Digital twin construction method for UAV wings based on model order reduction

This study addresses the pressing requirement for real-time condition monitoring of unmanned aerial vehicle (UAV) wings, which serve as primary load-bearing components and necessitate precise dynamic performance evaluation to ensure flight safety. Although digital twin technology provides a promising framework for wing performance analysis, conventional finite element models impose substantial computational burdens that restrict real-time applicability. To mitigate this limitation, a hybrid modeling approach is developed by integrating Proper Orthogonal Decomposition with surrogate modeling techniques, wherein the POD-radial basis function methodology is identified as the most effective. Through extensive case studies evaluating the multi-condition performance responses of a representative UAV wing, a digital twin visualization platform is constructed. The findings indicate that the proposed method sustains high stress prediction accuracy while markedly enhancing computational efficiency in comparison to traditional finite element models. This advancement facilitates real-time structural performance assessment during UAV operation and presents a feasible solution for intelligent monitoring of complex aerospace structures, underscoring its strong potential for practical deployment in safety-critical aerospace systems.