摘要
针对现有作战效能评估方法中评估指标复杂、主观因素强的问题,提出一种基于自编码器(AE)和BP神经网络的评估方法.首先通过自编码器对输入数据进行特征提取,减少冗余信息并降低数据相关性;然后将自编码器提取的特征向量作为BP神经网络的输入信息进行训练,得到BP神经网络权值;最后基于OODA环理论构建地空导弹武器系统作战效能评估体系,并通过仿真验证所提方法的有效性.仿真结果表明,该方法误差达到预期标准,能够有效完成对地空导弹武器系统作战效能的评估.
Aiming at the problems of complex evaluation indexes and strong subjective factors in the existing operational effectiveness evaluation methods,this paper proposes an evaluation method based on auto-encoder(AE)and BP neural network.Firstly,feature extraction of input data is carried out by AE to reduce redundant information and lower data correlation.Then,the feature vector extracted from AE is trained as the input information of BP neural network,so as to obtain the weight of BP neural network.Finally,the operational effectiveness evaluation system of surface-to-air missile weapon system is constructed based on the OODA loop theory,and the effectiveness of the proposed method is verified by simulations.The simulation results show that this method has the error reach the expected standard and can be used to effectively evaluate the operational effectiveness of ground-to-air missile weapon system.
作者
麻晓伟
张琳
MAXiaowei;ZHANG Lin(Air Force Engineering University,Xi’an 710051,China)
出处
《空天预警研究学报》
2022年第6期425-429,共5页
JOURNAL OF AIR & SPACE EARLY WARNING RESEARCH
基金
国家自然科学基金资助项目(62106283)。
关键词
地空导弹武器系统
效能评估
BP神经网络
自编码器
深度学习
ground-to-air missile weapon system
effectiveness assessment
BP neural network
auto-encoder
deep learning
