论文部分内容阅读
针对航空结构可靠性优化设计问题,提出了一种基于灵敏度的可靠性优化(RBO)解耦方法。首先将高效求解的可靠性灵敏度用于失效概率函数(FPF)的快速构建,其优点在于仅需要一次可靠性分析即可得到失效概率函数的局部近似,克服了常规求解方法中需要多次可靠性分析的缺点;然后将得出的FPF近似代入RBO模型中,将原问题解耦成确定性优化子问题,可用常规优化方法来求解。另外,采用了序列近似优化策略来保证可靠性优化解的正确性。文中给出了复合材料梁和机翼三盒段结构的优化求解算例来说明本文方法的可行性和正确性。
Aiming at the problem of aeronautical structural reliability optimization design, a sensitivity-based reliability optimization (RBO) decoupling method is proposed. First of all, the reliability of efficient solution is applied to the fast construction of the failure probability function (FPF). The advantage of this method is that only one reliability analysis is needed to obtain the local approximation of the failure probability function, which overcomes the need of multiple reliability Then the FPF approximation is substituted into the RBO model to decouple the original problem into a deterministic one, which can be solved by the conventional optimization method. In addition, a sequence approximate optimization strategy is adopted to ensure the correctness of the reliability optimization solution. In this paper, an example of the optimal solution of composite beam and wing box structure is given to illustrate the feasibility and correctness of this method.