卷烟制叶丝段不同生产线产品感官质量差异的工艺特征参数提取方法
|
摘要
为了解决卷烟制叶丝段不同生产线同一牌号产品感官质量存在差异等问题,提出了一种基于改进Fisher判别分析和参数贡献度评价的工艺特征参数提取方法。采用改进Fisher判别分析提取制叶丝段优秀生产线和问题生产线同一牌号工艺参数数据的最大分离方向,进而沿最大分离方向计算不同生产线的相似度,并根据优秀生产线的相似度计算控制限。如果问题生产线相似度超出控制限的占比大于给定置信值,计算问题生产线各工艺参数的差异贡献度,将本次迭代中贡献度最大的工艺参数作为特征参数,通过迭代提取获得导致感官质量差异的所有工艺特征参数。基于某特定牌号对不同生产线的工艺参数数据进行实验验证,结果表明:该方法提取的工艺特征参数更加科学和完备,通过量化问题生产线工艺参数的差异贡献度,有效提高了不同生产线多参数差异分析的可靠性。该方法可为制叶丝段不同生产线相同工艺路径下的工艺参数优化提供技术支持。
In order to minimize the sensory quality differences of cigarettes made from cut tobacco processed by different production lines,a method for extracting the characteristic technical parameters in cut strip processing was proposed based on revised Fisher discriminant analysis and evaluation of parameter contribution.The maximum discrete direction of technical parameter data between a typical production line and a questioned production line in cut strip processing section was extracted via revised Fisher discriminant analysis,then the similarity of each line was calculated along the maximum discrete direction,and the control limit was determined on the basis of the similarity of the typical line.If the similarity of the questioned line exceeding the determined control limit with a proportion larger than the given confidence value,then the contribution of variation of each technical parameter of the questioned production line should be calculated,the technical parameter with the largest contribution in this iteration was taken as one of the characteristic parameters.The results of validation showed that the characteristic technical parameters extracted by the proposed method were more scientific and comprehensive,the reliability of multi-parameter variation analysis for different production lines was effectively improved via quantifying the contribution of technical parameter variation of questioned production line.This method provides a theoretical support for the adjustment of the cut strip processing technology in different production lines of the same processing route.
In order to minimize the sensory quality differences of cigarettes made from cut tobacco processed by different production lines,a method for extracting the characteristic technical parameters in cut strip processing was proposed based on revised Fisher discriminant analysis and evaluation of parameter contribution.The maximum discrete direction of technical parameter data between a typical production line and a questioned production line in cut strip processing section was extracted via revised Fisher discriminant analysis,then the similarity of each line was calculated along the maximum discrete direction,and the control limit was determined on the basis of the similarity of the typical line.If the similarity of the questioned line exceeding the determined control limit with a proportion larger than the given confidence value,then the contribution of variation of each technical parameter of the questioned production line should be calculated,the technical parameter with the largest contribution in this iteration was taken as one of the characteristic parameters.The results of validation showed that the characteristic technical parameters extracted by the proposed method were more scientific and comprehensive,the reliability of multi-parameter variation analysis for different production lines was effectively improved via quantifying the contribution of technical parameter variation of questioned production line.This method provides a theoretical support for the adjustment of the cut strip processing technology in different production lines of the same processing route.
引文
[1]李山,郭春生,叶亚军,等.制丝在线加工工艺批次间差异性评价[J].农产品加工,2017(1):51-54.LI Shan,GUO Chunsheng,YE Yajun,et al.The difference between the preparation of tobacco online processing technic[J].Farm Products Processing,2017(1):51-54.
[2]李涛,管仲,马路.卷烟制丝生产的柔性制造初探[J].烟草科技,2006(12):20-22,25.LI Tao,GUAN Zhong,MA Lu.Preliminary discussion on flexible manufacturing in tobacco primary processing[J].Tobacco Science&Technology,2006(12):20-22,25.
[3]罗登山,陈良元,杨忠文,等.关于卷烟柔性制丝的几点思考[J].烟草科技,2010(1):5-6,17.LUO Dengshan,CHEN Liangyuan,YANG Zhongwen,et al.Making tobacco primary processing more flexible[J].Tobacco Science&Technology,2010(1):5-6,17.
[4]郝喜良,熊晓敏,万敏,等.利用多特征相似度法分析异地加工卷烟产品均质化[J].烟草科技,2009(10):17-20.HAO Xiliang,XIONG Xiaomin,WAN Min,et al.Consistency analysis of cigarettes manufactured by different factories with multi-feature similarity[J].Tobacco Science&Technology,2009(10):17-20.
[5]石凤学,张涛,张强,等.基于多元质量控制限的卷烟均质化评价方法研究[J].陕西科技大学学报,2013,31(1):10-14.SHI Fengxue,ZHANG Tao,ZHANG Qiang,et al.Research on the homogenization evaluation method of cigarette based on multivariate quality control limit[J].Journal of Shaanxi University of Science&Technology,2013,31(1):10-14.
[6]陶永峰,张胜华,李文璟,等.基于层次分析和灰色关联分析的评价模型在卷烟多点加工质量评价中的应用[J].中国烟草学报,2017,23(1):43-49.TAO Yongfeng,ZHANG Shenghua,LI Wenjing,et al.Application of AHP and GRA-based models to quality evaluation of cigarettes from different production sites[J].Acta Tabacaria Sinica,2017,23(1):43-49.
