YING Binzhou;LI Huayao;LIU Huan(School of Integrated Circuits,Huazhong University of Science and Technology,Wuhan National Laboratory for Optoelectronics,Optics Valley Laboratory,Wuhan 430074,China)

机构地区：[1]华中科技大学集成电路学院,武汉光电国家研究中心,湖北光谷实验室,湖北武汉430074

出　　处：《仪表技术与传感器》2023年第11期1-6,11,共7页Instrument Technique and Sensor

基　　金：国家自然科学基金项目(61904062);湖北省自然科学基金项目(2022CFA035)。

摘　　要：多组分气体检测在工业和环境监测、人体呼出气检测等领域需求广泛。通过对多组分气体的快速检测分析,有助于保障工业生产安全、保护生态环境。电子鼻系统是一种重要的多组分气体检测技术,它能够模拟人类嗅觉,对多种气体或气味进行快速识别。文中针对人体呼出气中常见气体NO_(2)、乙醇和甲醛,通过电子鼻动态配气系统配制不同浓度的单一和混合气体,重点研究电子鼻信号特征提取与多组分气体检测算法,实现了对目标气体的可靠检测,具有较高的定性识别和定量分析准确性,为多组分气体检测技术研究提供了思路和手段。Multi-component gas detection is widely demanded in industries,environmental monitoring,human breath analysis and other fields.Rapid detection and analysis of multi-component gases can help ensure industrial safety and protect the ecological environment.The electronic nose system is an important technology of multi-component gas detection that can simulate human olfaction and rapidly identify various gases or odors.In this paper,aiming at the common gases NO_(2),ethanol,and formaldehyde in human breath,by using an electronic nose dynamic gas mixing system to prepare different concentrations of single and mixed gases,the signal feature extraction and multi-component gas detection algorithm of the electronic nose system was studied.Reliable detection of the target gas is achieved with high qualitative identification and quantitative analysis accuracy,providing ideas and means for the research of multi-component gas detection technology.

关 键 词：多组分气体 动态配气系统 电子鼻 传感器阵列 特征提取 算法 模型 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]