造纸法再造烟叶涂布工序的智能预测与优化  

Intelligent Prediction and Optimization of Reconstituted Tobacco Leaf Coating Process in Papermaking Process

作　　者：李浥尘 王海同 曾伟[4] 胡波 陈丽舟[1,2] 唐天明 潘亚雄 周贤钢 李鹏 LI Yi-chen;WANG Hai-tong;ZENG Wei;HU Bo;CHEN Li-zhou;TANG Tian-ming;PAN Ya-xiong;ZHoU Xian-gang;LI Peng(China Tobacco Hubei Industrial LLC Wuhan,Hubei,430040 China;Hubei Xinye Reconstituted Tobacco Development Co.,Ltd.,Wuhan,Hubei,430056 China;Hubei Key Lab of Applied Research Institute of Reconstituted Tobacco Wuhan,Hubei,430040 China;School of Artifitial Intelligence and Automation,Huazhong University of Science and Technology Wuhan,Hubei,430074 China)

机构地区：[1]湖北中烟工业有限责任公司,湖北武汉430040 [2]湖北新业烟草薄片开发有限公司,湖北武汉430056 [3]重组烟叶应用技术研究湖北省重点实验室,湖北武汉430040 [4]华中科技大学人工智能与自动化学院,湖北武汉430074

出　　处：《纸和造纸》2023年第5期11-15,共5页Paper and Paper Making

摘　　要：涂布是重组烟草产品制造的一道重要工序。在涂布工序中,涂布率对产品质量有相当大的影响,是衡量产品涂布质量的重要指数。为使用参数优化方法解决涂布工序中涂布率的优化问题,选择了12个相关参数进行研究,并利用灰色关联分析方法分析其中关键的影响因素。基于反向传播网络算法提出模型,并结合了DE-ABC算法,对涂布率进行预测分析。实验结果表明,DE-ABC-BP网络模型对涂布率的预测准确率显著高于原BP网络模型。在此基础上,使用DE算法求解出该模型输出最优涂布率时的各参数,为今后的智能控制提供参考。Coating is an essential process in the production of reconstructed tobacco.Coating rate has a significant impact on production quality in the process,which makes it an important index to measure production quality.To solve the optimization problem of coating rate in the coating process using parameter optimization methods,12 parameters are taken into consideration,and grey relation analysis is utilized to analyze the key influence factor.The proposed model is based on Backward-Propagating neural network combined with a DE-ABC algorithm,in order to predict the coating rate.Research results show that the accuracy of the DE-ABC-BP model is significantly higher than that of original BP model.Based on that,DE algorithm is used to solve the optimal parameters when the model outputs the best coating rate,which could be a reference in the intelligent control in the future.

关 键 词：关联分析 涂布率 神经网络 人工蜂群 差分进化 

分 类 号：TB484[一般工业技术—包装工程] TS43[农业科学—烟草工业]