蜂窝夹芯板贯穿孔损伤修理容限上限确定方法

A method for determining the upper tolerance limit of perforation damage repair of honeycomb sandwich panels

为确定蜂窝夹芯板贯穿孔损伤能修与不能修理的理论界限(即修理容限上限),提出了一种基于参数化建模与遗传算法搜索的修理容限上限确定方法。首先采用ABAQUS有限元软件创建蜂窝夹芯板贯穿孔损伤修理分析模型,其中使用Hashin失效准则分析复合材料面板的损伤起始与扩展,并在修理铺层与母板间设置了cohesive内聚力单元,以预测修理结构的强度。同时开展了无损伤、含穿透损伤、修理板的对比压缩试验,验证有限元模拟结果的准确性。在此基础上,采用Python编程语言对结构修理模型进行参数化建模,并与基于NSGA-II遗传算法的modeFRONTIER优化软件进行集成,快速搜索出优化修理方案,再采用二分法最终确定蜂窝夹芯板贯穿孔损伤的修理容限上限。结果表明:蜂窝夹芯板修理分析模型的预测结果与试验值吻合较好,证明了建模方法的准确性;参数化建模结合遗传算法的方法能够快速确定蜂窝夹芯板贯穿孔损伤的修理容限上限,可大幅度减少计算工作量。

In order to determine the theoretical repairable threshold(i.e.the upper limit of repair tolerance)of honeycomb sandwich panels suffered with perforation damage,a method based on parametric modeling and genetic algorithm search is proposed.First,a finite element(FE)model simulating perforation damage repair of honeycomb sandwich panels was built using the ABAQUS software,in which the Hashin failure criterion was used for analyzing damage initiation and propagation of composite panels,and built-in cohesive elements were set between repair plies and the parent panel for predicting the strength of repaired structures.Meanwhile,compression tests on intact,perforation damaged,and repaired honeycomb sandwich panels were comparatively performed to verify the accuracy of the FE simulation results.Based on these results,the FE repair model was parameterized by the Python programming language,and then integrated with the modeFRONTIER optimization software based on NSGA-II genetic algorithm,which yields optimized repair scheme after quick search.The repairable threshold of honeycomb sandwich panel with perforation damage was finally obtained using the method of bisection.The results show that the simulation results of the repair model are in good agreement with those of the tests,which proves the accuracy of the FE modeling method.Parametric modeling combined with genetic algorithm can quickly determine the theoretical repairable threshold of honeycomb sandwich panels with perforation damage by greatly reducing calculation work.