The projectile deflects and even ricochets after an oblique impact on the concrete. However, research on the oblique impact of projectiles on concrete targets mainly focuses on oblique penetration and the critical ricochet angle, and there are few experimental studies on ricochets. Deflection and its influencing factors remain undefined. This study conducted experiments and LS-DYNA numerical simulations on projectiles obliquely impacting C60 concrete targets. The experimental research visually revealed deflection and ricochet phenomena after the oblique impact. The ricochet caused by large-angle impacts can effectively reduce the damaged area of concrete targets. Subsequently, the main governing parameters affecting the deflection angle of the projectile were identified through dimensional analysis, and a sensitivity analysis was performed on these parameters using an orthogonal experimental design. On this basis, the influence of the incident angle, impact velocity, and projectile length-to-diameter ratio on the projectile deflection was further clarified. The results showed that the maximum deflection angle was achieved when a 30 mm caliber projectile obliquely impacted a C60 concrete at an incident angle of ∼45°. In the case of ricochets, the deflection angle increased with an increase in the impact velocity and decreased with an increase in the length-to-diameter ratio. This study aids in predicting and controlling projectile deflection and provides a reference for the innovative design of concrete protective structures.

A quasi-static finite cavity cylindrical expansion model is proposed to investigate the lateral boundary effect during long-rod penetration. Analytical solutions for the cavity pressure in the elastic–plastic stage and plastic stage are obtained, through which a decay function for the lateral boundary effect is constructed. The resistance of the target with finite diameter R_T is obtained by multiplying the decay function with the resistance of infinite target R_T*. Then, a theoretical model for the penetration into the metal targets with finite diameters by long rods is proposed by substituting R_T into the Alekseevskii–Tate model. The proposed model is verified by comparison with existing experimental results. Furthermore, the effect of the target diameter, initial impact velocity and yield criterion on the penetration depth and resistance are investigated. The results showed that the lateral boundary effect should be taken into account if the ratio of the target radius to the rod radius (r_t0/R_P) is less than 25.

With the development of various emerging structures, concrete-filled steel tubular (CFST) columns have become critical load-bearing components in key infrastructures such as subways and underground utility tunnels. Accurately predicting their ultimate bearing capacity (N_u) is essential for guaranteeing structural safety. To address the limitations of traditional empirical formulas and code-based calculation approaches, this paper proposes a prediction model for ultimate bearing capacity based on the CatBoost algorithm optimized by Random Search. Furthermore, the marginal contribution of each key feature to the prediction results is measured through interpretability analysis. First, a database containing 438 CFST column ultimate bearing capacity test cases was established, with key parameters such as geometric dimensions and material properties as input variables. Second, the predictive performance of six machine learning algorithms—CatBoost, LightGBM, Random Forest (RF), Gradient Boosting (GB), K-Nearest Neighbors (KNN), and XGBoost—was compared. A five-fold cross-validation integrated with a Random Search strategy was employed for joint hyperparameter optimization. The results show that the optimized CatBoost model significantly outperforms other algorithms and conventional design codes, achieving a coefficient of determination (R²) as high as 0.99 and a root mean square error (RMSE) of 174.29 kN. Furthermore, the SHAP (Shapley Additive exPlanations) method was used to perform global and local interpretability analyses of the prediction model. This not only quantified the individual contribution and interaction effects of each feature parameter on the bearing capacity but also revealed that geometric parameters are the primary influencing factor. This finding confirms a high degree of consistency between the prediction mechanism of the data-driven model and classical mechanical theories, effectively validating the model’s reliability. This study provides an efficient and reliable tool for the optimal design and rapid evaluation of CFST columns and establishes a new data-driven paradigm for the design and reinforcement of key components in underground structures.