基于图像识别的叶丝形态表征及其数值模拟OA北大核心CSTPCD

[Background and Purpose]This study investigates a method for the characterization of particle size and shape of cut tobacco,which can be used for modeling in its numerical simulation.[Method]Image analysis technology is used to determine the equivalent diameter of the area circle,maximum Feret diameter,circularity,and aspect ratio of each cut tobacco strand dispersed on a two-dimensional plane.[Results](1)Using the proposed method of cut tobacco particle image acquisition and processing,and through statistical analysis of cut tobacco sample data,the probability density distribution curves of particle size and shape of cut tobacco materials were determined.This achieved the characterization of particle size and shape for cut tobacco materials of four different lengths:long,medium,short,and broken strands.The characterization parameters quantitatively reflect the differences in particle size and shape of different samples.(2)Numerical simulation results using the discrete element method show that the particle size and shape properties of cut tobacco have a significant impact on the material's stacking height.Based on the experimentally measured particle size and shape probability distribution curves,a numerical model of cut tobacco particles consistent with the actual morphological distribution was generated.The deviation of numerical simulation results from experimental results is within 1.3%.[Conclusion]The proposed method quantitatively characterizes cut tobacco particles in terms of particle size and shape.On one hand,the characterization results reflect the differences in properties of different materials,and on the other hand,they can be directly used to establish numerical models close to actual conditions,thereby accurately predicting the stacking height of cut tobacco particles of different lengths.