航空学报 > 2020, Vol. 41 Issue (9): 323983-323983   doi: 10.7527/S1000-6893.2020.23983

基于通道校准和HMM的机载雷达健康状态评估

陈建平, 徐皓吉, 张勇   

  1. 中国航空工业集团公司 雷华电子技术研究所, 无锡 214063
  • 收稿日期:2020-03-17 修回日期:2020-04-12 出版日期:2020-09-15 发布日期:2020-05-28
  • 通讯作者: 陈建平 E-mail:chenjianping@buaa.edu.cn

Health state assessment of airborne radar based on channel calibration and HMM

CHEN Jianping, XU Haoji, ZHANG Yong   

  1. AVIC Leihua Electronic Technology Research Institute, Wuxi 214063, China
  • Received:2020-03-17 Revised:2020-04-12 Online:2020-09-15 Published:2020-05-28

摘要: 性能退化与间歇故障是机载雷达健康状态的重要反映。针对传统机载雷达健康评估方法缺乏整机级的监测指标、未能综合考虑性能退化和间歇故障,传统BIT电路难以监测间歇故障等工程难题,提出了基于通道校准链路,综合运用通道校准仿真技术、正态云、HMM建立起健康状态评估流程。首先,对通道误差与间歇故障进行分析,得到4类校正系数,建立误差与间歇故障注入仿真流程;然后,基于正态云,利用校正系数的样本方差来表征雷达系统的健康状态,提出统一健康模型,并给出HMM拓扑结构与参数设计方法;最后,建立起基于数据驱动的评估流程并开展应用研究。仿真实验验证了误差和间歇故障仿真流程的可行性、健康状态评估方法的有效性,可获得大于95%的评估准确率;应用案例表明所提方法可行,能解决工程上机载雷达健康状态评估的问题。

关键词: 健康状态评估, 机载雷达, 间歇故障, 通道校准, 隐马尔可夫模型, 正态云

Abstract: Performance degradation and intermittent faults are important indicators of airborne radar health state. Traditional airborne radar health state assessment methods lack monitoring indicators at the overall system level and fail to comprehensively consider performance degradation and intermittent faults, and it is difficult for the traditional BIT circuit to monitor intermittent faults. Aiming at the above engineering problems, this paper puts forward a method based on the channel calibration circuit, combining the channel calibration simulation technology, the normal cloud model and HMM to build a health state assessment process. First, the channel errors and intermittent faults are analyzed to obtain four kinds of correction coefficients, thus establishing the simulation process of errors and intermittent fault injection. Then, based on the normal cloud model, the sample variance of the correction coefficients is used to represent the health state of the radar system, a unified health model is proposed, and the HMM topology and parameter design method are given. Finally, the data-driven evaluation process is established and its application studied. The simulation experiment verifies the feasibility of the errors and intermittent fault simulation process and the validity of the health state evaluation method, reaching an assessment accuracy of over 95%. The application case shows that the proposed method is feasible and can solve the problem of health state evaluation of airborne radars in engineering.

Key words: health state assessment, airborne radars, intermittent faults, channel calibration, HMM, normal cloud models

中图分类号: