常见食用油和煎炸食用油的激光诱导荧光光谱特性  

Laser Induced Fluorescence Spectrum Characteristics of Common Edible Oil and Fried Cooking Oil

作　　者：牟涛涛[1] 陈思颖[1] 张寅超[1] 陈和[1] 郭磐[1] 葛宪莹[1] 高丽蕾[1] 

机构地区：[1]北京理工大学光电学院,北京100081

出　　处：《光谱学与光谱分析》2013年第9期2448-2450,共3页Spectroscopy and Spectral Analysis

基　　金：国家自然科学基金项目(61178072)资助

摘　　要：为了实现地沟油的快速检测,利用激光诱导荧光探测技术搭建了食用油类快速检测系统。对多种食用油和煎炸食用油(地沟油的一种)进行了光谱采集,建立了多种食用油和煎炸食用油的荧光光谱数据库,发现煎炸食用油的特征荧光光谱与常见食用油之间存在较大差异,在此基础上利用主成分分析法和BP神经网络实现了油类识别、地沟油的快速检测,总体识别率高达97.5%。实验证明激光诱导荧光光谱技术具有快速非接触和灵敏度高等优点,与BP神经网络相结合能够实现油类的快速识别,成为地沟油快速检测的一种新方法。In order to detect the trench oil the authors built a trench oil rapid detection system based on laser induced fluorescence detection technology.This system used 355nm laser as excitation light source.The authors collected the fluorescence spectrum of a variety of edible oil and fried cooking oil（a kind of trench oil）and then set up a fluorescence spectrum database by taking advantage of the trench oil detection system.It was found that the fluorescence characteristics of fried cooking oil and common edible oil were obviously different.Then it could easily realize the oil recognition and trench oil rapid detection by using principal component analysis and BP neural network,and the overall recognition rate could reach as high as 97.5%.Experiments showed that laser induced fluorescence spectrum technology was fast,non-contact,and highly sensitive.Combined with BP neural network,it would become a new technique to detect the trench oil.

关 键 词：地沟油 食用油 激光诱导荧光(LIF) 主成分分析 BP神经网络 

分 类 号：O433.4[机械工程—光学工程]