系统仿真学报 ›› 2019, Vol. 31 ›› Issue (12): 2678-2684.doi: 10.16182/j.issn1004731x.joss.19-0126

• 仿真系统与技术 • 上一篇    下一篇

基于主成分分析的仿真结果评估方法

鞠儒生, 蔡子民, 杨妹, 王松   

  1. 国防科技大学系统工程学院,湖南 长沙 410072
  • 收稿日期:2019-03-26 修回日期:2019-05-13 发布日期:2019-12-13
  • 作者简介:鞠儒生(1976-),男,江苏泰兴,博士,副教授,研究方向为系统仿真数据评估分析,人工智能及其应用。
  • 基金资助:
    国家自然科学基金(61673388)

Evaluation Method of Simulation Results Based on Principal Component Analysis

Ju Rusheng, Cai Zimin, Yang Mei, Wang Song   

  1. College of Systems Engineering, National University of Defense Technology, Changsha 410072, China
  • Received:2019-03-26 Revised:2019-05-13 Published:2019-12-13

摘要: 复杂系统仿真往往会产生海量的结果数据,这些数据类型多样、关联复杂,用户通常不能直接使用,必须经过有效评估分析,才能寻找到仿真结果数据背后隐藏的规律。根据复杂仿真系统结果数据的特点,利用主成分分析方法降低数据分析的维度,探索系统输入输出数据之间的关联,寻找到影响作战效能的关键要素,通过案例进行应用验证,对提炼出的各个主成分因素进行深入探索,说明了该方法的有效性,分析策略可供广大同行研究参考。

关键词: 复杂系统仿真, 评估, 主成分分析, 作战效能

Abstract: A large amount of result data are produced by complex system simulation. These data types are diverse and complex. Usually they cannot be used directly. The inner pattern can only be found by effective evaluation. Based on the characteristics of complex combat simulation results, this paper utilizes the Principal Component Analysis method to reduce the dimension of data analysis and explore the relationship between the system input data and output data. The key influence factors of the measure of force effectiveness is found. The application verification is done by typical examples. The selected principle factors are explored in deepth to verify the effectiveness of the method. The strategy can be referred by other peer.

Key words: complex system simulation, evaluation, principle component analysis, force effectiveness

中图分类号: