基于智能与融合的静止轨道卫星容错方法  

Fault-Tolerant Computing for GEO Satellite Based on Intelligence and Fusion

作　　者：韩笑冬[1] 于慧[1] 张登峰 王志杰[1] HAN Xiaodong YU Hui ZHANG Dengfeng WANG Zhijie(Institute of Telecommunication Satellite, CAST, Beijing 100094, China School of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China)

机构地区：[1]中国空间技术研究院通信卫星事业部,北京100094 [2]南京理工大学机械学院,江苏南京210094

出　　处：《系统仿真技术》2017年第3期215-219,共5页System Simulation Technology

基　　金：国家自然科学基金(61374133;61471360)

摘　　要：基于计算智能和融合方法设计了一种静止轨道卫星容错结构,并提出了一个由模糊神经网络和专家系统模型构成的智能容错算法,该算法利用双滤波器对卫星状态进行估计和预测。首先对卫星测量信息的有效性进行判断,然后通过计算智能模型计算各敏感器的置信度,最后根据卫星在轨工作健康状态选择相应融合方法。仿真算例验证了该算法的有效性。The fault-tolerant module of geosynchronous orbit（ GEO） satellites was designed based on computational intelligence and information fusion. Then,an intelligent fusion fault-tolerant scheme selection algorithmwas proposed based on fuzzy neural network and expert systemmodel. The algorithmcan estimate and predict the state of satellite using dual-filter,judge the validity of satellite measurement information,compute the confidence of attitude determination systemby model,and select the matching fusion scheme according to actual in-orbit situation. Numerical examples are given to illustrate the validity of the algorithm.

关 键 词：静止轨道卫星 容错方法 信息融合 计算智能 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]