计算机应用 ›› 2018, Vol. 38 ›› Issue (6): 1535-1541.DOI: 10.11772/j.issn.1001-9081.2017122944

• 人工智能 • 上一篇    下一篇

基于云模型的仿真可信度评估方法

郑垚宇1, 方洋旺1, 魏贤智1, 陈少华1, 高翔2, 王宏柯1, 彭维仕1   

  1. 1. 空军工程大学 航空航天工程学院, 西安 710038;
    2. 95889部队 武器系统与运用工程部, 甘肃 酒泉 735018
  • 收稿日期:2017-12-15 修回日期:2018-02-07 出版日期:2018-06-10 发布日期:2018-06-13
  • 通讯作者: 郑垚宇
  • 作者简介:郑垚宇(1993-),男,河南林州人,硕士研究生,主要研究方向:火力指挥控制与评估工程;方洋旺(1966-),男,安徽安庆人,教授,博士,主要研究方向:导弹制导与控制、随机最优理论与应用、非线性控制;魏贤智(1962-),男,四川泸县人,教授,博士,主要研究方向:武器系统总体技术及作战使用;陈少华(1965-),男,浙江余姚人,副教授,硕士,主要研究方向:武器系统应用工程;高翔(1985-),男,河南信阳人,工程师,博士,主要研究方向:武器系统效能评估及作战使用;王宏柯(1965-),男,陕西扶风人,副教授,硕士,主要研究方向:武器系统应用工程;彭维仕(1987-),男,广西桂林人,博士研究生,主要研究方向:航空武器系统总体设计、仿真与控制。
  • 基金资助:
    国家自然科学基金重大项目(61627901);国家自然科学基金面上项目(2014JQ8339)。

Evaluation method for simulation credibility based on cloud model

ZHENG Yaoyu1, FANG Yangwang1, WEI Xianzhi1, CHEN Shaohua1, GAO Xiang2, WANG Hongke1, PENG Weishi1   

  1. 1. Institute of Aeronautics and Astronautics Engineering, Air Force Engineering University, Xi'an Shaanxi 710038, China;
    2. Department of Weapon Systems and Utilization Engineering, Troops No. 95889 of PLA, Jiuquan Gansu 735018, China
  • Received:2017-12-15 Revised:2018-02-07 Online:2018-06-10 Published:2018-06-13
  • Supported by:
    This work is partially supported by the Major Projects of National Natural Science Foundation of China (61627901), the General Projects of National Natural Science Foundation of China (2014JQ8339).

摘要: 针对云模型在非正态分布条件下不适用的问题,提出了基于均匀分布的一维逆向云算法,并将其应用于仿真系统的可信度评估体系。首先,阐述了仿真可信度的重要性,并以实际工程为背景建立了某型装备抗干扰能力评估结果可信度评估指标;其次,运用基于云模型的仿真可信度评估方法对系统进行评估,并对该评估方法进行改进;最后,为了完善该评估方法,推导出基于均匀分布的一维逆向云算法,并且设计实验验证了该算法的有效性。仿真实验结果表明,该逆向云算法在较大数据时平均绝对误差小于5%,具有较高实用性,为云模型理论的完善提供一种思路。此外仿真可信度评估结果表明,该评估方法精度高,包含数据的分散度和凝聚度信息,可以进行更全面评估和错误数据预测。

关键词: 仿真可信度评估, 云模型, 逆向云算法, 均匀分布, 中心矩, 残差

Abstract: A cloud model is not suitable for non-normal distribution. In order to solve the problem, a new one-dimensional backward cloud algorithm based on uniform distribution was proposed and applied to the credibility evaluation system of simulation system. Firstly, the importance of simulation credibility was expounded, and the credibility evaluation index of the evaluation results for a type of equipment concerning anti-jamming capability was established based on the actual project background. Secondly, the system was evaluated by using the evaluation method for simulation credibility based on cloud model, and the evaluation method was improved. Finally, in order to improve the evaluation method, a one-dimensional backward cloud algorithm based on uniform distribution was derived, and the experiment was designed for verifying the validity of the algorithm. The simulation experimental results show that, the average absolute error of the proposed backward cloud algorithm is less than 5% for large data, which has high applicability and provides a way of thinking for the perfection of cloud model theory. In addition, the simulation credibility evaluation results show that, the proposed method has high accuracy and contains the data information of dispersion and agglomeration, which can provides more comprehensive evaluation and the prediction of error data.

Key words: simulation credibility evaluation, cloud model, backward cloud algorithm, uniform distribution, central moment, residual error

中图分类号: