| 聚酯工业生产缩聚过程的多目标优化 |
| |
| 引用本文： | 江沛,曹柳林.聚酯工业生产缩聚过程的多目标优化[J].计算机与应用化学,2007,24(11):1493-1496. |
| |
| 作者姓名： | 江沛 曹柳林 |
| |
| 作者单位： | 北京化工大学信息科学与技术学院自动化系,北京,100029 |
| |
| 基金项目： | 国家自然基金(60704011) |
| |
| 摘 要： | 介绍一种聚酯(PET)生产缩聚过程的多目标优化技术,以最终产品的产量最大和质量最佳为优化目标,以第三和第四缩聚釜为优化对象,运用复合神经网络技术及相关机理知识建立优化模型。借鉴非劣分层方法、精英策略和群智思想,建立混合优化算法,通过该算法及惩罚函数寻优,设定生产过程中操作变量的最优参数,采用实际工业生产数据仿真和验证,建立周期性的两级优化结构来实现优化控制。
|
| 关 键 词： | 多目标优化 复合神经网络 非劣分层 粒子群 聚酯 |
| 文章编号： | 1001-4160（2007）11-1493-1496 |
| 收稿时间： | 2007-08-20 |
| 修稿时间： | 2007-10-20 |
Multi-objective optimal operation in polyester polymerization |
| |
| Jiang Pei,Cao Liulin.Multi-objective optimal operation in polyester polymerization[J].Computers and Applied Chemistry,2007,24(11):1493-1496. |
| |
| Authors: | Jiang Pei Cao Liulin |
| |
| Abstract: | A hybrid algorithm was developed for multi-objective optimization in chemical industry.It contained non-dominated sorting method,elitist strategy and swarm intelligence idea.The objectives in polyester polymerization were maximum yield and best quality: desired Intrinsic Viscosity and Molecular Weight Distribution.Models for optimization were founded by multi-neural networks.The hy- brid algorithm has quick convergent rate and better diversity of solutions.By using this algorithm and industrial data,we get best solu- tions in polyester polymerization.We also put this optimization into practice by two-order optimal operation. |
| |
| Keywords: | multi-objective optimization hybrid neural networks non-dominated sorting PSO polyester |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
