弹道学报2025，Vol.37Issue(3)：49-57,100,10.DOI:10.12115/ddxb.2024.06008
基于RL-NSGAⅡ算法的埋头式膨胀波火炮内弹道性能优化
Optimization of Interior Ballistic Performance of Submerged Expansion-wave Artillery Based on RL-NSGAⅡ Algorithm
彭碧荣 1薛晓春 1曹永杰 2常人九 1黄磊 1余永刚1
作者信息
- 1. 南京理工大学 能源与动力工程学院,江苏 南京 210094
- 2. 西北机电工程研究所,陕西 咸阳 712099
- 折叠
摘要
Abstract
In order to solve the problem of excessive recoil of a 40 mm case-telescoped ammunition gun,the expansion wave technology was adopted to reduce recoil by activating the afterburner during a specific bolt opening time.The opening latch time of the expansion-wave artillery is crucial to its performance.Although the rear spraying device can keep the initial velocity of projectile unchanged under the optimal opening latch time,proper adjustment of the opening latch time can significantly improve the recoil reduction effect while slightly reducing the initial velocity of projectile.Based on the firing characteristics of the submerged expansion wave gun,an internal ballistic model was established,and the numerical calculations was carried out.By optimizing the charge quality,powder condition and bolt opening time of main charge,the projectile initial-velocity decay-rate and the recoil reduction efficiency of the body tube were taken as the optimization objectives.To overcome the NSGAⅡ algorithm's lack of diversity and falling into local optimization in the iterative process,the reinforcement learning was introduced to form the RL-NSGAⅡ algorithm.Reinforcement learning improves the convergence speed and accuracy by dynamically adjusting the algorithm execution steps.The results show that the RL-NSGAⅡ algorithm is significantly better than the NSGAⅡ algorithm,which verifies the effectiveness of reinforcement learning and successfully optimizes the internal ballistic performance.After optimization,the projectile velocity decay-rate is 1.43%,and the body-tube recoil-reduction-efficiency reaches 64.73%,which not only improves the decay rate of projectile muzzle-velocity,but also significantly increases the recoil reduction efficiency.关键词
埋头弹/膨胀波技术/减后坐/多目标优化/强化学习Key words
case-telescoped ammunition gun/expansion-wave technology/recoil reduction/multi-objective optimization/reinforcement learning分类
军事科技引用本文复制引用
彭碧荣,薛晓春,曹永杰,常人九,黄磊,余永刚..基于RL-NSGAⅡ算法的埋头式膨胀波火炮内弹道性能优化[J].弹道学报,2025,37(3):49-57,100,10.
