| 基于组合优化的卷烟叶组配方设计 |
| |
| 引用本文： | 郭科,薛源,胡丹,白林.基于组合优化的卷烟叶组配方设计[J].中国烟草学报,2007,13(2):21-23,32. |
| |
| 作者姓名： | 郭科 薛源 胡丹 白林 |
| |
| 作者单位： | 成都理工大学信息管理学院,成都市成华区东三路1号,610059 |
| |
| 摘 要： | 应用BP神经网络建立了叶组配方的主要化学成分含量与其感官质量和烟气化学成分含量之间的非线性映射关系,在此基础上建立了相关的约束最优化模型并求解,由此得到所选取烟叶的最佳配方比例。使用此组合优化方法克服了以往在设计卷烟叶组配方中完全以人的主观判断为标准所带来的相当的盲目性和主观性,从而实现了卷烟叶组配方经验设计与计算机智能设计相结合的组合交互式设计,使设计出的叶组配方更具科学性和合理性。
|
| 关 键 词： | BP神经网络 约束最优化 卷烟叶组配方 设计 |
| 文章编号： | 1004-5708（2007）02-0021-03 |
| 修稿时间： | 2006-09-05 |
Designing cigarette blending formulation based on combined optimization |
| |
| GUO Ke,XUE Yuan,HU Dan,BAI Lin.Designing cigarette blending formulation based on combined optimization[J].Acta Tabacaria Sinica,2007,13(2):21-23,32. |
| |
| Authors: | GUO Ke XUE Yuan HU Dan BAI Lin |
| |
| Affiliation: | College of Information Management, Chengdu University of Technology, Chengdu 610059, China |
| |
| Abstract: | Non-linear relationship between primary chemical composition of blended tobacco and sensory quality as well as smoke components was built by applying BP network. Based on that a constrained optimization model of blending formulation to determine the best proportion of leaf tobacco was obtained. This method can solve the problem of blindness and human influence in cigarette design as far as blending formulation is concerned therefore can help to improve cigarette design by combining computer technology and experience. |
| |
| Keywords: | BP network constrained optimization cigarette blending formulation design |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
