摘要
为解决作战效能指标繁多、维度高、复杂性强,指标间存在冗余和相关性等问题,结合武器装备体系作战效能指标非线性特征,在创建作战效能评估指标体系基础上,采用改进的拉普拉斯特征映射算法(ISOLE)实现作战效能指标的约简。目的是将复杂的指标体系非线性映射到低维的指标体系中,减少效能指标的维数,保留关键性指标数据,去除冗余的指标数据。通过自编码神经网络算法实现约简指标重构、仿真检验,证实了应用改进后的拉普拉斯特征映射方法的可行性和有效性以及约简后效能指标的准确性。为开展武器装备体系作战效能评估时降低时间和空间的复杂度,提高作战效能评估效率提供了技术支持。
The combat effectiveness index,as the key to evaluating the victory or defeat of weapon and equipment systems,is an important basis for accurately evaluating the combat effectiveness of weapon and equipment systems.To solve the problems of multiple,high-dimensional,and complex combat effectiveness indicators,as well as redundancy and correlation between indicators,combined with the nonlinear practical characteristics of combat effectiveness indicators in weapon and equipment systems,an improved Laplace feature mapping algorithm(ISOLE)is used to reduce combat effectiveness indicators based on the creation of a combat effectiveness evaluation indicator system.The purpose is to non-linear map complex indicator systems to low dimensional indicator systems,reduce the dimensionality of performance indicators,retain key indicator data,and remove redundant indicator data.The feasibility and effectiveness of applying the improved Laplace feature mapping method,as well as the accuracy of the reduced performance indicators,have been confirmed through the implementation of a self coding neural network algorithm for reconstruction of reduction indicators and simulation verification.This provides auxiliary support for reducing the complexity of time and space in the subsequent evaluation of the combat effectiveness of weapon and equipment systems,and significantly improving the efficiency of combat effectiveness evaluation.
作者
张鹏
冯柯
侯林健
褚德天
赵凯
ZHANG Peng;FENG Ke;HOU Linjian;CHU Detian;ZHAO Kai(College of Field Operations Engineering,Army Engineering University of PLA,Nanjing 210000,China;Chinese People’s Liberation Army 94221 Unit,Rizhao 276800,China;Chinese People’s Liberation Army 93459 Unit,Tianjin 300000,China)
出处
《兵器装备工程学报》
2024年第S2期102-110,共9页
Journal of Ordnance Equipment Engineering
基金
全军军事类研究生资助课题(JY2023C089)。
作者简介
张鹏(1993-),男,硕士研究生,助理工程师,E-mail:zhangpeng002519@163.com;通信作者:冯柯(1968-),男,博士,博士生导师,E-mail:shuiyuantou2519@163.com。
