In recent years, several studies have demonstrated that aeronautical structures can still function after buckling. By harnessing buckling, it is possible to develop airfoils that are lighter, more efficient and exhibit enhanced functionality. Some preliminary studies have investigated the design of shape-adaptive airfoils operating in the post-buckling regime [1, 2]. However, accurately and efficiently predicting the deformation of structures in the post-buckling state is challenging due to the nonlinear nature of buckling phenomena. To address this challenge, we propose a Physics-Informed Neural Network (PINN)-based model for designing shape-adaptive airfoils under buckling loads. The PINN model acts as a surrogate for the finite element model. Given the geometric features and the query point locations as inputs, the PINN model predicts displacements. Strain energy is integrated as a physical loss term to enhance the reliability of the PINN model and automatic differentiation is utilized to efficiently compute the strains. Moreover, boundary conditions and geometry continuous constraints are added in the loss terms to improve the performance of displacement prediction. To train the PINN model, a configuration generating algorithm is developed, and the training dataset is built using the finite element method. The proposed PINN provides fast and accurate predictions of the deformation of the structures in the post-buckling regime, thereby accelerating the design of shape-adaptive airfoils. REFERENCES [1] J. Zhang and C. Bisagni. Buckling-driven mechanisms for twisting control in adaptive composite wings. Aerospace Science and Technology, Vol. 118, p. 107006, 2021. [2] D. Hahn , M. Haupt, S. Heimbs. Passive load alleviation by nonlinear stiffness of airfoil structures. AIAA SCITECH 2022 Forum, San Diego, CA & Virtual: American Institute of Aeronautics and Astronautics, p. 0318, 2022.