基于改进双向渐进结构优化法的钢节点拓扑优化研究OA北大核心CSTPCD
Research on topology optimization of steel joints based on improved bi-directional progressive structural optimization method
为提升双向渐进结构优化法(BESO)优化钢结构节点时的计算效率,充分解决优化过程中存在的棋盘格和网格依赖现象,提出了一种改进双向渐进结构优化法SJ-BESO.该方法引入灵敏度滤波半径,对过滤半径范围内各个单元的灵敏度值进行加权处理,并通过设置灵敏度阈值段确定元素的保留、删除或添加,同时引入收敛参数确保结构计算趋优的稳定性.通过钢牛腿节点算例将SJ-BESO与BESO进行对比分析,发现SJ-BESO有效提升了计算迭代效率并克服了棋盘格和网格依赖现象.再通过分叉节点算例将SJ-BESO与变密度法(SIMP)比较,表明SJ-BESO的优化方案材料利用效率更高.SJ-BESO适用于钢节点的拓扑优化,具有良好的计算效率和优化效果.
In order to improve the computational efficiency of bi-directional progressive structural optimization(BESO)of steel structural joints,and fully solve the problem of checkerboard and grid dependence in the optimization process,an improved bi-directional progressive structural optimization(SJ-BESO)method is proposed.In this method,the sensitivity filtering radius is introduced,and after the sensitivity values of each element within the filtering radius are weighted,the element retention,deletion or addition is determined by setting the sensitivity threshold segment.Meanwhile,the convergence parameter is established to ensure the stability of the structural gradual optimization calculation.A comparative analysis between SJ-BESO and BESO was conducted using a corbel joint example.The results indicate that SJ-BESO significantly enhances computational iteration efficiency and overcomes issues related to checkerboard patterns and grid dependency.Then,an example with a bifurcated node is used to compare SJ-BESO with the variable density method(SIMP),and the results show that the optimized scheme of SJ-BESO has a higher material utilization efficiency.SJ-BESO is suitable for topology optimization of steel joints,and has good computational efficiency and optimization performance.
汤朋山;杜文风;李少龙;顾金超;高博青
河南大学钢与空间结构研究所,开封 475004河南大学钢与空间结构研究所,开封 475004河南大学钢与空间结构研究所,开封 475004河南大学钢与空间结构研究所,开封 475004浙江大学空间结构研究中心,杭州 310027
双向渐进结构优化法钢节点灵敏度滤波有限元分析增材制造
bi-directional progressive structure optimization methodsteel structural jointsensitivity filteringfinite element analysisadditive manufacturing
《计算力学学报》 2024 (6)
991-997,7
国家自然科学基金(52478166)河南省自然科学基金重点项目(232300421133)资助.
评论