基于枝状超结构的蒸汽管网优化及禁忌搜索策略  

OPTIMIZATION AND TABOO SEARCH OF STEAM NETWORK BASED ON TREE-SHAPED SUPERSTRUCTURE

作　　者：王金阳[1] 崔国民[1] 涂惟民[1] 李瑜[1] 

机构地区：[1]上海理工大学热工程研究所432信箱,上海200093

出　　处：《工程热物理学报》2011年第2期255-257,共3页Journal of Engineering Thermophysics

基　　金：教育部博士点基金(No.200802520007);上海市人才发展基金(No.2009022);国家自然科学基金项目(No.200406011);上海市重点学科建设项目(No.S30503)

摘　　要：本文通过对蒸汽管网枝状超结构的分析,利用惩罚函数的优化技术,建立了一种基于混合整数非线性模型的蒸汽管网优化方法;同时,鉴于整型变量数目庞大所造成的最优解搜索效率下降的问题,提出了一种"禁忌搜索"的方法,提高了最优化算法的效率。After analyzing the superstructure of tree-shaped steam network, using the optimized technology of penalty function, established a new method for optimal design of steam network, which based on the model of mixed integer nonlinear; at the same time, in view of enormous number of integer variables and complicated combinations of optimization variables, resulting in the problem of declining efficiency in search the optimum solution, proposed a ＂taboo search＂ method to improve the efficiency of optimization algorithm.

关 键 词：蒸汽管网 枝状超结构 优化 惩罚函数 禁忌搜索 

分 类 号：TK172[动力工程及工程热物理—热能工程]