基于非参数核密度估计的卷烟配方相似性评价方法研究
Research on cigarette blend similarity evaluation method based on non-parametric kernel density estimation
-
摘要: 【目的】 提出一种基于非参数核密度估计(KDE)的配方相似性评价方法,以提高卷烟叶组配方维护效率。【方法】 通过历史配方数据构建单项化学成分的KDE分布模型,计算单项化学成分相似性;采用不同权重分配策略,比较实验配方与历史配方的相似度;基于相关性分析评估模型的性能,分析叶组配方整体相似性与感官质量之间的规律。【结果】 KDE拟合的概率密度估计函数,能较好地反映历史维护配方的化学成分数据分布;通过评审的叶组配方与实验叶组相比,糖碱比、总氮含量、还原糖含量3项指标与历史配方的整体相似性均值分别提升了30.57%,18.97%,13.06%;虽然4种赋权方法的相关性相近,但它们在单项化学成分相似性相关程度的倾向性上表现不同,其中博弈论组合,赋权法在多数感官指标上的相关性介于主观、客观赋权之间,可降低主客观赋权间的综合权重偏差。【结论】 通过历史配方数据构建KDE分布模型,以评估实验配方与历史配方的相似性,为评估和调控配方维护过程中叶组的相似性与稳定性提供参考。Abstract: 【Objective】 To enhance the efficiency of cigarette blend formulation maintenance, this study proposes a method based on non-parametric kernel density estimation (KDE) for evaluating the similarity of cigarette blend formulations. 【Methods】 A KDE distribution model for individual chemical components was constructed using historical formulation data to calculate the similarity of individual chemical components. Different weight allocation strategies were employed to compare the similarity between the experimental formulation and historical formulations. The model’s performance was evaluated using correlation analysis, and the relationship between the overall similarity of leaf group formulations and sensory quality was analyzed. 【Results】 The probability density estimation function fitted using KDE effectively reflects the distribution of chemical component data from historical formulations. Compared with the experimental leaf groups, the average overall similarity of the leaf group formulations that passed evaluation to historical formulations in three indicators (sugar-alkaloid ratio, total nitrogen content, and reducing sugar content) increased by 30.57%, 18.97%, and 13.06%, respectively. Although the four weighting methods exhibited similar correlation levels, they differed in their tendency regarding the correlation degree of individual chemical component similarity. Specifically, the game theory-based combined weighting method showed correlations with most sensory indicators that were intermediate between those of subjective and objective weighting methods, thereby reducing the overall weight deviation between these two approaches. 【Conclusion】 Constructed a KDE distribution model using historical formulation data to evaluate the similarity between experimental formulations and historical formulations. The results provide a reference for assessing and controlling the similarity and stability of leaf groups during the formulation maintenance process.
-
-
[1]
朱亚昆,梅吉帆,郭文孟,等.基于PPF投影算法和高光谱技术的卷烟牌号识别模型[J].轻工学报,2024,39(4): 118-126.
ZHU Y K,MEI J F,GUO W M,et al.Cigarette brand identification model based on PPF projection algorithm and hyperspectral technology[J].Journal of Light Industry,2024,39(4):118-126. -
[2]
冯润泽,雒兴刚,张忠良,等.基于单料片烟互信息的卷烟配方维护方法[J].烟草科技,2021,54(3):65-71.
FENGR Z,LUO X G,ZHANG Z L,et al.Cigarette blend maintenance based on mutual information from single-gradetobaccostrips[J].TobaccoScience&Technology,2021,54(3):65-71. -
[3]
李巧灵,陈昆焱,邓小华,等.基于烟草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 Q L,CHEN K Y,DENG X H,et al.Method of tobacco substitution based on differential analysis of tobacco pyrolysis[J].Tobacco Science & Technology,2018,51(8):77-84. -
[4]
李晓,张浩博,郭朋玮,等.不同规格卷烟配方叶丝柔软性与物理质量的关联性研究[J].中国烟草学报,2024,30(4):33-43.
LI X,ZHANG H B,GUO P W,et al.Study on the correlation between softness and physical quality of formula tobacco in cigarettes of different specifications[J].Acta Tabacaria Sinica,2024,30(4):33-43. -
[5]
郝贤伟,帖金鑫,何文苗,等.基于近红外光谱-感官评价的巴西烟叶风格模拟及替代[J].烟草科技,2018,51(10): 83-89.
HAO X W,TIE J X,HE W M,et al.Style simulation and substitution of Brazilian tobacco based on near infrared spectrum and sensory evaluation[J].Tobacco Science & Technology,2018,51(10):83-89. -
[6]
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 S T,LIAO F,HE W M,et al.Tobacco substitution and cigarette blend maintenance based on near infrared spectral similarity[J].Tobacco Science & Technology,2020,53(2):88-93. -
[7]
龙明海,李长昱,向虎,等.支持向量机在卷烟配方原料分组中的应用[J].云南农业大学学报(自然科学),2017,32(6): 1045-1051.
LONG M H,LI C Y,XIANG H,et al.Applications of support vector machine in tobacco blend grouping of chinese type cigarette[J].Journal of Yunnan Agricultural University (Natural Science),2017,32(6): 1045-1051. -
[8]
田震,许强,苗晨琳,等.基于配方功能的烟叶原料动态化平衡预测及分配方法探索[J].中国烟草学报,2023,29(1):1-10.
TIAN Z,XU Q,MIAO C L,et al.Exploration of dynamic equilibrium prediction and distribution method of tobacco raw materials based on formula function[J].Acta Tabacaria Sinica,2023,29(1):1-10. -
[9]
杨超,万里兴,杨盛刚,等.卷烟叶组化学成分分析及与配方维护的关系[J].烟草科技,2019,52(4): 76-84.
YANG C,WAN L X,YANG S G,et al.Analysis of chemical component in blended tobacco and relationship between chemical component and cigarette blend maintenance[J].Tobacco Science & Technology,2019,52(4):76-84. -
[10]
王楠,雒兴刚,张忠良,等.基于非负矩阵分解的卷烟配方维护方法[J].烟草科技,2019,52(8): 67-76.
WANG N,LUO X G,ZHANG Z L,et al.Cigarette blend maintenance based on non-negative matrix factorization [J].Tobacco Science & Technology,2019,52(8):67-76. -
[11]
吴喜之,赵博娟.非参数统计[M].北京: 中国统计出版社,2019. WU X Z,ZHAO B J.Nonparametric Statistics[M].Beijing: China Statistics Press,2019.
-
[12]
杨瑞君,张楚,杨评,等.非参数核密度估计模型预测双酚A的物种敏感度分布规律[J].生态毒理学报,2024,19(4):120-130.
YANG R J,ZHANG C,YANG P,et al.Application of non-parametric kernel density estimation model in prediction of species sensitivity distribution of bisphenol A[J].Asian Journal of Ecotoxicology,2024,19(4):120-130. -
[13]
蒋佳磊,陆扬,苏燕,等.我国主要烟叶产区烤烟化学成分特征与可用性评价[J].中国烟草学报,2017,23(2):13-27.
JIANG J L,LU Y,SU Y,et al.Chemical characteristics and usability of flue-cured tobacco growing in main tobacco producing areas in China [J].Acta Tabacaria Sinica,2017,23(2):13-27. -
[14]
国家烟草专卖局.烟草及烟草制品 淀粉的测定 连续流动法:YC/T 216—2013[S].北京:中国标准出版社,2013. State Tobacco Monopoly Administration.Tobacco and tobacco products-Determination of starch-Continuous flow method:YC/T 216—2013[S].Beijing:Standards Press of China,2013.
-
[15]
国家烟草专卖局.烤烟 烟叶质量风格特色感官评价方法:YC/T 530—2015[S].北京:中国标准出版社,2015. State Tobacco Monopoly Administration.Flue-cured tobacco-Sensory evaluation methods for quality and style characteristics of tobacco leaves:YC/T530—2015[S].Beijing: China StandardsPress,2015.
-
[16]
XU S G,CUI Y X,YANG C X,et al.The fuzzy comprehensive evaluation (FCE) and the principal component analysis (PCA) model simulation and its applications in water quality assessment of Nansi Lake Basin,China[J].Environmental Engineering Research,2021,26(2): 200022.
-
[17]
王晓宾,王军,刘兰,等.广东烟区烤烟主要化学成分可用性评价[J].中国烟草科学,2019,40(2): 64-72.
WANG X B,WANG J,LIU L,et al.Chemical component usability of flue-cured tobacco leaves in Guangdong tobacco-growing area[J].Chinese Tobacco Science,2019,40(2):64-72. -
[18]
汪璇,卞建锋,杨民烽,等.基于综合赋权法的四川省烤烟化学成分可用性时空差异评价 [J].中国农学通报,2021,37 (22): 131-138.
WANG X,BIAN J F,YANG M F,et al.Evaluation of temporal and spatial differences of flue-cured tobacco chemical composition usability in Sichuan province based on comprehensive weighting method[J].Chinese Agricultural Science Bulletin,2021,37 (22): 131-138. -
[19]
樊艳翔,雷社平,解建仓.广东省河流水体富营养化综合评价及分异特征: 基于博弈论组合赋权法与VIKOR模型[J].生态环境学报,2023,32(10): 1811-1821.
FAN Y X,LEI S P,XIE J C.Comprehensive evaluation and differentiation characteristics of eutrophication in river waters of Guangdong province: Based on game theory combined empowerment method and VIKOR model [J].Ecology and Environmental Sciences,2023,32(10): 1811-1821. -
[20]
吴瑞英,孙怀卫,严冬,等.基于格网数据和博弈论组合赋权的中巴经济走廊洪灾风险评价[J].农业工程学报.2021,37(14):145-154.
WU R Y,SUN H W,YAN D,et al.Evaluation of flood disaster risk in China-Pakistan Economic Corridor by combination weighting based on improved game theory and grid data[J].Transactions of the Chinese Society of Agricultural Engineering,2021,37(14): 145-154. -
[21]
王颖,冯承莲,穆云松,等.非参数核密度估计在铜、银物种敏感度分布中的应用[J].中国环境科学,2017,37(4): 1548-1555.
WANG Y,FENG C L,MU Y S,et al.Application of non-parametric kernel density estimation for developing species sensitivity distributions of copper and silver[J].China Environmental Science,2017,37(4): 1548-1555. -
[22]
李智冈,吕莉,谭德坤,等.基于加权核密度估计与微簇合并的密度峰值聚类算法[J].信息与控制,2024,53(3):302-314.
LI Z G,LYU L,TAN D K,et al.Density peaks clustering algorithm based on weighted kernel density estimation and micro-cluster merging[J].Information and Control,2024,53(3):302-314. -
[23]
韩忠成,林金官,汪红霞.基于局部多项式展开的多元非参数模型贝叶斯带宽选择[J].数理统计与管理,2020,39(1): 93-103.
HAN Z C,LIN J G,WANG H X.Bayesian bandwidth selection for multivariate nonparametricregression models based on local polynomial expansion[J].Journal of Applied Statistics and Management,2020,39(1): 93-103. -
[24]
兰博,田喆,牛纪德,等.基于TRNSYS仿真数据的平板型太阳能集热器净得热非参数核密度估计概率模型[J].太阳能学报,2021,42(2): 1-7.
LAN B,TIAN Z,NIU J D,et al.Probabilistic model of nonparametric kernel density estimation for useful energy gain of flat-plate solar collector based on trnsys simulation data[J].Acta Energiae Solaris Sinica,2021,42(2):1-7. -
[25]
李淑艳,翟友邦,王小龙,等.基于核密度估计的拖拉机传动轴载荷外推方法[J].中国农业大学学报,2021,26(10): 175-184.
LI S Y,ZHAI Y B,WANG X L,et al.An extrapolation method of tractor propeller shaft load based on kernel density estimation[J].Journal of China Agricultural University,2021,26(10):175-184. -
[26]
张晓英,张晓敏,廖顺,等.基于聚类与非参数核密度估计的风电功率预测误差分析[J].太阳能学报,2019,40(12): 3594-3604.
ZHANG X Y,ZHANG X M,LIAO S,et al.Prediction error analysis of wind power based on clustering and non-parametric kernel density estimation [J].Acta Energiae Solaris Sinica,2019,40(12):3594-3604. -
[27]
杨超,丁根胜,万里兴,等.基于烟叶化学成分的卷烟叶组感官质量和风格研究[J].烟草科技,2018,51(8): 85-91.
YANG C,DING G S,WAN L X,et al.Relationships of main chemical components in tobacco blend of cigarette with its sensory quality and style[J].Tobacco Science & Technology,2018,51(8):85-91. -
[28]
褚旭,许强,殷瑜东,等.基于不同聚类分析方法的卷烟叶组配方模块分类[J].江苏农业科学,2023,51(15): 149-154.
CHU X,XU Q,YIN Y D,et al.Classification of formula module of cigarette blend based on different clustering analysis methods[J].Jiangsu Agricultural Sciences,2023,51(15):149-154.
-
[1]
-
点击查看大图
计量
- PDF下载量: 2
- 文章访问数: 83
- 引证文献数: 0
下载:
