Improving bi-level system reliability of multi-disciplinary design synthesis optimization method

For large-scale uncertainty multidisciplinary systems, we present an optimization (bi-Level system reliability of the multi-disciplinary design synthesis optimization) algorithm, called by us RBMDOSB algorithm, where RB stands for Reliability-Based, MDO stands for Multidisciplinary Design Optimization, S stands for SORA (Sequential Optimization and Reliability Assessment) and B stands for BLISS (Bi-Level Integrated System Synthesis). Sections 1 through 3 of the full paper explain our RBMDOSB algorithm. RBMDOSB decomposes the reliability analysis from MDO and performs the deterministic MDO and reliability analysis sequentially under the SORA framework. In this algorithm, performance measure approach (PMA) based on multidisciplinary model is used for reliability analysis, BLISS method is introduced for deterministic MDO, and the amount of movement related to the coupling variables is added to amend the constraints of the MDO problems. Sections 4 gives a numerical example, whose results, presented in Table 1, verify the feasibility and efficiency of our RBMDOSB algorithm.