航空学报 > 2024, Vol. 45 Issue (12): 329453-329453   doi: 10.7527/S1000-6893.2023.29453

基于相关证据推理规则的激光惯组健康评估

唐帅文1, 曹友2(), 张朋3, 姜江1   

  1. 1.国防科技大学 系统工程学院,长沙 410073
    2.火箭军士官学校 控制技术教研室,青州 262500
    3.西安卫星测控中心 宇航动力学国家重点实验室,西安 710043
  • 收稿日期:2023-08-16 修回日期:2023-08-30 接受日期:2023-09-08 出版日期:2024-06-25 发布日期:2023-09-20
  • 通讯作者: 曹友 E-mail:936756268@qq.com
  • 基金资助:
    国家自然科学基金(62303474);山东省自然科学基金(ZR2023QF010)

Health assessment of LIMU based on evidential reasoning rule with dependent evidence

Shuaiwen TANG1, You CAO2(), Peng ZHANG3, Jiang JIANG1   

  1. 1.College of Systems Engineering,National University of Defense Technology,Changsha  410073,China
    2.Control Technology Teaching and Research Office,Rocket Army Sergeant School,Qingzhou  262500,China
    3.State Key Laboratory of Aerospace Dynamics,Xi’an Satellite Control Center,Xi’an  710043,China
  • Received:2023-08-16 Revised:2023-08-30 Accepted:2023-09-08 Online:2024-06-25 Published:2023-09-20
  • Contact: You CAO E-mail:936756268@qq.com
  • Supported by:
    National Natural Science Foundation of China(62303474);Natural Science Foundation of Shandong Province(ZR2023QF010)

摘要:

激光惯组(LIMU)是复杂装备系统的重要导航单机,其良好的健康状态对于确保系统安全稳定工作具有重要意义。针对激光惯组健康评估中面临的评估指标体系不健全、评估框架不一致、评估指标不独立和评估模型参数不确定问题,在激光惯组工作机理分析的基础上,构建健康评估指标体系,提出基于相关证据推理规则(ERR-DE)的激光惯组健康评估方法。首先,提出转换矩阵的概念,实现了不同评估框架的统一转换。其次,采用变异系数法确定证据权重,采用基于距离的方法确定证据可靠度。然后,基于证据可靠度提出证据融合顺序确定机制,提出基于相对总体相关系数(RTDC)的证据相关性分析方法,并在证据推理规则(ERR)的基础上提出了相关证据推理规则。再者,构建参数优化模型以确定最优评估参数,提高评估精度,进一步对评估算法进行了总结。最后,通过某型激光惯组健康评估案例和对比研究,验证了所提方法的有效性。

关键词: 证据推理规则, 激光惯组, 健康评估, 证据相关性, 参数优化

Abstract:

The Laser Inertial Measurement Unit (LIMU) is an important navigation unit for the complex equipment system, and its good health state is of great significance for ensuring safe and stable operation of the system. In the health assessment of LIMU, there exist the problems of imperfect assessment indicator system, inconsistent assessment framework, dependent assessment indicators and uncertain parameters of assessment model. Based on the analysis of working mechanism of LIMU, a health assessment indicator system is constructed, and a health assessment method of LIMU is proposed based on the Evidential Reasoning Rule with Dependent Evidence (ERR-DE). Firstly, the concept of transformation matrix is proposed to achieve unified transformation of different assessment frameworks. Secondly, the coefficient of the variation-based weighting method is used to determine evidence weights, and the distance-based method is employed to determine evidence reliabilities. Then, based on the evidence reliability, a mechanism to determine the aggregation sequence of evidences is proposed, and an evidence dependence analysis method is proposed based on the Relative Total Dependence Coefficient (RTDC). Based on the Evidential Reasoning Rule (ERR), the ERR-DE is proposed. Furthermore, a parameter optimization model is constructed to determine the optimal assessment parameters and improve the assessment accuracy. The assessment algorithm is further summarized. Finally, the effectiveness of the proposed method is verified through a health assessment case study and comparative studies on a certain type of LIMU.

Key words: evidential reasoning rule, laser inertial measurement unit, health assessment, evidence dependence, parameter optimization

中图分类号: