| 266 | 1 | 162 |
| 下载次数 | 被引频次 | 阅读次数 |
【目的】提高卷烟配方原料分类分档的工作效率，帮助配方人员客观掌握配方原料的年度间波动。【方法】提出一种热分析图谱结合支持向量机算法（SVM）对卷烟配方原料进行分类分档的新技术。采用python3中sklearn中的SVM模块，通过核函数和one-against-all方法并选择合适的惩罚参数，对2016—2018年度的129个样品进行训练学习，并对33个样品进行测试。当模型预测准确率达到应用需求后，对2019年度46个卷烟配方原料进行分类预测。【结果】129个训练集样品的分类准确率为93.02%,33个样品测试集样品的分类准确率为84.85%,46个卷烟配方原料的分类准确率为84.78%。
Abstract：[Background] This study aims to improve the work efficiency of classification and grading of cigarette blend raw materials, and help formula personnel to objectively grasp the annual fluctuation of formula raw materials. [Methods] A new technique is proposed for classifying and grading cigarette blend raw materials by combining thermal analysis spectra with the Support Vector Machine(SVM)algorithm. The SVM module in sklearn of python3 is used, and through the kernel function and one-against-all method and selecting appropriate penalty parameters, 129 samples from the years 2016-2018 are trained, and 33 samples are tested. After determining the prediction accuracy of the model, the model was applied to the classification evaluation of 46 samples in 2019. [Results] The classification accuracy rate of the 129 training set samples was 93.02%, the classification accuracy rate of the 33 test set samples was 84.85%, and the classification accuracy rate of the 46 cigarette blend raw materials was 84.78%.
[1]李石头，廖付，何文苗，等.基于近红外光谱相似的烟叶替代与卷烟配方维护[J].烟草科技，2020,53(2):88-93.LI Shitou, LIAO Fu, HE Wenmiao, et al. Tobacco substitution and cigarette blend maintenance based on near infrared spectral similarity[J]. Tobacco Science&Technology, 2020, 53(2):88-93.
[2]姚小龙.打叶复烤模块化加工的探索[J].质量探索，2016,13(2):32,31.YAO XiaoLong. Exploration on modular processing of beating leaf re-baking[J]. Quality Exploration, 2016, 13(2):32, 31.
[3]胡静宜，杨永锋，刘茂林，等.不同配打模块中各尺寸片烟化学成分的变化[J].西北农林科技大学学报，2020,48(6):30-38,51.HU Jingyi, YANG Yongfeng, LIU Maolin, et al. Changes of chemical composition with different batching modules among different sizes of tobacco[J]. Journal of Northwest A&F University,2020,48(6):30-38,51.
[4]张建平，陈江华，束茹欣，等.近红外信息用于烟叶风格识别及卷烟配方研究的初步探索[J].中国烟草学报，2007,13(5):1-5.ZHANG Jianping, CHEN Jianghua, SHU Ruxin, et al. Tobacco characteristics identification and blending formula study by using NIRs[J]. Acta Tabacaria Sinica,2007,13(5):1-5.
[5]刘晓萍，李斌，于川芳，等.基于近红外光谱的卷烟配方结构识别[J].烟草科技，2006(10):16-18.LIU Xiaoping, LI Bin, YU Chuanfang, et al. Recognition of cigarette blend make-up based on NIR spectrum[J]. Tobacco Science&Technology, 2006(10):16-18.
[6]刘剑君，申金媛，张乐明，等.基于红外光谱的烟叶自动分级研究[J].激光与红外，2011, 41(9):986-990.LIU Jianjun, SHEN Jinyuan, ZHANG Leming, et al. Study on automatic classification of tobacco based on infrared spectrum[J].Laser&Infrared, 2011, 41(9):986-990.
[7] Ni LJ, Zhang LG, Xie J, Luo JQ. Pattern recognition of Chinese flue-cured tobaccos by an improved and simplified K-nearest neighbors classification algorithm on near infrared spectra[J]. Anal Chim Acta. 2009;633:43-50.
[8] Wang D, Xie L, Yang SX, Tian F. Support vector machine optimized by genetic algorithm for data analysis of near-infrared spectroscopy sensors[J]. Sensors(Switzerland). 2018, 18:3222-34.
[9] Zhang F, Zhang X. Classification and quality evaluation of tobacco leaves based on image processing and fuzzy comprehensive evaluation[J]. Sensors. 2011, 11:2369-84.
[10]薛芳，施丰成，李耀光，等.基于电子鼻的星座图聚类分析法在烟叶分类中的应用[J].烟草科技，2015, 48(7):25-28.XUE Fang, SHI Fengcheng, LI Yaoguang, et al. Classifying flue-cured tobacco leaves with electronic nose-based constellation diagram clustering analysis[J]. Tobacco Science&Technology,2015, 48(7):25-28.
[11]杨超，惠建权，王芳，等.基于因子分析及曲线估计的卷烟叶组化学成分品质评价[J].烟草科技，2017, 50(12):77-84.YAO Chao, HUI Jianquan, WANG Fang, et al. Quality evaluation of chemical components in cigarette blend based on factor analysis and curve estimation[J]. Tobacco Science&Technology, 2017,50(12):77-84.
[12]李超，刘秀明，秦云华，等.基于挥发性特征组分的卷烟烟丝质量趋势分析[J].烟草科技，2017, 50(11):39-47.LI Chao, LIU Xiuming, QIN Yunhua, et al. Quality trend analysis on cigarette cut filler based on characteristic volatile components in cut filler[J]. Tobacco Science&Technology, 2017, 50(11):39-47.
[13] Lin X, Sun L, Li Y, et al. A random forest of combined features in the classification of cut tobacco based on gas chromatography fingerprinting[J]. Talanta Elsevier B.V, 2010, 82:1571-5.
[14] Gu L, Xue LC, Song Q. Classification of the fragrant style and evaluation of the aromatic quality of flue-cured tobacco leaves by machine-learning methods[J]. Bioinform Comput Biol, 2016, 14:1650033.
[15]杨卿，武书彬.麦草的热失重特性及动力学[J].农业工程学报，2009, 25(3):193-197.YANG Qing, WU Shubin. Thermogravimetric properties and kinetic analysis of wheat straw[J]. Transactions of the Chinese Society of Agricultural Engineering, 2009, 25(3):193-197.
[16]张乾，向欣宁，梁丽彤，等.热重分析煤焦恒温气化过程中气体切换的影响[J].化工进展，2022, 41(1):166-173.ZHANG Qian, XIANG Xinning, LIANG Litong, et al. Gas switch effect on the isothermal gasification reaction of coal char in thermogravimetric analyser[J]. Chemical Industry and Engineering Progress, 2022, 41(1):166-173.
[17]李巧灵，王辰琨，吴清辉，等.烟草热失重质量损失与苯酚释放量的关系[J].烟草科技，2015, 48(12):41-47.LI Qiaoling, WANG Chenkun, WU Qinghui, et al. Relationship between thermal mass loss and phenol release of tobacco during pyrolysis process[J]. Tobacco Science&Technology, 2015, 48(12):41-47.
[18] Qiaoling Li, Quanxing Zheng, Xiaohua Deng, et al. Numerical simulation of the burning process in a king-size cigarette based on experimentally derived reaction kinetics, Beitr?ge zur Tabakforschung International, 2020, 29(3):156-179.
[19]李巧灵，陈昆焱，刘泽春，等.基于热重的烟叶热解差异度分析[J].烟草科技，2017, 50(8):75-79.LI Qiaoling, CHEN Kunyan, LIU Zechun, et al. TGA-based analysis of pyrolysis differential between different tobacco samples[J]. Tobacco Science&Technology, 2017, 50(8):75-79.
[20]李巧灵，陈昆焱，邓小华，等.基于烟叶热解差异度分析的烟叶替代方法[J].烟草科技，2018, 51(8):77-84.LI Qiaoling, CHEN Kunyan, DENG Xiaohua, et al. Method of tobacco substitution based on differential analysis of tobacco pyrolysis[J]. Tobacco Science&Technology, 2018, 51(8):77-84.
[21]何伟，范坚强，牟文君，等.片烟四段式气调绿色醇化技术体系的构建[J].中国烟草学报，2020,26(6):32-44.HE Wei, FAN Jianqiang, MU Wenjun, et al. Construction of green aging technology system for flue-cured tobacco strips using four-stage controlled atmosphere method[J]. Acta Tabacaria Sinica,2020,26(6):32-44.
[22]罗登山，王兵，乔学义.《全国烤烟烟叶香型风格区划》解析[J].中国烟草学报，2019,25(4):1-9.LUO Dengshan, WANG Bing, QIAO Xueyi. Explanation of national regionalization of leaves style of flue-cured tobacco[J].Acta Tabacaria Sinica, 2019,25(4):1-9.
[23] Yin C, Deng XH, Yu ZQ, et al. Auto-classifcation of biomass through characterization of their pyrolysis behaviors using thermogravimetric analysis with support vector machine algorithm:case study for tobacco[J], Biotechnology for Biofuels, 2021, 14(1):106.
[24] Corinna C, Vladimir V. Support Vector Networks[J]. Mach. Learn,1995, 20(3):273-297.
[25] Hsu C W, Lin C J. A comparison of methods for multiclass support vector machines[J]. IEEE Trans, 2002. 13(2):415-425.
[26] Yuan G X, Ho C H, Lin C J. Recent advances of large-scale linear classification[J]. Proc IEEE, 2012, 100(9):2584-2603.
[27] Hsu C W, Chang C C, Lin C J. A practical guide to support vector classification. Technical report, University of National Taiwan,Department of Computer Science and Information Engineering,(2003), pp. 1-12.
基本信息:
DOI：10.16472/j.chinatobacco.2022.T0236
中图分类号:TS452
引用信息:
[1]何伟,杨韧强,刘泽春等.基于热分析图谱的卷烟配方原料分类分档模型构建与应用[J].中国烟草学报,2024,30(02):1-10.DOI:10.16472/j.chinatobacco.2022.T0236.
基金信息:
福建中烟工业有限公司科技项目“片烟质量热分析技术数字化评价模型的构建及应用”（FJZYJH2021ZD007）