基于无单元伽辽金法和Matlab遗传算法工具箱的结构形状优化研究  

Structural Shape Optimization Using Element-free Galerkin Method and Genetic Algorithm Toolbox

作　　者：陈仁科[1] 龚曙光[1] 黄云清[2] 张建平[1] 

机构地区：[1]湘潭大学机械工程学院CAE研究所,湘潭411105 [2]湘潭大学科学工程计算与数值仿真湖南省重点实验室,湘潭411105

出　　处：《机械科学与技术》2008年第6期827-830,833,共5页Mechanical Science and Technology for Aerospace Engineering

基　　金：国家自然科学基金项目(50475143);高等学校博士学科点专项科研基金项目(20040530001);国家重点基础研究专项经费(2005CB321701);湖南省教育厅优秀青年项目(06B096);湘潭大学博士启动基金项目(06QDZ16)资助

摘　　要：将无网格Galerkin法(element-free Galerkin method,EFGM)和改进的Matlab遗传算法工具箱(matlab genetic toolbox,MGT)相结合,提出了一种新的连续体结构形状优化设计方法。针对形状优化设计的特点,建立了一种规则网格与有限元网格混合的EFG积分背景网格法。详细介绍了M函数文件的修改方法,解决了现有Matlab遗传算法工具箱不能直接施加的非线性与隐性约束条件的难题。最后,完成了一个工程实例的结构形状优化,优化结果表明该方法是有效可行的。A new method is proposed for shape optimization of continuum structure by integrating Matlab GAT（ Genetic Algorithm Toolbox） with EFGM（Element-Free Galerkin Method）. Taking into account the characteristic of shape optimal design, a new background mesh for integration is presented through combining regular mesh with fi- nite element mesh. The revision method of M file is introduced in detail, and the difficulty that the non-linear, implicit constraint cannot be imposed directly in using the current genetic algorithm toolbox in Matlab is successfully overcome. Finally, the structural shape optimization of an engineering example is performed, and it shows that the present method is effective and feasible.

关 键 词：形状优化 无网格法 无单元伽辽金法 MATLAB遗传算法工具箱 背景网格 

分 类 号：TH122[机械工程—机械设计及理论]