基于PSO改进RBF神经网络的威胁评估方法研究

Research on Threat Assessment Method Based on PSO Improved RBF Neural Network

导弹威胁评估是飞机对抗威胁过程中的一个重要环节,是干扰决策的前提。本文介绍了神经网络在军事领域的应用,接着通过粒子群算法优化RBF神经网络,利用优化后的网络解决导弹威胁评估问题,提出了基于粒子群优化RBF神经网络的威胁评估方法,使用粒子群对RBF神经网络参数寻优。用此方法与BP神经网络、RBF神经网络算法性能进行比较,结果表明此方法更有优势,能够快速、准确地评估威胁。

Missile threat assessment is an important link in the process of aircraft countering threats and is a prerequisite for interference decision-making. This paper introduces the application of neural net-work in the military field, and then optimizes RBF neural network through particle swarm optimization algorithm. The optimized network is used to solve the problem of missile threat assessment. A threat assessment method based on Particle swarm optimization RBF neural network is proposed, which uses particle swarm optimization to optimize the parameters of RBF neural network. Comparing the performance of this method with BP neural network and RBF neural network algorithms, the results show that this method has more advantages and can quickly and accurately evaluate threats.