Reliability based aeroelastic design optimization of composite wings via surrogate modeling

This paper presents a framework based on surrogate modeling for the reliability-focused optimization of aeroelastic performance in composite wing structures. The optimization process centers on modifying the stacking sequences of composite materials to enhance aeroelastic performance, while explicitly accounting for uncertainties in ply angles resulting from manufac turing variability. To address the high computational cost of direct aeroelastic evaluations, the framework employs a modified high-dimensional Kriging surrogate model combined with an adapted Efficient Global Optimization (EGO) algorithm. This approach enables an efficient ex ploration of the design space, balancing accuracy with computational feasibility. By integrating reliability analysis within the optimization process, the framework also ensures robust aeroelas tic performance under uncertain conditions. Results demonstrate that the proposed method identifies optimal stacking sequences with improved aeroelastic reliability while significantly reducing computational costs. This methodology offers a novel approach to enhancing the reliability of composite aeroelastic structures, with implications for high-performance aerospace applications.