An efficient strategy for reliability-based multidisciplinary design optimization

As a powerful tool, reliability-based multidisciplinary design optimization (RBMDO) has gained increasingly attentions. However, integrating the reliability-based design and multidisciplinary design optimization (MDO) directly may cause tremendous implementation and low computational efficiency. In this work, an efficient strategy features two-layered for RBMDO was proposed. The sequential optimization and reliability assessment (SORA) was adopted in the upper layer to decouple the reliability analysis from deterministic MDO which avoided the nested loops of RBMDO and thus improved the efficiency. In the lower layer, the deterministic MDO was implemented in a parallel way with the concurrent subspace optimization strategy (CSSO). A performance measure approach based sequential multidisciplinary system analysis method (PMA-SMRA) is proposed in this paper, which enables the system analysis and system sensitivity analysis executed sequentially to search the most probable point (MPP). The effectiveness of the proposed strategy is illustrated with the shock absorber of landing gear.