应用双层混合遗传算法优化大规模换热网络  

Application of Double-level Hybrid Genetic Algorithm to Optimize Large-scale Heat Exchanger Networks

作　　者：毛政涛 刘新文[1] 

机构地区：[1]宁波工程学院,浙江宁波315016

出　　处：《宁波工程学院学报》2014年第1期77-82,共6页Journal of Ningbo University of Technology

基　　金：宁波市教育科学规划课题成果(项目编号:YGH032);"王伟明助创基金"项目(项目编号:2013006)

摘　　要：换热网络综合优化问题是典型的混合整数非线性规划(MINLP)问题,存在局部极值众多的特点,特别是对于大规模换热网络来说,要做到全局寻优,难度极大。混合遗传算法能够对换热网络进行优化,但对大规模的换热网络的优化效率并不高。本文采用先通过混合遗传算法对换热网络进行初步优化,然后划分换热网络为若干换热独立的子网络,对产生的子网络分别进行单性遗传进化。结果表明,该方法对大规模换热网络的优化具有较高的优化效率。The optimization of heat exchanger networks （HEN） is a typical mixed integer nonlinear program （MINLP） problem with numerous local peaks. Especially for large-scale HEN, it is difficult to solve this problem globally. Hybrid genetic algorithm could optimize HEN, but the efficiency of optimizing the large-scale HEN is not high. In this paper, the hybrid genetic algorithm is applied to optimize the whole HEN, then it is divided into several independent sub- networks and the monogenetic algorithm is applied to optimize these sub-networks. The resuks show that this method could optimize the large-scale HENs efficiently.

关 键 词：大规模 换热网络 双层遗传算法 单性遗传 优化 

分 类 号：TK513[动力工程及工程热物理—热能工程]