Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.
Identification of the marked ages of rice wines by combining novel e-tongue based on conducting polymer nanocomposites modified electrodes and smartphone e-nose based on cloud platform
Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org
Citation: 2018 ASABE Annual International Meeting 1801186.(doi:10.13031/aim.201801186)
Authors: Yanan Yang, Zhenbo Wei
Keywords: E-tongue; E-nose; Rice wine; Wine age; Modified electrodes; Smartphone.
Abstract. Aroma and taste are the most important attributes of alcoholic beverages. In the study, a novel electronic tongue (e-tongue) and a smartphone electronic nose (e-nose) were combined to identify the marked ages of rice wines. Three types of conducting polymer nanocomposites modified electrodes were fabricated to compose the sensor array of the novel e-tongue, and two types of chronoamperometry were applied to the modified electrodes for recording the taste information through a standard three-electrode configuration. The e-nose was self-developed with 12 MOS sensors which have high sensitive to volatile organic compounds (VOCs) of rice wines, and a smartphone application was designed to control all operations and display the olfactory information (it also included the after taste information) of rice wines through wireless communication. The area under the response curves were taken as the feature data, and six feature datasets (e-tongue dataset, e-nose dataset, directed-fusion dataset, weighted-fusion dataset, optimized directed-fusion dataset and optimized weighted-fusion dataset) with different pattern recognition methods were applied for the identification of wine ages. Principal component analysis (PCA) and locality preserving projections (LPP) were used for classification. LPP performed better than PCA based on each dataset, and the best results was obtained by LPP based on the weighted-fusion dataset. Therefore, the weighted-fusion dataset was applied as independent variables of partial least squares regression, extreme learning machine and library for support vector machines (LIBSVM) for the prediction of the wine ages. All the methods performed well with high correlation coefficient (R2 > 0.99) based on both of the training and testing sets, and LIBSVM performed best in the prediction work.
(Download PDF) (Export to EndNotes)