Abstract: Geometric deformations and performance drifts resulting from long-term service or modifications present significant challenges to the lifecycle assessment and maintenance of aircraft. To address this, we propose an integrated assessment framework based on measured point clouds that synthesizes reverse modeling, free-form deformation (FFD), and aerodynamic simulation. In the reverse modeling phase, we develop a robust point cloud reconstruction workflow incorporating normal vector information. By combining Screened Poisson Reconstruction with aerodynamic topological logic, unstructured point clouds are converted into watertight meshes that satisfy the boundary conditions required for computational fluid dynamics. To mitigate the sensitivity of the Poisson algorithm to normal noise, a feature-preserving smoothing strategy is introduced to jointly optimize point positions and normal vector fields. This approach effectively suppresses high-frequency noise while preserving the geometric continuity of critical aerodynamic features, such as the wing trailing edge. Subsequently, in the deformation phase, control volumes are constructed using FFD technology without the need for original parameterized geometric models, enabling the parametric manipulation of key aerodynamic shapes including wing twist and dihedral. Finally, utilizing the meshless Lattice Boltzmann Method (LBM), the reconstructed geometry is directly employed for multi-condition flow simulations to analyze the response patterns of aerodynamic parameters and flow structures before and after deformation. Results demonstrate that this framework reconstructs high-fidelity watertight geometric models from high-precision scanned data and facilitates parameterized geometric deformation, offering a robust engineering pathway for digital twin construction and multidisciplinary performance assessment of complex aircraft configurations.

Key words: Reverse engineering, Point cloud reconstruction, Free-Form Deformation(FFD), Aerodynamic characteristics, Lattice Boltzmann Method(LBM)