BP神经网络导弹攻击区火控工作式拟合

Fire Control Formula Fitting of Missile Attack Envelope Based on BP Neural Network

采用BP神经网络和改进的BP神经网络方法对某型空空导弹理论攻击区进行处理来构造火控计算机导弹攻击区火控工作式,通过BP神经网络和改进的BP神经网络仿真,获得典型发射条件下的失机发射概率、超界发射概率和平均成功发射概率。结果表明BP神经网络解算导弹攻击区的正确性和有效性,改进的BP神经网络有更快的收敛速度,更高拟合精度和平均发射成功率。

In order to obtain control formula,air-to-air missile theoretical attack envelope is processed,utilizing BP neural network and the improved BP neural network method.The simulation gains outside probability of missed launch opportunity,probability of out of bounds launch and mean successful launch opportunity.Results show it is valid to compute the missile attack envelope by BP neural network.comparing with the BP neural network method,the improved BP neural network method can increase the accuracy of fitting,mend convergence velocity and successful launch probability.