A flexible fiber model was established and incorporated into the coupled method of computational fluid dynamics (CFD) and discrete element method (DEM) for the study of air separation characteristics of flexible fiber biomass particles in a fluidized bed. The effects of operating and equipment parameters, including gas superficial velocity, fiber inlet velocity, fiber volume fraction, and equipment size, on separation efficiency were systematically explored. A clustering criterion with a dimensionless parameter kappa correlating the length, width, and fiber inlet velocity of the air separator was proposed to determine the outcome of separation and the probability of clustering in the fluidized bed, and a phase diagram of clustering behavior was created based on kappa. When kappa is greater than 1.2, no significant clusters are observed and the separation extent is large. When kappa is less than 1.2, distinct clusters and a lower level of separation are obtained. The clustering index kappa can be used to optimize the structural and operational parameters of vertical air classifiers for flexible fiber biomass particles, reducing clustering probability and increasing separation efficiency.

The fine control over the moisture content of tobacco filaments is critical for the manufacture of cigarettes in the tobacco industry, which is one of the most important factors affecting tobacco quality. In this work, the hygroscopic properties of four types of tobacco filaments were investigated on a multi-sample dynamic sorption and desorption apparatus under the conditions mimicking the production environment, and moreover, their surface and cross-sectional morphologies were examined using scanning electron microscopy. Results show that four tobacco filaments differ significantly in the surface and pore structures. The rate of hygroscopicity increases with the relative humidity. The rate of moisture absorption is primarily governed by diffusion within the pore structure of the tobacco; the greater the humidity, the more pronounced the effect. Moreover, the moisture adsorption process of tobacco is an exothermic process based on isothermal results, fitted using different models, among which GAB and Oswin models were found to be two of the most suitable for fitting the adsorption results with GAB having the slightly better goodness-of-fit. Our findings provide a thermodynamic and kinetic basis for optimizing the tobacco processing technology. Specifically, the equilibrium moisture content of tobacco can be predicted using the Oswin model, which is closely dependent on the pore structure of tobacco filaments.

Accurately predicting processing intensity is essential for controlling moisture and enhancing taste quality in tobacco during drum drying. However, owing to the scarcity of available processing intensity data, the predictive accuracy of conventional models is significantly constrained under varying operating conditions. This study proposes a few-shot prediction method for the processing intensity of tobacco drum drying, incorporating an attention-enhanced Time-series Generative Adversarial Network (TimeGAN) and physical constraints. To address the challenge posed by the limited availability of historical data, a TimeGAN model incorporating multi-head self-attention mechanisms is designed for generating high-fidelity synthetic sequences of drum drying processing intensity. This method improves the model's the model's capacity to effectively capture key dynamic variations during operational transitions, thereby improving the modeling of non-stationarity, dynamic variation, and strong nonlinearity tobacco drum drying processes. Subsequently, a physically-constrained Long Short-Term Memory (LSTM) is developed by embedding domain-specific physical laws governing processing intensity into the network architecture. This ensures that predictions are consistent with underlying physical principles and enhances model interpretability. Empirical evaluations on industrial drum drying datasets indicate that the proposed method substantially out performs traditional deep learning.

The uniformity of the flow field and cut tobacco distribution inside the airflow dryer seriously affects the stability of the moisture content of the dried cut tobacco and the sensory quality of the product. The physical field, motion trajectory, and physical state of the cut tobacco inside the HDT3 (High-Speed Drying Tube) airflow drying machine are analyzed using numerical simulation methods. And the flow field structure inside the equipment has been improved by adding flow deflectors at critical locations. The average residence time of the cut tobacco in the HDT3 airflow dryer is about 6.48 s, with a residence time of about 0.378 s in the riser. The temperature difference of cut tobacco with different particle sizes is significant, and the outlet particle temperature gradually decreases as the size increases. The average temperature of cut tobacco at the outlet of the dryer is 371.7K. There is segregation phenomenon at the bend of the cut tobacco, and the average segregation rate at the outlet of the riser is 1.819. The deflector can significantly improve the flow field and particle distribution uniformity inside the riser. After adding the deflector, the velocity deviation of rear sections decreases by 14.47%, 30.29%, 24.06%, and 20.18%, respectively; The cut tobacco segregation degree at the outlet of the riser is reduced to 0.997, and the uniformity of the riser outlet section was improved by about 48.7%.

