首页|基于APSO-BP神经网络的末敏弹作战效能评估方法

基于APSO-BP神经网络的末敏弹作战效能评估方法

扫码查看
针对末敏弹结构复杂、影响因素多、作战效能分析困难等问题,以末敏弹命中概率作为目标函数,建立了作战效能评估指标体系,提出了一种基于APSO-BP神经网络的作战效能评估模型,并构建了BP神经网络和PSO-BP神经网络2种对比模型,利用MATLAB工具对3种模型进行了仿真分析.结果显示,APSO-BP神经网络的运行耗时为0.651 3 s,均方误差为0.003 2,相关系数为0.978 9;PSO-BP神经网络的运行耗时为2.015 4 s,均方误差为0.007 5,相关系数为0.968 8;BP神经网络的运行耗时为14.137 5 s,均方误差为0.015 9,相关系数为0.890 0.APSO-BP神经网络评估模型运行耗时更短,预测精度更高,对于末敏弹的作战运用具有重要的理论意义和现实价值.
Evaluation method of operational effectiveness of terminal sensitive projectiles based on APSO-BP neural network
In view of the complex structure,many influencing factors and difficulties in operational effectiveness analysis of terminal sensitive projectiles,the operational effectiveness evaluation index system was established by taking the hit probability of the terminal sensitive projectiles as the objective function.An operational effectiveness evaluation model based on APSO-BP neural network was proposed,two models of BP neural network and PSO-BP neural network were constructed for comparison,and the three models were simulated by Matlab.The results show that the running time of APSO-BP neural network is 0.651 3 s,the mean square error is 0.003 2,and the correlation coefficient is 0.978 9.The running time of PSO-BP neural network is 2.015 4 s,the mean square error is 0.007 5,and the correlation coefficient is 0.9688.The running time of BP neural network is 14.137 5 s,the mean square error is 0.015 9,and the correlation coefficient is 0.890 0.APSO-BP neural network evaluation model has shorter running time and higher prediction accuracy,which has important theoretical significance and practical value for the operational application of terminal sensitive projectiles.

terminal sensitive projectileshit probabilityperformance evaluationBP neural networkparticle swarm optimizationAPSO-BP algorithm

唐永果

展开 >

陆军工程大学石家庄校区,石家庄 050000

末敏弹 命中概率 效能评估 BP神经网络 粒子群算法 APSO-BP算法

&&

装供[2023]920号

2024

兵器装备工程学报
重庆市(四川省)兵工学会 重庆理工大学

兵器装备工程学报

CSTPCD北大核心
影响因子:0.478
ISSN:2096-2304
年,卷(期):2024.45(10)