基于案例推理的航天测控设备故障诊断  

Fault Diagnosis of Space TT&C Equipment Based on Case-based Reasoning

作　　者：刘涌[1] 李海潮[1] 弓喜忠 詹炳光 

机构地区：[1]西安卫星测控中心,西安710043

出　　处：《电讯技术》2017年第2期236-242,共7页Telecommunication Engineering

摘　　要：针对航天测控设备故障诊断难度日益增大的问题,将基于案例推理方法应用于设备故障诊断中。首先提出用故障信息规范描述、特征属性集和属性字典的三元组合结构来表示案例的方法,利用数据库管理系统建立故障案例库;然后通过对目标案例的特征属性集元素进行删补及位序调整,构造出与源案例特征属性集同构的特征属性阵列,以此计算案例的局部相似度;最后利用最近邻搜索策略计算案例间的全局相似度,完成对目标案例的搜索和问题的求解过程。应用实例表明,该方法案例表示结构清晰,易于建立案例库,能快速获得故障问题的解决方案,且诊断精度较高,有效提高了设备故障诊断效率。With the ever-increasing difficuhy of the fault diagnosis on space tracking, telemetry and command（TT＆C） equipment, case-based reasoning（CBR） methodology is applied to the fault diagnosis of equipment. Firstly,a method to express fault cases is proposed by using a three-element composite structure, which consists of the normalized description of fault, the set of feature properties and the attribute dic- tionary of faults, and a database of fault cases is built by using database management system. Secondly, a feature attribute array is constructed as the set of feature properties of the target fault case, which is rebuilt to be isomorphic to the set of feature property of the source fault case by deleting or adding and sorting its elements. The array is used to calculate the partial similarity. Finally, the nearest neighbor search method is adopted to calculate the overall similarity between two fault cases, which is used for the completion of the target fault case search and the process of troubleshooting. The application shows that the expressing struc- ture of fault case is clear in this method, and this method can easily establish the database of fault cases and rapidly find solution to the equipment fault with high accuracy, and it can effectively improve the efficiency of fault diagnosis.

关 键 词：航天测控设备 故障诊断 基于案例推理 特征属性阵列 案例相似度 

分 类 号：V55[航空宇航科学与技术—人机与环境工程] TP182[自动化与计算机技术—控制理论与控制工程]