基于BP神经网络的烟草制丝工艺参数优化研究
Research on optimization of tobacco silk processing parameters based on BP neural network
-
摘要: 为了提高烟丝质量,以烟丝物理特性、烟支物理特性和卷烟化学成分为质量控制指标,采用BP神经网络(BPNN)对烟草制丝工艺参数进行了优化,将BPNN优化结果与正交试验结果进行对比,并对优化后的参数进行验证。结果表明:所建BPNN模型预测结果具有较高的可靠性和准确性,通过BPNN预测的最优制丝工艺参数组合唯一且准确;通过正交试验方法得出的烟草制丝工艺参数最优组合在蒸汽流量和热风温度的参数设置上存在差异。烟草制丝工艺参数优化后,烟丝整丝率、填充值提升,碎丝率、卷烟单支质量标准差、烟支吸阻标准差、CO释放量、焦油释放量、烟碱释放量均降低,整体优化效果明显。BPNN对烟草制丝最优参数预测准确,避免了误判现象,提高了加工效率,降低了时间成本和资源浪费。Abstract: In order to improve the quality of tobacco silk, the processing parameters of tobacco silk were optimized by using BP Neural Network (BPNN), taking the physical properties of tobacco silk, the physical properties of cigarettes and the chemical composition of cigarettes as quality control indicators. The BPNN optimization results were compared with the orthogonal test results, and the optimized parameters were verified. The results showed that the prediction results of the established BPNN model had high reliability and accuracy, and the optimal silk-making processing parameters combination predicted by BPNN was unique and accurate. There were differences in the parameter settings of steam flow and hot air temperature for the optimal combination of tobacco silk-making process parameters obtained by the orthogonal test method. After the parameters of the tobacco silk-making processing parameters were optimized, the whole cut rate and filling value of tobacco silk had been improved, the broken cut rate had been reduced, the standard deviation of single cigarette weight and cigarette suction resistance had decreased, and the CO, tar, and nicotine releases had been reduced, and the overall optimization effect was obvious. BPNN accurately predicted the optimal parameters of tobacco silkmaking, which avoided misjudgment, improved processing efficiency, and reduced time cost and resource waste.
-
-
[1]
LEHMANN K,KUHN S,REIMER J.E-electronic cigarettes in germany: Patterns of use and perceived health improvement[J]. European Addiction Research,2017,23(3):136-147.
-
[2]
刘晓东,陈竺,陈洁,等.烟草制丝生产线人机交互界面设计研究[J].电气自动化,2020,42(2):116-118.
-
[3]
冯煜.优化烟草制丝设备提高烟丝质量[J].中小企业管理与科技(下旬刊),2020(4):176-177.
-
[4]
窦志.预测PID控制在烟草工业松散回潮工序中的应用[D].绵阳:西南科技大学, 2020.
-
[5]
聂铭,周冀衡,杨荣生,等.基于MIV-SVM的烤烟评吸质量预测模型[J].中国烟草学报,2014,20(6):56-62.
-
[6]
汪文良,范兴,朱雪峰,等.基于帕累托图的卷烟制丝工艺质量分析[J].现代农业科技,2021(13):235-239,243.
-
[7]
张馨予,严莉红,易小丽,等.不添加料液的混合型膨胀烟丝生产工艺及质量分析[J].中国烟草学报,2020,26(1):36-41
,47. -
[8]
邓羽翔,罗诚,李东亮,等.基于BP神经网络的烟叶醇化感官质量仿真模拟[J].食品与机械,2020,36(3):161-165.
-
[9]
卓鸣,汪鹏,望开奎.基于MIV-BP神经网络的成品烟丝质量预测模型构建[J].食品与机械,2021,37(12):161-166
,214. -
[10]
陆玉浩,王彬彬,万云飞,等.基于神经网络的烟丝填充值预测模型研究[J].中国烟草科学,2016,37(5):82-86.
-
[11]
温延,林敏,罗民,等.烟块分散装置的设计及应用[J].中国烟草学报,2018,24(5):130-132.
-
[12]
李秀芳.烟片松散回潮关键工艺参数过程控制系统的优化设计[J].中国烟草学报,2015,21(3):34-41.
-
[13]
BO M.Moisture control methods in silk reeling process of tobacco based on the random forest regression[J]. Open Journal of Statistics,2015,5(5):393-402.
-
[14]
张丽娜,许利平,樊虎,等.高精度自动化加料加香烘丝一体化微型设备的研制和应用[J].中国烟草学报,2018,24(4):27-34.
-
[15]
RENÉ H,NICOLA W,ANGELIKA S,et al.Low-tech, pilot scale purification of a recombinant spider silk protein analog from tobacco leaves[J].International Journal of Molecular Sciences,2016,17(10):1687.
-
[16]
田忠,陈闯,许宗保,等.制丝关键工序对细支卷烟燃烧温度及主流化学成分的影响[J].中国烟草学报,2015,21(6):19-26.
-
[17]
曹家升,王海明,吴龙元.烟包垂直分切机卸料装置的优化设计[J].烟草科技,2021,54(9):108-112.
-
[18]
和智君,王理珉,张强,等.云南烤烟品种加工性能评价[J].中国烟草科学,2013,34(4):72-75.
-
[19]
罗军,赵科文,王强春,等.薄板烘丝工艺参数对卷烟焦油释放量的影响[J].轻工学报,2016,31(4):76-81
,87. -
[20]
张新锋.基于ANP的卷烟制丝质量评价方法[J].郑州轻工业学院学报(自然科学版),2015,30(3/4):34
-38.
-
[1]
计量
- PDF下载量: 50
- 文章访问数: 2469
- 引证文献数: 0