研究了基于自适应网格技术的结构拓扑优化.采用有限元离散设计域,单元节点密度作为设计变量.优化迭代过程中,根据设计域密度场信息对结构网格进行自适应加密和稀疏,使得材料分布边界处的网格加密,远离材料边界处的网格稀疏.同时,优化设计变量空间也随着网格的变化而变化.给出了拓扑优化中网格自适应加密和稀疏的准则,以及网格变化时结构密度场更新算法.算例结果表明提出的拓扑优化策略可以减少结构分析和优化求解的计算量,在同等结构分析和优化求解计算量下能够得到更好的拓扑结果.
The structural topology optimization based on adaptive mesh refinement is researched. The design domain is discretized into element, where the node density is the design variable. As the optimization proceeds, the structural mesh gets refined or coarsened depending on the density field information of the design domain in order to refine the mesh near the material boundary and coarsen the mesh away from the material boundary. At the same time, the design space changes with the change of structural mesh. The rule of adaptive mesh refinement and the algorithm of updating density field are proposed. The example result shows that the computation efficiency of structural analysis and optimization can be improved and the better topology result can be obtained with the same computational effort for structural analysis and optimization.