基于机器学习的侵彻弹安全性量化分级方法初探
Preliminary Study on Quantitative Classification Method of Penetrating Projectile Safety Based on Machine Learning
王锋 1汪衡 2刘宗伟 2王昭明 2谭力犁 2陆晓宇3
作者信息
- 1. 南京理工大学 机械工程学院,江苏,南京 210094;重庆红宇精密工业集团有限公司,重庆 402760
- 2. 重庆红宇精密工业集团有限公司,重庆 402760
- 3. 重庆红宇精密工业集团有限公司,重庆 402760;南京理工大学 网络空间安全学院,江苏,南京 210094
- 折叠
摘要
针对侵彻弹在子弹撞击、破片撞击及殉爆等机械冲击刺激条件下的安全性响应分析方法不健全、量化分级标准缺乏等问题,提出了一种基于机器学习的侵彻弹冲击安全性响应量化分级方法,基于Lee-Tarver装药反应度仿真数据和试验数据集,获得了装药反应度与冲击安全性响应等级的映射关系,实现了冲击安全性响应等级的定量预测,并开展了典型侵彻弹的子弹撞击、破片撞击及殉爆验证试验.结果表明:仿真预测结果与试验结果吻合较好,证明了侵彻弹冲击安全性响应量化分级方法的有效性,可为侵彻弹的冲击安全性工程设计提供重要支撑.
Abstract
Under the mechanical impact stimulation condition,such as the bullet impact,fragment impact and martyr explosion,there still exist the problems that the safety response analysis method is imperfect and the quantitative classification method is lacking.To tackle these problems,a machine learning based quantitative classification method for the penetrating projectile under the impact safety response was proposed.With the sim-ulation data of Lee-Tarver charge reactivity and historical experimental data,a mapping relation was established between the charge reactivity and the impact safety response classification for quantitative classification under the impact safety response.In addition,experiments,including the bullet impact,fragment impact and martyr ex-plosion tests,were used to verify the performance of the proposed method.The results show that the simulation results and experimental results are consistent,which verifies the effectiveness of the proposed method and can support the impact safety engineering design for penetrating projectile in the future.
关键词
侵彻弹/冲击安全性/量化分级/数值模拟/试验验证Key words
penetrating projectile/impact safety/quantitative classification/numerical simulation/experiment-al validation引用本文复制引用
出版年
2025
NETL
NSTL
万方数据