基于麻雀搜索算法优化BP神经网络的弹丸射程预测研究

Research on Projectile Range Prediction Based on SSA-BP Algorithm

弹丸发射参数、气象条件等会影响弹丸射程,其构成的影响体系复杂并难以准确预测。针对BP预测算法会因初始权值和阈值取值不当导致陷入局部最优的问题,建立了麻雀搜索算法(SSA)优化BP神经网络的弹丸射程预测模型,以弹丸射程作为输出指标,选取弹丸初速、发射角度和风力条件作为影响因素输入,经过数据预处理后进行弹丸射程预测;同时与粒子群算法(PSO)和遗传算法(GA)优化BP神经网络预测模型的预测精度进行对比,验证SSA优化BP神经网络模型的预测效果。结果表明,SSA-BP预测模型的平均绝对误差、均方根误差和平均绝对百分比误差分别为10.456 4 m、11.831 3 m和0.058 13%,低于BP、PSO-BP、GA-BP预测模型的相应评估指标,所以SSA-BP模型的预测精度高于BP、PSO-BP、GA-BP预测模型,其可以为弹丸射程预测和远程火力打击研究提供支持。

Projectile launch parameters and meteorological conditions will affect range,and the influe-nce system is complex and difficult to predict accurately.Aiming at the problem that BP prediction algorithm will fall into local optimum due to improper initial weight and threshold value,a projectile range prediction model based on SSA algorithm optimized BP neural network is established.Projectile range is used as the output index,and projectile muzzle velocity,firing angle and wind conditions are selected as the input of influencing factors.After data pre-processing,projectile range is predicted.At the same time,the prediction accuracy of the BP neural network prediction model optimized by PSO and GA algorithms is compared with that of the SSA optimized BP neural network model to verify the prediction effect of the latter.The results show that the MAE,RMSE and MAPE of the SSA-BP prediction model are 10.4564 m,11.8313 m and 0.05813%,respectively,which are lower than the corresponding evaluation indexes of BP,PSO-BP and GA-BP prediction models.Therefore,the prediction accuracy of the SSA-BP model is higher than that of the BP,PSO-BP and GA-BP prediction models.The results can provide support for projectile range prediction and long range fire strike research.