系统工程与电子技术 ›› 2022, Vol. 44 ›› Issue (8): 2621-2627.doi: 10.12305/j.issn.1001-506X.2022.08.27

• 制导、导航与控制 • 上一篇    下一篇

基于正态-逆伽马分布的反巡航导弹命中概率估计方法

李康, 史宪铭*, 李广宁, 刘昊邦   

  1. 陆军工程大学石家庄校区, 河北 石家庄 050003
  • 收稿日期:2021-04-16 出版日期:2022-08-01 发布日期:2022-08-24
  • 通讯作者: 史宪铭
  • 作者简介:李康(1997—), 男, 硕士研究生, 主要研究方向为装备管理与保障、系统工程|史宪铭(1975—), 男, 副教授, 博士, 主要研究方向为系统工程、供应工程理论与应用|李广宁(1992—), 男, 硕士研究生, 主要研究方向为装备管理与保障、系统工程|刘昊邦(1998—), 男, 硕士研究生, 主要研究方向为装备管理与保障、系统工程
  • 基金资助:
    军内科研基金项目(LJ20202C050369);全军军事类研究生项目(JY2020B086)

Estimation method of hit probability of anti-cruise missile weapon based on normal-inverse Gamma distribution

Kang LI, Xianming SHI*, Guangning LI, Haobang LIU   

  1. Shijiazhuang Campus, Army Engineering University, Shijiazhuang 050003, China
  • Received:2021-04-16 Online:2022-08-01 Published:2022-08-24
  • Contact: Xianming SHI

摘要:

有效估计反巡航导弹武器系统命中概率是确定反巡航导弹武器系统弹药消耗的基础, 反巡航导弹试验成本高, 导致数据不足, 给估计反巡航导弹武器系统的命中概率带来困难。本文以防空试验中靶面落点服从二维联合正态分布的现象为切入点, 结合巡航导弹等效矩形域的特点, 提出了基于正态-逆伽马分布的反巡航导弹武器系统命中概率估计方法。选取正态分布和逆伽马分布作为反巡航导弹武器系统命中概率特征参数的先验分布, 并通过贝叶斯分析方法得到相应的后验分布函数。采用参数估计得到先验分布中的超参数, 确定命中概率的最大后验估值, 示例表明了方法的可操作性, 具有广阔应用前景。

关键词: 命中概率, 贝叶斯推断, 正态逆伽马分布, 最大后验估计

Abstract:

Effective estimation of hit probability of anti-cruise missile weapon system is the basis of determining ammunition consumption of anti-cruise missile weapon system. The problem of insufficient data caused by the high cost in the anti-cruise missile test makes it difficult to estimate the hit probability of the anti-cruise missile weapon system. Considering that the target surface's landing point obeys the two-dimensional joint normal distribution in the air defense test, and with the characteristics of equivalent rectangular domain for cruise missile, a hit probability estimation method of anti-cruise missile weapon system based on normal-inverse Gamma distribution is proposed. Normal distribution and inverse Gamma distribution are selected as the prior distribution of hit probability characteristic parameters of the anti-cruise missile weapon system, and the corresponding posterior distribution function is obtained by the Bayesian analysis method. Parameter estimation is used to obtain the hyper-parameter in the prior distribution, and the maximum posterior estimate of the hit probability is determined. The example shows that the method is operable and has broad application prospects.

Key words: hit probability, Bayesian inference, normal-inverse Gamma distribution, maximum posterior estimation

中图分类号: