A synergistic physics-data multiscale assembly deformation and damage analysis method for large-size CFRP structures

Large-size Carbon Fiber Reinforced Polymer (CFRP) structures are critical in aircraft, yet assembly-induced deformation and damage significantly impact service performance. Therefore, predicting assembly deformation and damage is crucial for ensuring CFRP structural performance. However, due to the large disparity of macro–micro scales and complex geometries, the efficient and accurate prediction of multiscale assembly deformation and damage for large-size CFRP structures remains a challenge. To address this issue, this paper proposed a synergistic physics-data multiscale analysis method for assembly deformation and damage of large-size CFRP structures. The method employed the hierarchical structural computation strategy, in which the structure is divided into critical and non-critical regions based on pre-simulation results. During the analysis, critical regions were solved using the concurrent multiscale method (physics driven), while non-critical regions were computed using machine learning surrogate model (data driven). This synergistic approach enabled efficient and accurate prediction of the multiscale mechanical behavior of large-size CFRP structure. The validity of the multiscale model was verified through specifically designed experiments on CFRP laminate and panel. The results demonstrated that the proposed method successfully captures the multiscale mechanical responses and damage states, providing an efficient and accurate tool for the deformation and damage analysis of large-size CFRP structures.