[7]文杰,唐着宽,杨明权,等.过程能力分析在制丝工艺技术改进中的应用[J].烟草科技,2012(3):12-16.WEN Jie,TANG Zhaokuan,YANG Mingquan,et al.Application of process capability analysis in upgrading of primary processing[J].Tobacco Science&Technology,2012(3):12-16.
[8]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.
[9]Wang W,Zhao C H,Sun Y X,et al.A variant fisher discriminant analysis algorithm and its application to fault diagnosis[C]//Proceedings of the 10th Asian Control Conference(ASCC).Kota Kinabalu,Malaysia:IEEE,2015:1-6.
[10]任磊,韦徵,龚春英,等.电力电子电路功率器件故障特征参数提取技术综述[J].中国电机工程学报,2015,35(12):3089-3101.REN Lei,WEI Zheng,GONG Chunying,et al.Fault feature extraction techniques for power devices in power electronic converters:a review[J].Proceedings of the CSEE,2015,35(12):3089-3101.
[11]陈华坤,章卫国,史静平,等.航空电子设备故障预测特征参数提取方法研究[J].西北工业大学学报,2017,35(3):364-373.CHEN Huakun,ZHANG Weiguo,SHI Jingping,et al.Research on feature extraction method for fault prediction of avionics[J].Journal of Northwestern Polytechnical University,2017,35(3):364-373.
[12]丁美宙,熊安言,马宇平,等.叶丝滚筒干燥工艺参数对卷烟香气风格的影响[J].烟草科技,2015,48(5):74-79.DING Meizhou,XIONG Anyan,MA Yuping,et al.Effects of technical parameters in cylinder drying on aroma style of cigarette[J].Tobacco Science&Technology,2015,48(5):74-79.
[13]Simoglou A,Martin E B,Morris A J.Statistical performance monitoring of dynamic multivariate processes using state space modelling[J].Computers&Chemical Engineering,2002,26(6):909-920.
[2]李涛,管仲,马路.卷烟制丝生产的柔性制造初探[J].烟草科技,2006(12):20-22,25.LI Tao,GUAN Zhong,MA Lu.Preliminary discussion on flexible manufacturing in tobacco primary processing[J].Tobacco Science&Technology,2006(12):20-22,25.
[3]罗登山,陈良元,杨忠文,等.关于卷烟柔性制丝的几点思考[J].烟草科技,2010(1):5-6,17.LUO Dengshan,CHEN Liangyuan,YANG Zhongwen,et al.Making tobacco primary processing more flexible[J].Tobacco Science&Technology,2010(1):5-6,17.
[4]郝喜良,熊晓敏,万敏,等.利用多特征相似度法分析异地加工卷烟产品均质化[J].烟草科技,2009(10):17-20.HAO Xiliang,XIONG Xiaomin,WAN Min,et al.Consistency analysis of cigarettes manufactured by different factories with multi-feature similarity[J].Tobacco Science&Technology,2009(10):17-20.
[5]石凤学,张涛,张强,等.基于多元质量控制限的卷烟均质化评价方法研究[J].陕西科技大学学报,2013,31(1):10-14.SHI Fengxue,ZHANG Tao,ZHANG Qiang,et al.Research on the homogenization evaluation method of cigarette based on multivariate quality control limit[J].Journal of Shaanxi University of Science&Technology,2013,31(1):10-14.
[6]陶永峰,张胜华,李文璟,等.基于层次分析和灰色关联分析的评价模型在卷烟多点加工质量评价中的应用[J].中国烟草学报,2017,23(1):43-49.TAO Yongfeng,ZHANG Shenghua,LI Wenjing,et al.Application of AHP and GRA-based models to quality evaluation of cigarettes from different production sites[J].Acta Tabacaria Sinica,2017,23(1):43-49.
[7]文杰,唐着宽,杨明权,等.过程能力分析在制丝工艺技术改进中的应用[J].烟草科技,2012(3):12-16.WEN Jie,TANG Zhaokuan,YANG Mingquan,et al.Application of process capability analysis in upgrading of primary processing[J].Tobacco Science&Technology,2012(3):12-16.
[8]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.
[9]Wang W,Zhao C H,Sun Y X,et al.A variant fisher discriminant analysis algorithm and its application to fault diagnosis[C]//Proceedings of the 10th Asian Control Conference(ASCC).Kota Kinabalu,Malaysia:IEEE,2015:1-6.
[10]任磊,韦徵,龚春英,等.电力电子电路功率器件故障特征参数提取技术综述[J].中国电机工程学报,2015,35(12):3089-3101.REN Lei,WEI Zheng,GONG Chunying,et al.Fault feature extraction techniques for power devices in power electronic converters:a review[J].Proceedings of the CSEE,2015,35(12):3089-3101.
[11]陈华坤,章卫国,史静平,等.航空电子设备故障预测特征参数提取方法研究[J].西北工业大学学报,2017,35(3):364-373.CHEN Huakun,ZHANG Weiguo,SHI Jingping,et al.Research on feature extraction method for fault prediction of avionics[J].Journal of Northwestern Polytechnical University,2017,35(3):364-373.
[12]丁美宙,熊安言,马宇平,等.叶丝滚筒干燥工艺参数对卷烟香气风格的影响[J].烟草科技,2015,48(5):74-79.DING Meizhou,XIONG Anyan,MA Yuping,et al.Effects of technical parameters in cylinder drying on aroma style of cigarette[J].Tobacco Science&Technology,2015,48(5):74-79.
[13]Simoglou A,Martin E B,Morris A J.Statistical performance monitoring of dynamic multivariate processes using state space modelling[J].Computers&Chemical Engineering,2002,26(6):909-920.