Prediction Model for Projectile Ballistic Characteristics in Multi-Layered Spaced Concrete Thin Targets Based on CNN

To address the high computational cost of traditional ballistic prediction methods and their difficulty in meeting rapid assessment needs, this paper proposes an efficient predictive model based on a Convolutional Neural Network (CNN) for the penetration ballistics of multi-layer thin concrete targets. First, using a numerically simulated method validated through experiments, the significant influence of projectile angular velocity on trajectory deflection was analyzed and confirmed. This parameter was subsequently identified as a key projectile-target engagement condition. By systematically adjusting initial parameters, a dataset containing 127 cases of single-layer thin concrete target penetration was constructed. Building on this, a high-precision ballistic prediction model for single-layer target penetration was developed, with projectile parameters, target parameters, and engagement conditions as inputs, and post-impact projectile motion parameters as outputs. Furthermore, by integrating rigid-body kinematic equations describing the projectile's flight between targets, a complete penetration-flight iterative prediction framework was established, enabling rapid prediction of ballistic characteristics for multi-layer spaced thin concrete targets. The results indicate that an increase in counterclockwise angular velocity leads to a positive rise in radial residual velocity behind the target and upward trajectory deflection, while clockwise angular velocity produces the opposite effect. These findings clearly demonstrate that projectile angular velocity is a critical and non-negligible parameter in thin-target penetration. For single-layer target scenarios, the model demonstrated strong predictive performance, with mean MSE values for the training and test sets stabilizing around 0.0012 and 0.0019, respectively. In multi-layer target predictions, while maintaining accuracy (maximum relative error in residual velocity of 10.65% and maximum absolute error in attitude angle of 3.47°), the model's computation time was only 0.05% of that required by traditional numerical simulation methods. This study not only reveals the influence of the key factor—projectile angular velocity—on penetration ballistics but also offers a novel modeling paradigm of "data-driven + physics-equation fusion," providing an important methodological reference for weapon damage effectiveness assessment and design optimization.