期刊文献+

基于BP神经网络和遗传算法的环形药型罩优化 被引量:2

Optimization of Annular Liner Based on BP Neural Network and Genetic Algorithm
下载PDF
导出
摘要 环形聚能装药相比传统的聚能装药具有侵彻口径大的优势,为了得到可以形成稳定的环形聚能射流的最优环形药型罩结构,提出了一种将数值模拟结果与BP神经网络相结合,并通过遗传算法对环形药型罩进行优化设计的方法。首先,利用正交试验法对环形药型罩进行方案设计,得到各因素对环形聚能射流稳定性的重要程度,其次利用LS-DYNA软件进行数值模拟得到最初的样本数据,然后通过MATLAB软件拟合出神经网络训练所需的样本数据,接着将环形药型罩结构参数作为BP神经网络的输入,射流头部速度、射流横向速度、射流长度分别作为输出进行训练,同时将测试值作为适应度,最后结合遗传算法选择最优的环形药型罩结构参数。研究结果表明:影响环形聚能射流成形的主要因素是药型罩口径和锥顶角,次要因素为药型罩罩顶厚、内罩偏移量和外罩偏移量;当药型罩罩顶高为0.81 mm,药型罩口径为15.43 mm,罩顶角为61.89°,内罩偏移量为11.38%,外罩偏移量为14.36%时所形成的环形射流形态比正交实验所得环形聚能射流更好。 Compared with traditional shaped charge,annular shaped charge has the advantage of penetrating large aperture.In order to obtain the optimal annular liner structure that can form a stable annular shaped charge jet,a method was proposed to optimize the annular liner by combining the numerical simulation results with BP neural network and genetic algorithm.Firstly,the scheme of annular shaped charge liner was designed by the orthogonal test method,and the importance of each factor to the stability of annular shaped charge jet was obtained.The numerical simulation was carried out to obtain the test data by LS-DYNA.The sample date,which was required for neural network training,was fitted by MATLAB.Next,the structural parameters of the annular liner and the jet tip velocity,horizontal velocity and jet length were used as the input and output of BP neural network,respectively.At the same time,the achieved test values were adopted as the fitness.Combining with genetic algorithm,the optimal structural parameters of the liner can be selected.The results show that the liner diameter and cone angle are the major factors affecting the annular shaped charge,while the liner top thickness,inner liner offset and outer liner offset are the minor factors.When the top thickness of the liner is 0.81 mm,the diameter of the liner is 15.43 mm,the cone angle of the liner is 61.89°,the offset of the inner liner is 11.38%,and the offset of the outer liner is 14.36%,the annular shaped jet formed by the shaped charge liner is better than the annular shaped charge jet obtained by the orthogonal experiment.
作者 陈浩 祖旭东 黄正祥 CHEN Hao;ZU Xudong;HUANG Zhengxiang(School of Mechanical Engineering,Nanjing University of Science and Technology,Nanjing 210094,China)
出处 《弹道学报》 CSCD 北大核心 2022年第4期1-7,共7页 Journal of Ballistics
基金 基础加强计划领域基金(2020-JCJ2-JJ-405)。
关键词 环形聚能装药 正交试验法 BP神经网络 遗传算法 结构优化 药型罩 annular shaped charge orthogonal experimental method BP neural network genetic algorithm structural optimization liner
  • 相关文献

参考文献9

二级参考文献53

共引文献68

同被引文献10

引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部