华东交通大学学报2017，Vol.34Issue(3)：125-130,6.
基于多目标模拟退火算法的微管网拓扑优化
Topology Optimization of Micro-vascular Network Based on Multi-objective Simulated Annealing Algorithms
摘要
Abstract
In the self-healing research, the microcapsule(or the hollow glass giber) is served as the carrier of self-healing agent. However, this carrier has some deficiencies such as being independent and uncorrelated with each other. In this paper, a novel carrier of micro-vascular network based on multi-objective simulated anneal-ing (MOSA) algorithms was studied and topology of micro-vascular network was optimized to design a carrier with low volume fraction, high flow efficiency and inter-connective structure. Firstly, the mathematical model was established. Secondly, the multi-objective optimization function was designed by iteration of Hardy Cross. And then the topology of micro-vascular by MOSA algorithms combined with non-dominated sorting was studied. Fi-nally, the superiority of MOSA was verified compared with non-dominated sort genetic algorithm (NSGA-Ⅱ) al-gorithms. The results show that volume fraction and flow efficiency of micro-vascular network are negatively cor-related. The convergence time of NSGA-Ⅱ is slightly better than NSGA-Ⅱ, but the degree of dominant and uni-formity of MOSA are significantly better than that of NSGA-Ⅱ.关键词
微管网/多目标优化/模拟退火/非劣分层/哈迪克罗斯迭代Key words
micro-vascular network/multi-objective optimization/simulated annealing/non-dominated sorting/iteration of Hardy Cross分类
通用工业技术引用本文复制引用
李鹏,刘远,黄金勇..基于多目标模拟退火算法的微管网拓扑优化[J].华东交通大学学报,2017,34(3):125-130,6.基金项目
国家自然科学基金项目(51365012) （51365012）
研究生创新专项资金项目(YC2015-S237) （YC2015-S237）
