Identification Of Aroma Scent Based On Electronic Nose Technology And Its Material Basis

Eaglewood is the resiniferous wood of Aquilaria agallocha Roxb. and Aquilaria sinensis (Lour.) Gilg.. Agarwood is the resiniferous wood of Aquilaria agallocha Roxb.. And Chinese eaglewood is the resiniferous wood of Aquilaria sinensis (Lour.) Gilg, which mainly produced in Guangdong、Guangxi and Hainan province. China and many other countries have treated eaglewood as rare medicinal materials and natural flavor since ancient times. Nowadays, there is fewer and fewer real eaglewood and more and more fake and inferior eaglewood in the markets. The traditional identification method of eaglewood, which mainly focuses on the odors of eaglewood, has unique advantages, but it is vulnerable. And the modern identification methods have their own problems. So the objectification of the identification of eaglewood’s odors is very important.In this research, eaglewood samples from different areas are collected and identified; the detection method of eaglewood’s odors using electronic nose(EN) is established; the methods of data feature extraction, data pretreatment and pattern recognition to identify agarwood, Chinese eaglewood and fake and inferior eaglewood are screened; eaglewood’s odors are analysed by HS-GC-MS and the correlation between HS-GC-MS data and EN data is studied to get material basis of eaglewood’s odors. Here’s the result:1 Sample collection and identificationSamples of 42 batches eaglewood are collected and identified with traditional method and TLC method. There are 10 batches Chinese eaglewood,9 batches agarwood and 23 batches fake and inferior eaglewood in the collected samples.2 Detection methods of eaglewood’s odors using electronic noseThe injection volume and incubation time are studied by the single factor method, and then particle size of grinded sample, weight of sample and incubation temperature are studied by orthogonal experiments with the differences between maximum EN sensors response of Chinese eaglewood and agarwood as the study index to establish the detection method of eaglewood’s odors using electronic nose. The established method is accurate and stable. The established method is used to carry out experiment with all 42 eaglewood samples.3 Data feature extraction, pretreatment and pattern recognition screeningUsed multivariate statistics to screen maximum value、mean value and mid value. Choose maximum value as the data feature extraction method of electronic nose.Divide the samples of 42 batches eaglewood into training set and test set. Then used the raw data、normalized data and standardized data of the training set to train five different artificial neural networks (BP、RBF、GRNN、PNN、LVQ). Then optimize the parameters of ANN with the classification accuracy of test set as the study index. And the classification accuracy of test set of BP is the highest,96.30%, with optimized parameters.4 material bases of eaglewood’s odors8 samples of Chinese eaglewood and 8 samples of agarwood are studied by HS-GC-MS.33 components are detected by HS-GC-MS. And there is relevance between 9 of them and EN data. The 9 components are Thujopsene、4-ethenyl-1,2-dimethyl-benzene, valenca-1(10),8-dien-11-o1、α-bulnesene、eremophila-9,11(13)-dien-12-o1、2-(4 α,8-dimethyl-1,2,3,4,4a,5,6,7-octahydro-naphthalen-2-yl)-prop-2-en-l-o1、α-guaiene、jinkohol、 (S)-4α-methyl-2-(1-methylethyl)-3,4,4α,5,6,7-hexahydronaphthalene...