化工学报 ›› 2022, Vol. 73 ›› Issue (5): 2060-2072.DOI: 10.11949/0438-1157.20211681

• 过程系统工程 • 上一篇    下一篇

一种应用于换热网络综合的阻尼优化方法

刘薇薇(),崔国民(),张璐,肖媛,杨其国,张冠华   

  1. 上海理工大学能源与动力工程学院,上海市动力工程多相流动与传热重点实验室,上海 200093
  • 收稿日期:2021-11-24 修回日期:2022-01-18 出版日期:2022-05-05 发布日期:2022-05-24
  • 通讯作者: 崔国民
  • 作者简介:刘薇薇(1997—),女,硕士研究生,lww_1126@163.com
  • 基金资助:
    国家自然科学基金项目(21978171);中国博士后科学基金项目(2020M671171)

Damping optimization method for heat exchange network synthesis

Weiwei LIU(),Guomin CUI(),Lu ZHANG,Yuan XIAO,Qiguo YANG,Guanhua ZHANG   

  1. School of Energy and Power Engineering, University of Shanghai for Science and Technology,Shanghai Key Laboratory of Multiphase Flow and Heat Transfer in Power Engineering, Shanghai 200093, China
  • Received:2021-11-24 Revised:2022-01-18 Online:2022-05-05 Published:2022-05-24
  • Contact: Guomin CUI

摘要:

针对启发式方法优化换热网络在优化后期易陷入局部极值的问题,提出一种阻尼优化方法,即通过引入延缓概率的概念,以一定的概率不接受费用下降的结构,延缓该结构形成固定匹配,避免因连续变量优化过快导致整型变量优化不充分而陷入局部最优。通过探讨不同阶段的优化特点及优化陷入局部极值的成因,进而提出分阶段延缓策略,合理调控延缓条件以及延缓概率的取值,从而提高算法的全局搜索能力。最后采用四个不同规模的算例进行验证,结果表明该方法可有效地跳出局部最优解,促进结构的进一步优化。

关键词: 换热网络, 优化, 局部极值, 延缓, 结构进化, 模型, 计算机模拟

Abstract:

Aiming at the problem that the heat exchange network optimized by the heuristic method is easy to fall into the local extreme value in the later stage of optimization, a damping optimization method is proposed, that is, by introducing the concept of delay probability, the structure with a certain probability will not accept the cost reduction, the structure will be delayed from forming a fixed structure match, and will avoid falling into local optimum due to insufficient optimization of integer variables due to too fast optimization of continuous variables. By discussing the optimization characteristics of different stages and the causes of optimization falling into local extreme, a phased delay strategy is proposed to reasonably adjust the delay conditions and the value of delay probability, to improve the global search ability of the algorithm. Finally, four different scale examples are used to verify the results. The results show that the method can effectively jump out of the local optimal solution and promote the further optimization of the structure.

Key words: heat exchanger network, optimization, local extreme, delay, structural variation, model, computer simulation

中图分类号: