| Summary: | The electronic nose technology with 10 sensor channels was used to determine and analyze the flavor components of red Zanthoxylum bungeanum from 6 different major producing areas in China, including Hanyuan Zanthoxylum bungeanum, for the identification of Hanyuan Zanthoxylum bungeanum. Principal Component Analysis (PCA), Linear Discriminate Analysis (LDA) and Partial Least Squares Discriminant Analysis (PLS-DA) were performed on 6 Zanthoxylums bungeanum from different origins using the differential analysis of distribution profile of flavor components of sensor responses, and the discriminant model of Hanyuan Zanthoxylum bungeanum was established. The results showed that both PCA and LDA models could distinguish Hanyuan Zanthoxylum bungeanum from other 5 red Zanthoxylums bungeanum, respectively. The "Yes or No" discrimination model of Hanyuan Zanthoxylum bungeanum was established with PLS-DA. The prediction ability parameter (Q2) and goodness of fit parameter (R2) of the model were 0.947 and 0.968, respectively. The results of cross validation and permutation showed that the identification model did not overfitting, and the identification model was robust and effective. At the same time, Hanyuan Zanthoxylum bungeanum doped other kinds of red Zanthoxylum bungeanum samples with 5%, 10%, 30% and 50% were determined and analyzed. LDA and PLS-DA identification models could accurately identify the doped samples from pure Hanyuan Zanthoxylum bungeanum. In this study, by means of electronic nose technology and stoichiometry theory and method, other red Zanthoxylum bungeanum from different origins and adulterated Hanyuan Zanthoxylum bungeanum was accurately identified with pure Hanyuan Zanthoxylum bungeanum, which provided a new technical reference for the protection of China National Geographic Indication Products, quality control and origin tracing of Hanyuan Zanthoxylum bungeanum.
|
|---|