Voltammetric Electronic Tongue In The Detection Of Edible Oil

Edible oil consumption takes up large proportion in materialconsumption which is the necessity for making delicacies, in recent years,edible oil adulterate have repeatedly occurred, the quality of edible oil isvarious on the market, which is even more aroused the attention ofconsumers for edible oil detection, and developing a new detectionmethod has become significant.Electronic tongue system is made of the signal acquisition device,the signal processing device and the pattern recognition device. Since itappears, it is widely used in a variety of areas, among whichvoltammetric electronic tongue, potentiometric electronic tongue,light-addressable electronic tongue are the most widely used.This subjectis mainly about edible oil detection. Firstly, detecting edible oil in two traditional methods include conductivity method and peroxide acid valuemethod. Through analysis showed from these two methods, it can onlydistinguish the inferior oil from the normal edible oil, neither for theedible oil types. As a result, creating a new method is urgent and has animportant significance.In this paper, the signal acquisition device adopts three-electrodesystem: working electrode (gold electrode), reference electrode (saturatedcalomel electrode) and auxiliary electrode (platinum electrode). Signalacquisition device also includes CHI type electrochemical workstation,which selects experiment technology for cyclic voltammetry technology.By constructing the voltammetric electronic tongue system foredible oil types detection, this article chooses five different crops fromedible oil including soybean oil, corn oil, peanut oil, sesame oil,sunflower oil, using two different pretreatment methods of different cyclic voltammetric curve. It proves that voltammetric electronic tonguecan distinguish the types of edible oil through the feature extraction, andusing canonical correlation analysis, factor analysis and principalcomponent analysis, clustering methods use hierarchical clusteringalgorithm and mixed leapfrog clustering algorithm. Mixed leapfrogclustering algorithm is better.By constructing the voltammetric electronic tongue system formixed edible oil detection. This subject adopts homemade mixed edibleoil (normal edible oil and repeated fried old oil), which uses cyclicvoltammetric curve with different blending proportions. It proves thatvoltammetric electronic tongue can distinguish the types of mixed oilthrough the feature extraction, which uses canonical correlation analysis,factor analysis and principal component analysis, clustering methods usehierarchical clustering and artificial fish swarm clustering algorithm. Artificial fish swarm clustering algorithm is better. Then using partialleast square method to predict oil blending ratio and some valuableconclusions are drawn.