基于知识推理的航天器自主故障诊断方法  

Knowledge Inference Based Self Fault Diagnosis Method for Spacecraft

作　　者：龚健[1] 杨桦[1] 赵玮[1] 乔磊[1] 

机构地区：[1]北京控制工程研究所,北京100190

出　　处：《空间控制技术与应用》2011年第4期19-23,共5页Aerospace Control and Application

基　　金：国家重大专项

摘　　要：面向深空探测这类型号任务,提出一种基于知识推理的航天器自主故障诊断的智能方法,采取直接线性存储的知识库结构、主动监测事实更新并触发推理以及基于散列算法的事实管理三项关键技术,适用于航天器硬件处理能力和存储容量受限的应用环境.采用C语言在基于TSC695处理器的嵌入式系统上实现了该故障诊断方法,实验的结果表明该方法有效.Aiming at missions such as deep space exploration, an intelligent self-fault diagnosis method based on knowledge inference is presented for spacecrafts in this paper. Three key techniques, knowledge base structure with direct linear storage, active monitoring-based fact updating and triggering reasoning, and hash algorithm based fact management are adopted. These key techniques are fit for spacecraft＇ s hardware with limited processing capability and memory capacity in space application. In this paper, the fault diagnosis method is implemented by using C programming language in a TSC695 processor-based embedded system. Experimental result shows the validity of the method proposed.

关 键 词：故障诊断 知识推理 自主 人工智能 空间应用 

分 类 号：TP306.3[自动化与计算机技术—计算机系统结构]