NSSC OpenIR  > 空间技术部
基于遥测数据相关性的航天器异常检测
Alternative TitleAnomaly detection of spacecraft based on the telemetry data correlation
杨甲森; 孟新; 陈托; 智佳; 李虎
Department空间技术部
Source Publication仪器仪表学报
2018
Volume39Issue:8Pages:24-33
ISSN0254-3087
Language中文
Keyword遥测数据 相关性 知识发现 异常检测 数据驱动
Abstract

基于信息技术进行星载设备的异常检测始终是空间任务安全性、可靠性领域最重要的主题。传统航天器异常检测方法的知识和模型很难自动化构建,而纯数据驱动方法会忽略系统设计知识和遥测数据本身的特性,造成先验知识的浪费。针对这些问题,提出一种基于遥测数据相关性的航天器异常检测方法。首先,给出遥测数据相关性的定义,结合量子卫星任务实例,对单维度遥测数据的6种领域相关性进行建模和向量表示;随后,提出一种遥测数据驱动进行相关性知识发现、进而将知识应用于异常检测的方法;最后,利用量子卫星数据、基于真实故障特征的模拟数据对该方法进行了验证。实验与分析结果表明:该方法能够实现航天器异常检测知识构建的自动化,且可以有效检测出遥测数据的异常。

Other Abstract

Anomaly detection on satellite-carrying equipment has always been the most challenging task in the field of space task security and reliability.Currently,there are two available methods which are knowledge-based model method and data-driven method.The first approach is crippled by the inherent flaw-the difficulty to automate knowledge-based model construction.The second discriminates the design basis and data characteristics which makes the task inefficient.In this study,the definition of telemetry data correlation is firstly given.Six types of domain correlation for single dimension telemetry data are modeled and represented by the vector combined with quantum satellite examples.Then,an anomaly detection method based on the correlation knowledge is proposed,which can be acquired by telemetry data driven method automatically.Experimental results demonstrate that this method can automatically generate knowledgebased model that characterizes the design basis of the spacecraft and the anomaly in telemetry data can be effectively detected.The proposed method is evaluated by comparing the task data of quantum satellite and simulated data based on characteristics of real faults.

Indexed ByEI ; CSCD
Citation statistics
Document Type期刊论文
Identifierhttp://ir.nssc.ac.cn/handle/122/6598
Collection空间技术部
Recommended Citation
GB/T 7714
杨甲森,孟新,陈托,等. 基于遥测数据相关性的航天器异常检测[J]. 仪器仪表学报,2018,39(8):24-33.
APA 杨甲森,孟新,陈托,智佳,&李虎.(2018).基于遥测数据相关性的航天器异常检测.仪器仪表学报,39(8),24-33.
MLA 杨甲森,et al."基于遥测数据相关性的航天器异常检测".仪器仪表学报 39.8(2018):24-33.
Files in This Item:
File Name/Size DocType Version Access License
201839824-33.pdf(655KB)期刊论文出版稿开放获取CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[杨甲森]'s Articles
[孟新]'s Articles
[陈托]'s Articles
Baidu academic
Similar articles in Baidu academic
[杨甲森]'s Articles
[孟新]'s Articles
[陈托]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[杨甲森]'s Articles
[孟新]'s Articles
[陈托]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.