基于K-means聚类及模糊判别的卷烟包灰性能综合评价方法
Comprehensive evaluation method for ash characterization of burning cigarette based on K-means clustering and fuzzy discrimination
-
摘要: 为实现卷烟包灰性能的综合评价和评价结果具象化,以49个卷烟的灰色、裂口率、缩灰率、碳线宽度、碳线整齐度测定结果为原始变量,先运用K-means聚类、模糊判别法将原始变量转换为具象化的得分数据,再运用Critic赋权法赋予各项指标权重,建立了一种卷烟包灰性能综合评价方法。结果表明:将原始变量转换成区间为60~100、平均值在80左右的得分,可使评价结果具象化且更加符合认知习惯;5项指标的权重由高到低依次为裂口率(0.27) >缩灰率(0.25)>灰色(0.18)>碳线整齐度(0.16)>碳线宽度(0.14);卷烟包灰性能可划分为优、良、差三档,各档得分区间依次为(85,100]、[75,85]、[60,75);不同档次代表性卷烟的灰柱视觉效果对比结果证明,综合得分可客观反映卷烟包灰性能的优劣。Abstract: In order to achieve a comprehensive evaluation of ash characterization of burning cigarette and make the evaluate results concrete, the gray scale, crack rate, shrinkage rate, width of char line, and uniformity of char line measurement results of 49 cigarettes were used as the primitive variables. Firstly, K-means clustering and fuzzy discriminant analysis were used to convert the primitive variables into concrete score data, then the weight of each indicator was assigned by critical weighting method, and a comprehensive evaluation method was established. The results indicated that converting the primitive variables into scores ranging from 60 to 100 with an average value of around 80 could make the evaluate results concrete and consistent with cognitive habits. The weight of the five indicators, from high to low, was as follows: crack rate (0.27)> shrinkage rate (0.25)>gray scale(0.18)> uniformity of char line (0.16)> width of char line (0.14). The comprehensive ash characterization of burning cigarette could be divided into three levels: excellent, good, and poor, the score range of each level was (85,100], [75,85], and [60,75), respectively, the comparison of bar charts of representative cigarettes of different levels proved that the comprehensive score could objectively reflect the advantages and disadvantages of the ash characterization of burning cigarette.
-
Key words:
- cigarette /
- ash characterization /
- K-means clustering /
- fuzzy discriminant /
- critical weighting method
-
-
[1]
程占刚,叶明樵,胡素霞,等.影响卷烟包灰能力的因素研究[J].烟草科技,2011,44(2):9-12.
-
[2]
郭东锋,王孝峰,张劲,等.基于特征工程的卷烟包灰影响因素筛选及评价模型[J].烟草科技,2023,56(4):73-81.
-
[3]
唐小雪,黄宇亮,宾晖,等.基于感官符合度及卷烟纸参数的卷烟包灰性能改进研究[J].中国烟草学报,2023,29(3):21-33.
-
[4]
郑晗,詹建波,王浩,等.卷烟包灰性能的研究现状分析[J].新型工业化,2018,8(11):93-97.
-
[5]
王孝峰,张劲,李延岩,等.卷烟燃烧速率与其包灰性能关系分析[J].烟草科技,2021,54(10):70-76.
-
[6]
王孝峰,张劲,曹芸,等.卷烟燃烧参数、灰分外观和烟气成分相互关系研究[J].中国烟草学报,2022,28(5):17-22.
-
[7]
董浩,钟宇,王澍,等.基于计算机视觉的卷烟包灰检测方法[J].烟草科技,2024,57(5):91-97
,112. -
[8]
郑丰,肖翠翠,王小平,等.卷烟纸特性对卷烟静态包灰性能的影响[J].烟草科技,2020,53(3):82-88.
-
[9]
楚文娟,崔建华,王建民,等.基于卷烟纸参数的卷烟静态包灰综合评价方法[J].烟草科技,2021,54(11):69-76.
-
[10]
孙家政,姜红,刘新磊,等.差分拉曼光谱技术结合K-means聚类法对牙膏的快速分类[J].理化检验-化学分册,2022,58(1):84-89.
-
[11]
WANG X,SHAO Z W,SHEN Y C,et al.Research on fast marking method for indicator diagram of pumping well based on K-means clustering[J].Heliyon,2023,9(10):e20468.
-
[12]
李振永,杨彩玲,买自珍,等.基于CRITIC赋权法、模糊概率法评价萝卜品种在宁南山区栽培适应性[J].中国瓜菜,2023,36(9):102-107.
-
[13]
冯占科,钱旺,彭建川.基于综合权重——TOPSIS的露天矿边坡稳定性评价[J].工矿自动化,2023,49(S1):133-137.
-
[14]
陈胜可.SPSS统计分析从入门到精通[M].北京:清华大学出版社,2010.
-
[15]
杨倚奇.基于复相关——主成分分析的江苏高职院校人才竞争力指标筛选研究[J].江苏高职教育,2019,19(2):27-31.
-
[16]
邱盛媛,陈豪,申展,等.覆盖发热物质对雷竹林土壤化学性质和竹笋营养成分的影响[J].经济林研究,2023,41(2):234-241.
-
[17]
李庆贺.网球、羽毛球、乒乓球运动员灵敏素质构成因素的模型构建与比较研究[D].上海:上海体育学院,2023.
-
[18]
熊北辰,刘炳辰.省级地方政府债务风险指数的一种新构建方法[J].中国货币市场,2023(9):14-20.
-
[1]
-
点击查看大图
计量
- PDF下载量: 3
- 文章访问数: 238
- 引证文献数: 0
下载:
