To elucidate the formation mechanism of the closed pressure drop of cigarette rods and to develop a predictive model, CT-based three-dimensional reconstruction technology, physical model, and gradient boosting decision tree algorithm were jointly employed to systematically analyze the closed pressure drop of 34 cigarette rod samples. Data quality was improved through preprocessing using a physical model, and the predictive performances of four types of models, namely traditional physical model, basic theoretical model, empirical-physical model, and machine learning model, were systematically compared. Using cigarette rods with a circumference of 20 mm as a case study. The results showed that: 1) Among the three types of physical models, the empirical-physical model showed the best predictive performance and provided a basis for data cleaning and feature screening. Using this model, outliers with residuals exceeding 100 Pa, a threshold of approximately 2.6 times the median absolute deviation, were identified and removed, thereby improving the physical plausibility of the dataset. 2) The gradient-boosted decision tree model achieved the best predictive performance on the independent test set (MAE=38.58 Pa, R2 =0.850, MAPE=7.08%), outperforming other types of physical models. 3) The equivalent pore diameter extracted from three-dimensional reconstruction was the dominant factor affecting the closed pressure drop, with a feature importance of 76.73%. Removing this feature increased the mean absolute error of prediction by 78.6%. 4) The equivalent aperture was significantly influenced by the interaction between the width of cut tobacco and stalk position of tobacco leaves, suggesting that the closed pressure drop can be targetedly controlled through process adjustments. Although validated only for demi-slim cigarettes, the proposed physically guided and data-driven integrated framework provides a new method for optimizing the entire process from raw material to processing in the tobacco industry, and offers a reference for the flow analysis of porous media.
To address over-reliance on manual experience and the lack of quantitative indexes for regulating the piled fermentation of domestic cigar tobacco leaves, the temporal change patterns of the pile core temperature was analyzed. A total of 45 domestic cigar tobacco piles were studied, and their core temperature sequences were monitored. Dynamic time warping algorithm (DTW) was used to measure the similarity between the different temperature sequences, followed by k-means++ algorithm for temperature time series clustering analysis. By analyzing the dynamic characteristics of the various temperature change patterns, a multi-factor decision-making framework integrating temperature change patterns, sliding 24-hour temperature differences, temperature thresholds and fermentation time was constructed. The results showed that: 1) During the piled fermentation process, the core temperature exhibited five different change patterns including stable low temperature, steady warming, rapid warming, stable high temperature, and medium temperature fluctuation. 2) For the stable low temperature type, it was recommended to turn over the pile when the fermentation time exceeded 120 hours and the core temperature was below 30 ℃. For the steady warming type, it was recommended to turn over the pile when the fermentation time exceeded 168 hours and the core temperature was below 35 ℃. For rapid warming or stable high temperature type, when the core temperature exceeded 45 ℃ or the sliding 24-hour temperature differences was ≤0 ℃, a pile turnover warning could be triggered, and for the medium temperature fluctuation type, when the cumulative fermentation time of the entire cycle exceeded 30 days, the current stage fermentation time exceeded 168 hours, and the sliding 24-hour temperature difference was ≤0 ℃, it could be determined that the fermentation was over and the product could be discharged. 3) The use of decision framework method to assist in pile turnover enabled early prediction of the fermentation endpoint, reduced the constant temperature or cooling time at the end of each fermentation stage, and improved fermentation efficiency. This study provides a quantitative decision-making strategy for cigar tobacco piling, pile turnover, and pile discharge, and offers technical support for the digital control and standardized production of cigar tobacco fermentation.
To address brand mark misregistration, particularly under variable-speed operating conditions, and the high maintenance workload of the online printing system in the ZJ119 filter cigarette maker, an online registration and rod cutoff system was developed, and the original online printing unit was removed. The system uses pre-printed cigarette paper, i.e., special cigarette paper on which the brand mark pattern is printed in advance and equally spaced detectable registration marks are incorporated. A color mark sensor was used to detect the spacing of registration markers in real time. Meanwhile, a phase compensation scheme based on a paper feed virtual axis and a position-type PID control algorithm with self-tuning based on speed intervals were introduced to dynamically adjust the cigarette rod cutoff position during operation. The system was evaluated using a cigarette brand produced at Hangzhou Cigarette Factory. The results showed that after the adoption of the pre-printed cigarette paper and the online registration and rod cutoff system, the cigarette reject rate under variable-speed operating conditions decreased from 9.0%-10.0% to 1.5%-2.0%, while the average maintenance time of the paper feeding system decreased from 31.0 minute/shift to 1.3 minute/shift. The system can support the improvement in cigarette appearance quality and the effective operating rate of the equipment, and provide the equipment capability required for filter cigarette making during further development and wider adoption of pre-printed cigarette paper.
To improve local detail recovery and overall structural representation capability in tobacco leaf vein segmentation, a collaborative segmentation network for tobacco leaf veins with main vein guidance and detail enhancement (DiffUNet-VEUNet, DVNet) was constructed. DVNet guided the structural modeling of the vein segmentation network by treating the main vein as a high-confidence structural prior, and strengthened the feature representation of fine veins by incorporating detail enhancement mechanism. The results showed that the constructed DVNet achieved superior segmentation performance compared with existing mainstream methods for tobacco leaf vein segmentation. While maintaining the stability of the main vein structure, DVNet improved the recognition capability of fine veins and further enhanced the overall segmentation quality. Furthermore, qualitative visualization analysis and ablation experiments further verified the effectiveness of each core module. Overall, DVNet achieves more stable and refined tobacco leaf vein segmentation.
To comprehensively and objectively evaluate the quality and stability of online laser perforation in slim cigarettes, a high-precision laser scanning technique was used to measure seven key indicators: hole length, hole width, hole depth, upper surface area, internal surface area, the distance from the mouth end to the first row of performations, and filter ventilation. The efficiency coefficient method was used to assign a quantitative score to each indicator, and the CRITIC (Criteria importance through intercriteria correlation) method was used to determine the weight of each indicator, thereby a comprehensive quality evaluation method for online laser perforation in slim cigarettes was established. The results showed that the proposed method enabled comprehensive evaluation of laser perforation quality of slim cigarettes of different brands based on the seven indicators. Brand C had the highest mean comprehensive score and the smallest standard deviation, indicating better laser perforation quality and process stability. The mean comprehensive score for laser perforation quality was positively correlated with the filter ventilation conformity rate, whereas the standard deviation of the comprehensive score was negatively correlated with the process performance index (Ppk). The developed method enables objective evaluation of online laser perforation quality in slim cigarettes and provides a practical basis for process optimization and process quality control.