With rising demands for tobacco product quality and smoking experience, precise modeling of internal void structures in cigarettes has become critical for process optimization. This study proposes a novel ball-and-stick network reconstruction method for filterless cigarettes by optimizing the Maximal Ball Algorithm (MBA) with density-driven adaptive thresholds. Highresolution micro-CT scanning combined with Bayesian-optimized throat tolerance adjustment enables accurate quantification of 3D topological features. After grayscale segmentation and morphological filtering for noise removal, key parameters were extracted: void radii follow a left-skewed distribution (dominant peak at 0.4 mm), throat lengths concentrate near 1.25 mm, and coordination numbers average 3.1 (primarily $\lt 7$), indicating poor connectivity and low permeability. Isolated spheres exhibit uniform spatial distribution but higher frequency at the ignition end, with significant volume ratio mutations at both ends potentially affecting combustion stability. The proposed approach overcomes characterization bottlenecks in tobacco fiber-void composites, establishing a high-precision framework for predicting draw resistance and optimizing manufacturing parameters.

Biobased porous materials such as porous biochar or plant-based porous fiber are excellent candidates for applications in the fields of catalysts, energy, environment, etc. Porosity is the pivotal microstructural characteristic of these biomaterials, as it governs not only the capacity to accommodate metal-based active components, but also regulates the diffusion of target molecules. Currently, due to the lack of advanced characterization techniques and statistical analysis algorithms, comprehensive analysis of the nanoscopic porosity of biomaterials and the correlation between porosity and attributes like growth location and origin is often lacking. This results in a limited understanding of the pore structures in these materials. This study takes tobacco biomass as an example to reveal the correlation between microstructure and properties through electron microscopy and mercury intrusion methods, coupled with principal component analysis. The results reveal consistent pore structures across different bake tobacco samples, with an uneven distribution of pore sizes. Bake tobacco from upper and middle plant parts exhibit higher density compared to lower parts, and variations of porosity exist among bake tobaccos from different regions. The rich porous microstructure of bake tobacco based biomaterials has been systematically revealed. This research provides valuable insights for understanding microstructures of biobased porous materials, facilitating improvements of macroproperties. Furthermore, it establishes a foundation for interpreting the microstructure and macroproperties, paving a way for novel design of biobased porous materials for a broad range of applications.

Anomaly recognition in manufacturing is crucial for quality control, especially in the cigarette-rolling process, which involves high-volume output, high-speed operations, and significant costs due to downtime. However, existing methods for anomaly recognition in cigarette-rolling often rely heavily on prior knowledge, pay limited attention to interactions among multiple indicators, and provide inadequate interpretability of the root causes of anomalies. To address these limitations, we propose CR-GSL, a graph structure learning-based method for anomaly recognition in the cigarette-rolling process. The method includes: (1) Initializing a cigarette-rolling graph structure to capture the spatial characteristics of the production process; (2) Embedding learnable indicator features to account for diverse and potentially complex inter-indicator relationships; (3) Performing graph structure learning, leveraging hyperbolic space to measure similarities among indicator embeddings and applying graph sparsity to refine structure; (4) Propagating graph information via dynamic attention mechanisms tailored to the cigarette-rolling process, facilitating node feature extraction and indicator prediction; and (5) Recognizing anomalies based on prediction outcomes, utilizing the learned graph structure to enhance interpretability. In approximately five hours of real-world production, our method achieves state-of-the-art performance on the test set (Precision = 0.926, Recall = 0.950, F1 = 0.938). Further analysis indicates that abnormal cigarette draw resistance is caused by gas source pressure instability, while abnormal air leakage results from excessively low rolling plate temperatures.

The arrangement of filaments in cigarettes has an important impact on the combustion performance of cigarettes. In order to quantitatively characterize the spatial orientation of filaments, it is necessary to establish a measure of the main direction of filament arrangement, and then analyze the differences in the arrangement of different types of cigarettes. Due to the lack of non-destructive means of observing the morphology of the filaments inside the cigarette and the difficulty of analyzing the morphology due to the intertwining of the filaments, the industry currently lacks an effective means of measuring and analyzing the ordering rate of the filaments. In order to address this problem, the study uses high-resolution micro-CT scanning imaging technology to scan the cigarettes, three-dimensional reconstruction of two-dimensional cigarettes tomographic images obtained from scanning based on the filtered inverse projection method, and proposes a method for detecting the ordering rate of the tobacco filaments in the cigarettes based on the average normal vector of the point cloud. After the optimization of scanning resolution and sample penetration, the radial measurement accuracy of the investigated CT scanning imaging method is better than 0.06 mm with relative standard deviation RSD $<1 \%$, and the axial measurement accuracy is better than 0.24 mm with RSD $<1 \%$. Exploring the ordered rate of cigarette filaments of several specifications, it was found that the angle between the total filling direction of fine cigarette filaments and the axial line of the cigarette was about 43.9°, with an average ordered rate of 51.22 %, the angle between the total filling direction of medium cigarette filaments and the axial line of the cigarette was about 33.18°, with an average ordered rate of 63.14 %, and the angle between the total filling direction of regular cigarette filaments and the axial line of the cigarette was about 31.27°, with an average The ordering rate is 65.09 %, showing the distribution of high in the middle and low in the front and back, and the ordering rate of conventional cigarette is higher than that of medium cigarette, and the ordering rate of medium cigarette is higher than that of fine cigarette. This method provides a new means to analyze the physical distribution of tobacco inside the cigarette and to improve the filling process.

This study proposes an improved motion model for tobacco particles in a rotary drum dryer and presents a method to estimate their residence time based on known structural and operating parameters. By accounting for multiple movement forms and estimating the real-time position of particles, the model enables accurate prediction of residence time, which is essential for precise moisture and temperature control during tobacco processing. Simulation results show that, when structural parameters are fixed, residence time and contact heating time are more sensitive to drum rotation speed than to hot air velocity. Thus, adjusting the drum speed proves to be an effective strategy for optimizing moisture control in the drying process.

Electron tomography is employed to conduct a comprehensive porosity analysis of the biomass tobacco and its derived carbon materials.

Biomass and its derivatives have broad applications in the fields of bio-catalysis, energy storage, environmental remediation. The structure and components of biomass, which are vital parameters affecting corresponding performances of derived products, need to be fully understood for further regulating the biomass and its derivatives. Herein, tobacco is taken as an example of biomass to introduce the typical characterization techniques in unraveling the structural information, chemical components, and properties of biomass and its derivatives. Firstly, the structural information, chemical components and application for biomass are summarized. Then the characterization techniques together with the resultant structural information and chemical components are introduced. Finally, to promote a wide and deep study in this field, the perspectives and challenges concerning structure and composition charaterization in biomass and its derivatives are put forward.

Utilization of waste tobacco stems holds significant environmental and economic value. This study developed a hierarchical porous carbon matrix from waste tobacco stems via pyrolysis with FeCl3 activation. The resulting tobacco stem-derived porous carbon (TPC)-anchored Ru nanoparticles, demonstrating significant activity in the hydrogen evolution reaction (HER) in acid and alkaline electrolytes. The optimal catalyst, Ru/TPC-900, produced at 900℃ with 1.5-wt

Tobacco, a widely cultivated crop, has been extensively utilized by humans for an extended period. However, the tobacco industry generates a significant amount of organic waste, and the effective utilization of this tobacco waste has been limited. Currently, most tobacco waste is either recycled as reconstituted tobacco sheets or disposed of in landfills. However, tobacco possesses far more potential value than just these applications. This article provides an overview of the diverse uses of tobacco waste in agriculture, medicine, chemical engineering, and energy sectors. In the realm of agriculture, tobacco waste finds primary application as fertilizers and pesticides. In medical applications, the bioactive compounds present in tobacco are fully harnessed, resulting in the production of phenols, solanesol, polysaccharides, proteins, and even alkaloids. These bioactive compounds exhibit beneficial effects on human health. Additionally, the applications of tobacco waste in chemical engineering and energy sectors are centered around the utilization of lignocellulosic compounds and certain fuels. Chemical platform compounds derived from tobacco waste, as well as selected fuel sources, play a significant role in these areas. The rational utilization of tobacco waste represents a promising prospect, particularly in the present era when sustainable development is widely advocated. Moreover, this approach holds significant importance for enhancing energy utilization.

The cigarette draw resistance monitoring method is incomplete and single, and the lacks correlation analysis and preventive modeling, resulting in substandard cigarettes in the market. To address this problem without increasing the hardware cost, in this paper, multi-indicator correlation analysis is used to predict cigarette draw resistance. First, the monitoring process of draw resistance is analyzed based on the existing quality control framework, and optimization ideas are proposed. In addition, for the three production units, the cut tobacco supply (VE), the tobacco rolling (SE), and the cigarette-forming (MAX), direct and potential factors associated with draw resistance are explored, based on the linear and non-linear correlation analysis. Then, the correlates of draw resistance are used as inputs for the machine learning model, and the predicted values of draw resistance are used as outputs. Finally, this research also innovatively verifies the practical application value of draw resistance prediction: the distribution characteristics of substandard cigarettes are analyzed based on the prediction results, the time interval of substandard cigarettes being produced is determined, the probability model of substandard cigarettes being sampled is derived, and the reliability of the prediction result is further verified by the example. The results show that the prediction model based on correlation analysis has good performance in three months of actual production.

为研究松散润叶筒内部热湿环境分布,通过构建真实比例的三维松散润叶筒模型,采用计算流体力学(CFD)方法数值模拟松散润叶筒内部的三维传热传质过程,并根据温度和含水率对喷雾喷射角和筒体倾角进行优化.结果表明:模拟值与实验值整体吻合良好,温度平均相对误差为 0.6%,所建立的 CFD 模型真实有效;轴向气相温度呈阶段式变化,前半段温度迅速升高,后半段温度稳定在341.0 K左右;雾化水和水蒸气向烟叶持续传质传热,使得轴向烟叶温度、含水率逐步提高,最终出料口烟叶温度达 326.3 K,含水率达17.21%;筒体倾角为3°、喷雾喷射角为45°时,烟叶加湿加热效果最佳.

为确定片烟在离散元仿真模拟中的合理模型参数取值,更好地计算出片烟在松散润叶设备中的复杂运动,基于离散元软件(Rocky DEM),创建了与以往球形黏结刚性颗粒模型不同的柔性片烟壳体模型,并分析了柔性片烟壳体模型形变特点.以物理和仿真试验的堆积角为对比对象,选取"Hysteretic Linear Spring"法向接触模型、"Linear Spring Coulomb Limit"切向接触模型以及"Constant Adhesive Force"黏附力模型,对仿真接触参数进行标定.再通过对片烟本征参数测定、查阅文献及Rocky DEM说明手册,分析得到了片烟的离散元参数取值范围,设计Plackett-Burman、最陡爬坡以及Box-Behnken试验,确定了影响柔性片烟堆积角的显著性参数以及堆积角的拟合方程.以堆积角物理试验35.83°为目标值,得到显著性参数的最优解:片烟与片烟间静摩擦系数为0.67,片烟与片烟间滑动摩擦系数为0.31,片烟与片烟黏间黏附力系数为0.2,其余非显著性参数取平均值.仿真试验与物理试验相对误差为0.78%,从而验证了模型的准确性与真实性,可为工业生产进一步计算柔性片烟在松散润叶装置内的运动提供理论依据.

Biotemplates are often used in the preparation of catalysts due to availability of raw materials and environment-friendly. Herein, three biomorphic CeO2 were prepared from waste tobacco materials and impregnated with 1 wt% Ru for the catalytic combustion of vinyl chloride. The results showed that the biotemplates greatly affected the properties of the catalysts. 1Ru/CeO2-YS exhibited the best catalytic activity and superior stability because of its abundant oxygen vacancies, sufficient acidic sites and the optimal redox properties. SEM images suggested that CeO2-YS and CeO2-YM maintained the basic structure of the biotemplates. Moreover, a possible reaction mechanism over Ru/CeO2 was proposed.

本文介绍了国内8家实验室研究基于图像处理的烟丝宽度测量方法及ISO 20193:2019的精密度及检测效率.结果表明:(1)基于图像处理的烟丝宽度测量方法得到的精密度与ISO 20193:2019处于同一数量级,重复性标准差(S r)与烟丝宽度的比值小于5％,再现性标准差(S R)与烟丝宽度的比值小于8％;(2)基于图像处理的烟丝宽度测量方法比ISO 20193:2019的检测效率高.

切丝宽度是卷烟制丝过程的重要参数,是影响卷烟品质的主要因素之一.综述了切丝宽度对卷烟品质在烟丝结构、填充值、烟支物理指标、烟气及感官品质等方面的研究进展,以期为卷烟制丝工艺提供一定的理论基础及依据.

目前在卷烟研发过程中,实验室烟丝加料加香小样制样环节仍多采用手工配料配香,存在精度低、均匀性差,及烘烤后烟丝水分精度无法精准控制等问题,突出体现为中小样制样与大线中样存在品质差异,这是一个行业共性问题.我们研制了一台可实现实验室卷烟小样制样过程自动化配料、配香、加料、加香及烘丝功能的高精度自动化加料加香烘丝一体化微型设备.应用效果显示:1)其调配精度与调配中心大线生产的精度基本一致.2)微型设备调配料、香与大线生产调配料、香相似度分别为99.11％、98.26％;微型设备烟丝加料加香均匀度为98.79％;3)经感官评价,微型设备处理烟丝制成的卷烟感官质量与大线生产样差异值为0.5分,明显优于人工操作效果.说明该设备能有效模拟大线生产过程,有利于提高研发效率.