This study presents a data-driven finite element-machine learning surrogate model for predicting the end-to-end full-field stress distribution and stress concentration around an arbitrary-shaped inclusion. This is important because the model’s capacity to handle large datasets, consider variations in size and shape, and accurately replicate stress fields makes it a valuable tool for studying how inclusion characteristics affect material performance. An automatized dataset generation method using finite element simulation is proposed, validated, and used for attaining a dataset with one thousand inclusion shapes motivated by experimental observations and their corresponding spatially-varying stress distributions. A U-Net-based convolutional neural network (CNN) is trained using the dataset, and its performance is evaluated through quantitative and qualitative comparisons. The dataset, consisting of these stress data arrays, is directly fed into the CNN model for training and evaluation. This approach bypasses the need for converting the stress data into image format, allowing for a more direct and efficient input representation for the CNN. The model was evaluated through a series of sensitivity analyses, focusing on the impact of dataset size and model resolution on accuracy and performance. The results demonstrated that increasing the dataset size significantly improved the model’s prediction accuracy, as indicated by the correlation values. Additionally, the investigation into the effect of model resolution revealed that higher resolutions led to better stress field predictions and reduced error. Overall, the surrogate model proved effective in accurately predicting the effective stress concentration in inclusions, showcasing its potential in practical applications requiring stress analysis such as structural engineering, material design, failure analysis, and multi-scale modeling.

This paper presents an experimental and numerical investigation on the influence of pre-existing impact damage on the low-velocity impact response of Carbon Fiber Reinforced Polymer (CFRP). A continuum damage mechanics-based material model was developed by defining a user-defined material model in Abaqus/Explicit. The model employed the action plane strength of Puck for the damage initiation criterion together with a strain-based progressive damage model. Initial finite element simulations at the single-element level demonstrated the validity and capability of the damage model. More complex models were used to simulate tensile specimens, coupon specimens, and skin panels subjected to low-velocity impacts, being validated against experimental data at each stage. The effect of non-central impact location showed higher impact peak forces and bigger damage areas for impacts closer to panel boundaries. The presence of pre-existing damage close to the impact region leading to interfering delamination areas produced severe changes in the mechanical response, lowering the impact resistance on the panel for the second impact, while for non-interfering impacts, the results of the second impact were similar to the impact of a pristine specimen.