Digital Evaluation of Cigarette Smoke Clumping Based on Aerosol Particle Analysis

Abstract

Cigarette smoke clumping, an important indicator of smoke status, draws consumer attention due to its correlation with smoke concentration, thickness, fullness, and the clumping index, which significantly affects the sensory quality of cigarettes. Due to the difference of individuals and the influence of environment, the current manual evaluation method is greatly influenced by subjectivity, and the consistency and reproducibility of the evaluation results are poor. In order to solve this problem, this study takes clusteriness as a breakthrough point in the digital research of sensory evaluation. Firstly, the best-selling cigarettes are selected as the research object to form a consistent evaluation of clusteriness through expert evaluation. Then, from the microscopic point of view, that is, the particle size distribution of flue gas, the influence degree of flue gas clumping is explored. In the experiment, the diameter particle spectrometer was used to detect the aerosol diameter particles, and the characteristics of the diameter particle distribution curve, including the spectral shape area, the spectral peak intensity and the mean width, were further verified, and the correlation was proved. Finally, a model of granularity distribution and cluster relationship is constructed based on the feature engineering method, and the expert evaluation score is used as the expected value of the model for training, and finally the cluster digital evaluation system is realized.

Keywords

• Cigarette smoke clumping
• size distribution
• characteristic correlation
• digital evaluation