Reliability-based multidisciplinary design optimization of an aeroelastic unpowered guided aerial vehicle

Most Aeronautical and Astronautical Systems (AAS) are inherently complex, multidisciplinary, nonlinear, and computationally intensive for design and analysis. Utilizing the Reliability-Based Multidisciplinary Design Optimization framework can address the multidisciplinary nature of these systems while accounting for inherent uncertainties. In this paper, an efficient methodology for Reliability-Based Multidisciplinary Design optimization of an aerial vehicle is developed. The computational burden of reliability assessment could make its integration within a Multidisciplinary Design Optimization cycle a formidable task. In this respect, a multilevel Multidisciplinary Design Optimization architecture is proposed in which the computational cost is reduced by considering the reliability analysis, as needed only for critical subsystems. To this end, a single-level Reliability-Based Multidisciplinary Design Optimization is derived using the Performance Measure Analysis and the Karush-Kuhn-Tucker condition. The work demonstrates the integration of this formulation into the proposed multilevel Reliability-Based Multidisciplinary Design Optimization architecture. The proposed design architecture is implemented for an aeroelastic Unpowered Guided Aerial Vehicle whose outcomes are compared with previous results obtained via a mono-level Uncertainty-Based Multidisciplinary Design Optimization architecture.