摘要
结合光电干扰武器系统的工作过程,对影响目标威胁评估的各种因素进行了分析,讨论了常用威胁评估方法的缺点和不足,提出了基于神经网络的空中目标的威胁估计算法,利用神经网络良好的自适应能力和自学习能力,通过样本数据训练,确定各个因素之间的非线性复杂关系,并通过示例介绍了目标威胁值的解算过程;与层次分析法进行了比较,结果表明,神经网络可以很好地逼近各个因素之间的权重关系,提高了空中目标威胁估计算法的准确性和适应性。
This paper analyzed factors which affect threat assessment, discussed the defeat and shortage of general way for menace assessment and presented a threat assessment algorithm for Aerial Target based on Neural Network . By utilizing the good abilities of adapting and self-leaning and samples training, the networks approximated the nonlinear complex relationship of all factors. The process of solution is given by instantiation. Compared with AHP, the results show that BP neural network can successfully approximate the weights of all factors and the accuracy and adaptability of the threat assessment algorithm are improved.
出处
《计算机测量与控制》
CSCD
2008年第2期212-214,217,共4页
Computer Measurement &Control
基金
中科院二期创新项目(C04708Z)
关键词
威胁估计
神经网络
BP算法
光电干扰武器系统
threat assessment
neural network
BP algorithm
photoelectric interferential weapon system
