Abstract:

This study addresses the critical challenges in risk assessment of natural fragmentation warhead explosions by proposing a comprehensive parametric analytical methodology. Through systematic integration of key components including stochastic fragment generation, precise trajectory calculation, three-dimensional_h probability evaluation, and quantitative human damage assessment, we have established a multi-scale coupled risk evaluation system. In terms of fragment kinematics modeling, we developed an aerodynamic surrogate model based on artificial neural networks. By introducing fragment sphericity parameters and Mach number as dual variables, the model significantly improves the trajectory calculation accuracy for naturally fragmenting projectiles under random tumbling conditions. For hazard effect evaluation, we innovatively proposed a three-dimensional pie-shaped target model. By incorporating detailed human geometric parameters and considering the relative position between fragment trajectories and human targets, the model enables more accurate calculation of fragment hit probability on personnel. Regarding risk quantification, we constructed a multi-level probabilistic assessment framework that integrates AIS injury scales with fragment kinetic energy distribution. Validation using 155mm projectiles demonstrates that this method can not only describe fragment hazards through uniform annular safety distances but also generate two-dimensional spatial distributions of fragment hazard probability. The research achieves quantitative characterization of the entire process encompassing initial stochasticity, motion complexity, and progressive damage effects of fragments. It provides new analytical tools and decision support for dynamic safety distance determination in ammunition storage and transportation, optimized safety zoning design for firing ranges, and industrial explosion protection.

Key words: fragment hazard, terminal effects, risk assessment, safety distance