化工进展2017，Vol.36Issue(2)：442-450,9.DOI:10.16085/j.issn.1000-6613.2017.02.006
具有步长调整策略的强制进化随机游走算法优化换热网络
Optimizing heat exchanger network by random walking algorithm with compulsive evolution combined with step length adjustment strategy
摘要
Abstract
Random walking algorithm with compulsive evolution(RWCE) is a novel heuristic method to optimize heat exchanger networks,which has a powerful global optimizing ability in the process of evolution. In this paper,the effect of maximal step length on the performance of RWCE algorithm was studied. To efficiently control the global and local search ability of the algorithm,a decreasing maximal step length adjustment strategy based on a parabola opening downwards curve was proposed. Compared with the basic algorithm,the strategy is capable of jumping out of local optima in the late evolution stage and strengthening the local search ability. The optimal results of three HEN cases (10SP2,9SP and 15SP) from literatures were used to test the effectiveness of the RWCE algorithm cooperated with proposed strategy. The results of former two(10SP2 and 9SP)are better than the best results published,which is 20.98% and 1.11% lower than the original literature results. A new heat exchanger networks structure was found in case 3(15SP),which is better than the majority of optimal results of no stream splits and 4.6% lower than the literature results. The results of these three cases demonstrate that the method enjoys a better optimization capability in the global optimization of heat exchanger network.关键词
强制进化随机游走算法/换热网络优化/全局搜索能力/局部搜索能力Key words
random walking algorithm with compulsive evolution(RWCE)/heat exchanger network synthesis(HENS)/global searching ability/local search capability分类
能源科技引用本文复制引用
刘璞,崔国民,肖媛,陈家星,周剑卫..具有步长调整策略的强制进化随机游走算法优化换热网络[J].化工进展,2017,36(2):442-450,9.基金项目
上海市科委部分地方院校能力建设计划（16060502600）、国家自然科学基金（51176125）及沪江基金研究基地专项（D14001）项目。 （）
