带有区域性惩罚因子的双向渐进结构优化方法

Bidirectional Evolutionary Structural Optimization with Regional Penalty Factor

针对传统的拓扑优化方法欠缺考虑增材制造工艺约束,如果优化模型存在大面积悬垂、封闭空腔等工艺几何特征将增加打印及后处理操作的复杂度的问题,提出带有区域性惩罚因子的双向渐进拓扑优化方法.首先考虑增材制造工艺特征约束,使得优化结果具有良好的可打印性;然后对模型进行网格划分,根据单元体所在的区域定义区域性惩罚因子,保证区域性惩罚因子由外表面向内核逐渐增大;最后结合有限元分析,以区域性惩罚因子对单元体的计算应力进行惩罚,决策单元体的保留与删减.通过数值实验,与传统的双向渐进结构拓扑优化算法对比的结果表明,该方法可以改变空腔形貌特征、减少垂悬面积、均衡模型应力分布,提升了拓扑优化模型的可打印效果.

Traditional structural optimization methods do not take full account of the process constraints from additive manufacturing.Manufacturing process and post-processing will become more difficult if the optimized model has process geometric characteristics such as large overhangs and closed cavities.This paper presents bidirectional evolutionary structural optimization method with regional penalty factor.It makes optimized results printable by considering process feature constraints.Regional penalty factors were defined according to the region where elements belonged to.Factors should gradually increase from outer toward inner core while meshing.Stress of the element would be penalized with the factor to decide its retention or deletion after finite element analysis.Compared with traditional bidirectional evolutionary structural optimization method by numeric experiments,the method can change the morphology of cavity,reduce overhang area,balance the stress distribution of the model,and improve the printable effect of the topology optimization model.