公开(公告)号：US10895560B2

公开(公告)日：2021-01-19

申请号：US16651722

申请日：2018-03-18

Applicant: EAST CHINA UNIVERSITY OF SCIENCE AND TECHNOLOGY

Inventor： Daqi Gao ,  Zejian Wang ,  Xiaoqin Zhang ,  Jiamin Song

IPC: G01N33/00 ,  G01N1/24 ,  G06N3/08

Abstract: Provided is a sensory quality evaluation method for tobacco and tobacco products using an electronic nose instrument. The electronic nose instrument includes a gas sensor array module, an automatic smoke sampling system, a computer control and data analysis system, and an automatic ignition device. These components are integrated in a test box to make the instrument be structure miniaturization and work automation. The large tobacco data is established, in which the relationship between the responses of gas sensor array and the brand labels and sensory quality index scores by testing a large number of standard cigarette samples. A cascade type of modular neural networks is proposed with revised activation function and new decision-making and quantification rules to simulate the smoking and evaluating process of the professional panel. The electronic nose instrument and method realize on-site detection, discrimination and quality score estimation of a large number of tobacco and tobacco products.