化工学报 ›› 2008, Vol. 59 ›› Issue (4): 947-952.

• 过程系统工程 • 上一篇    下一篇

一种基于经验增强的实时优化方法MEO

张正江;邵之江;陈曦;周舟;钱积新   

  1. 浙江大学工业控制技术国家重点实验室,工业控制研究所,浙江 杭州 310027
  • 出版日期:2008-04-05 发布日期:2008-04-05

Real-time optimization method MEO based on mnemonic enhancement

ZHANG Zhengjiang;SHAO Zhijiang;CHEN Xi;ZHOU Zhou;QIAN Jixin   

  • Online:2008-04-05 Published:2008-04-05

摘要:

根据实时优化问题的特点,提出了一种实时优化方法——MEO(mnemonic enhancement optimization),它是一种高效的基于经验增强的实时优化方法。设计了MEO框架及其具体实现步骤,基于Aspen Plus平台开发了MEO通用代理服务器系统。结合Aspen Plus OOMF 脚本语言与AOS NLP/NLA接口混合编程实施了MEO通用代理服务器系统。并应用脱丙烷塔和脱丁烷塔的联塔系统及大规模乙烯分离系统进行测试,结果显示相比于传统方法,MEO不但具有很好的实时性和收敛性,而且具有很好的鲁棒性和开放性,为实时操作优化的进行奠定了基础。

关键词: 实时优化, MEO, 经验记忆库, 逼近方法, 联塔系统, 乙烯分离系统

Abstract:

A real-time optimization method MEO (mnemonic enhancement optimization) is introduced according to the characteristics of the real time optimization (RTO) problem, it is an efficient method based on mnemonic enhancement.The framework of MEO is designed, and the process of implementation is presented in this paper.The universal proxy of MEO is developed based on Aspen Plus.Aspen Plus OOMF script language and AOS NLP/NLA interface are used.With the simulations of multi-column system and ethylene separation process system, the effectiveness of MEO is demonstrated.It confirms that MEO is not only efficient in solution and convergence, but also robust and flexible.As a result, MEO can be very useful in the real-time optimization.

Key words: 实时优化, MEO, 经验记忆库, 逼近方法, 联塔系统, 乙烯分离系统