基于多岛遗传算法的连续体结构拓扑优化  

Topology Optimization of a Continuum Structure Using Multi-Island Genetic Algorithm (MIGA)

作　　者：牟淑志[1] 杜春江[2] 牟福元[1] 张玉[1] 

机构地区：[1]金陵科技学院机电工程学院,南京211169 [2]中国电子科技集团公司第14研究所,南京210013

出　　处：《机械科学与技术》2009年第10期1316-1320,共5页Mechanical Science and Technology for Aerospace Engineering

基　　金：金陵科技学院科研基金项目资助

摘　　要：为了能够更好的在优化域中寻找全局最优解,将多岛遗传算法(MIGA)应用到连续体结构拓扑优化中,以二进制编码映射离散化后优化域中的单元,采用有限单元法进行适应度的求解,并将结构中连通域的个数作为约束条件加入到适应度计算中,对典型算例进行了优化设计。In order to find the global optimal solution, continuum structure topology optimization is carried out by using multi-island genetic algorithm （MIGA） with binary code mapping elements in optimization domain, solving fitness with finite element method, and adding numbers of connection as constraint in fitness calculation. Simulation is done with typical samples. Result shows that our method is effective; it can achieve better result with less time consumption compared with the traditional genetic algorithm under the same conditions.

关 键 词：拓扑优化 遗传算法 多岛遗传算法 

分 类 号：TH11[机械工程—机械设计及理论]