防空作战中空袭兵器识别仿真研究

Air-Attack Weapons Identification Based on Variable Precision Rough Set and RBF Neutral Network Model

防空作战对空袭兵器进行准确、快速的识别,可以辅助指挥员迅速做出合理的作战决策,增强防空武器系统的作战效能,而在解决特征维度较高的空袭兵器识别问题时,使用径向基神经网络模型存在着"组合爆炸"的缺陷。针对"组合爆炸"问题,采用一种变精度粗糙集-径向基神经网络的改进模型,通过引入变精度粗糙集属性约简得到最优识别系统,然后系统输入到径向基神经网络中进行识别。由于模型有效地简化了神经网络的规模和结构,在保证可靠性的同时,缩短了网络运算时间,提高了系统的实时性,避免了"组合爆炸"的发生。仿真结果表明,改进后的模型是可行的,为防空作战中空袭兵器识别提供了参考。

The accurate and rapid air- attack weapon recognition during the air strikes can assist commanders with quick operational decisions and enhance combat effectiveness of air defense weapon system. While dealing with the problem of multi - dimension air - attack weapon recognition, the traditional model of neural network has the defect of combination explosion. To solve the problem, an improved model is proposed based on variable precision rough set （VPRS） model and the RBF neural network model. Firstly,the optimal decision system was determined by the at- tributed reduction of VPRS. Then, the RBF neural network model recognized the air - attack weapon through the system. The model effectively simplified the size and structure of the neural network. While ensuring the accuracy at the same time,the improved model reduced the network operation time and improved the real - time performance of the whole system. The results show that the improved model can provide the air - attack weapon recognition with a reference.