三维对称性破缺内凹拉胀蜂窝结构的弹性性能及逆向设计
ELASTIC PROPERTIES AND INVERSE DESIGN OF THREE-DIMENSIONAL SYMMETRY-BROKEN RE-ENTRANT HONEYCOMB
-
摘要: 拉胀蜂窝结构因具有重量轻、比强度高、阻尼性能好而受到了广泛的研究. 最近研究表明, 对称性破缺内凹蜂窝结构具有更好的吸能性能, 以及更优越的拉胀性. 然而, 传统二维结构的拉胀性能在三维空间应用上存在局限, 难以兼顾结构的轻量化设计及多方向负泊松比, 且目前缺乏对该种结构弹性性能的理论描述. 为此, 本文提出了三维双向正交对称性破缺内凹拉胀蜂窝结构. 基于小变形假设, 通过力法建立了该新型结构的理论模型, 并分析了结构几何参数对弹性性能的影响; 同时为了更好地应用所推导的理论模型, 满足结构实际应用力学性能的需求, 运用非支配排序遗传算法Ⅱ的优化功能, 在一定范围内给定等效弹性模量和泊松比, 通过附加该新型结构几何约束条件的逆向设计方法来获取实现目标性能值的几何参数组合. 相比于依赖尝试与经验的传统设计或计算量较大的人工神经网络方法, 该逆向设计框架能有效避免不满足几何相容性的参数组合产生, 显著提高了设计效率与可靠性. 最后, 通过有限元及实验验证理论模型和逆向设计方法的有效性与准确性.Abstract: Auxetic honeycomb structures have attracted significant research attention due to their lightweight nature, high specific strength, and superior damping properties. Recent studies have shown that the symmetry-broken re-entrant honeycomb exhibits higher energy absorption capabilities and auxetic performance. However, traditional two-dimensional configurations face limitations regarding spatial application in three-dimensional environments, it is difficult to take into account the lightweight design of the structure and the multi-directional negative Poisson's ratio, and a comprehensive description of the elastic properties for such structures is currently lacking. To address these issues, a three-dimensional bidirectional orthogonal symmetry-broken re-entrant auxetic honeycomb is proposed. Based on the small deformation hypothesis, a theoretical model of this novel structure is established using the force method to systematically analyze the influence of geometric parameters on elastic properties. To facilitate the practical application of the derived theoretical model and meet specific mechanical requirements, an inverse design approach based on the Non-dominated Sorting Genetic Algorithm II is implemented. By specifying the target equivalent elastic modulus and Poisson's ratio within a defined range, and incorporating geometric constraints, the optimal combinations of geometric parameters required to achieve the target properties are identified. Compared with the traditional trial and error design or artificial neural network method with large computational cost, the proposed inverse design framework can effectively avoid the generation of parameters that do not meet the geometric compatibility, and significantly improve the design efficiency and reliability. Finally, the validity and accuracy of the proposed theoretical model and the inverse design method are verified through Finite Element Analysis and experimental tests.
下载: