Journal of Systems Engineering and Electronics ›› 2011, Vol. 33 ›› Issue (5): 1055-.doi: 10.3969/j.issn.1001-506X.2011.05.19

• 系统工程 • 上一篇    下一篇

基于支持向量机的部队作战效能评估

程恺, 车先明, 张宏军, 张睿, 单黎黎   

  1. 解放军理工大学工程兵工程学院, 江苏 南京 210007
  • 出版日期:2011-05-25 发布日期:2010-01-03

Evaluation method of operational effectiveness based on support vector machine

CHENG Kai, CHE Xian-ming, ZHANG Hong-jun, ZHANG Rui, SHAN Li-li   

  1. Engineering Institute of Corps of Engineers, University of Science & Technology of the PLA, Nanjing 210007, China
  • Online:2011-05-25 Published:2010-01-03

摘要:

随着信息化战争复杂性与不确定性的增加,对部队作战效能的评估提出了更高要求。在分析支持向量机和增强最简半自治适应性作战神经网络仿真工具箱(enhanced irreducible semiautonomous adaptive combat neural simulation toolkit, EINSTein)特点的基础上,建立了基于支持向量机的部队作战效能评估模型,完成从评估指标到作战效能值的非线性映射。在一定作战想定背景下,结合EINSTein系统产生的仿真数据对模型进行了验证。仿真结果表明,与误差反向传播神经网络(back propagation neural networks, BPNN)相比,基于支持向量机的效能评估模型能够降低人为因素的影响,更准确地评估部队作战效能,是一种有效的评估方法。

Abstract:

With the increasing complexity and uncertainty of the information war, a higher requirement is put forward to the evaluation of operational effectiveness of the force. After analyzing the characteristics of support vector machine and enhanced irreducible semi-autonomous adaptive combat neural simulation toolkit (EINSTein), the evaluation model of operational effectiveness is established to accomplish nonlinear mapping from evaluation index to combat effectiveness based on support vector machine. In a certain battle scenario, the model with the simulation data produced by EINSTein system is validated. Simulation results show that compared with error back propagation neural networks(BPNN), the support vector machine model which can reduce the impact of human factors and assess the operational effectiveness more accurately is an effective evaluation method.