基于PCA-BP神经网络的烟叶含水率预测研究  

Research on the Prediction of Tobacco Water Content Based on PCA-BP Neural Network

作　　者：吴宏 孔泽栋 王若方 马松 WU Hong;KONG Ze-dong;WANG Ruo-fang(Huahuan International Tobacco Co.,Ltd.,Chuzhou,Anhui 239000;School of Smart City and Transportation,Southwest Jiaotong University,Chengdu,Sichuan 611730)

机构地区：[1]华环国际烟草有限公司,安徽滁州239000 [2]西南交通大学智慧城市与交通学院,四川成都611730

出　　处：《安徽农业科学》2024年第14期219-222,241,共5页Journal of Anhui Agricultural Sciences

摘　　要：为了实现对复烤下机烟叶含水率的准确预测,提出了基于主成分分析法和BP神经网络的烟叶含水率预测模型。首先,采用主成分分析法提取最具表征意义的复烤烟叶含水率特征因子,获得特征矩阵。然后将特征矩阵输入BP神经网络,构建包括特征矩阵与复烤下机烟叶含水率的预测模型。仿真结果表明,提出的模型在复烤烟叶含水率预测方面呈现出显著的预测能力,决定系数达0.92。文中方法可辅助优化烟叶复烤控制参数,提升复烤烟叶品质。In order to realize the accurate prediction of the water content of the tobacco under the re-roasting machine,a tobacco water content prediction model based on principal component analysis and BP neural network was proposed.First,principal component analysis was used to extract the most characteristic factors of water content of re-roasted tobacco,and the feature matrix was obtained.Then,the feature matrix was input into BP neural network to construct a prediction model including the feature matrix and the water content of tobacco under re-roasting.The simulation results showed that the proposed model presented significant prediction ability in the prediction of water content of re-baking tobacco,and the coefficient of determination reached 0.92.By using this method,we could assist in the optimization of the control parameters of tobacco re-baking and the improvement of the quality of re-baking tobacco.

关 键 词：烟叶 含水率 主成分分析 神经网络 预测模型 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TS443[自动化与计算机技术—控制科学与工程]