论文部分内容阅读
针对采用离散材料优化(DMO)模型优化复合材料纤维分布时角度优化结果收敛率低的问题,将连续化惩罚策略与Heaviside惩罚函数引入传统DMO模型中,提出了一种改进的HPDMO(Heaviside Penalization of Discrete Materi-al Optimization)模型,从而提高结构的收敛率。建立了复合材料单层板基于最小柔顺性设计的优化列式,给出了多种离散材料构成结构的灵敏度信息求解方法。分析比较了DMO模型、连续化惩罚模型和HPDMO模型对最终优化结果收敛率、目标函数、迭代历史的影响。数值算例表明,改进的离散材料惩罚模型不但可以显著地提高优化结果的收敛率,给出清晰的优化构型,而且可以通过较少的优化迭代步数实现这一结果,为纤维增强复合材料的优化设计在工程中的应用提供了新的技术手段。
In order to solve the problem of low convergence rate of optimal angle distribution of the composite fiber with discrete material optimization (DMO) model, the continuous penalty strategy and Heaviside penalty function are introduced into the traditional DMO model. An improved HPDMO (Heaviside Penalization of Discrete Materi-al Optimization model to improve the rate of structural convergence. Based on the optimal formulation of the minimum flexibility design of composite monolayer, a method to solve the sensitivity information of a variety of discrete materials is given. The effects of DMO model, continuous penalty model and HPDMO model on the convergence rate, objective function and iterative history of final optimization results were analyzed and compared. Numerical examples show that the improved discrete material penalty model can not only significantly improve the convergence rate of the optimization results but also give a clear optimization of the configuration, and can achieve this result with fewer optimized iteration steps. For the fiber reinforced composites The optimal design in the engineering application provides a new technical means.