人工神经网络在卫星姿态测量系统故障诊断中的应用研究  

Application study of ANN for Fault Diagnosis in the Satellite Attitude Measurement System

作　　者：田蔚风[1] 金志华[1] 陆恺[1] 

机构地区：[1]上海交通大学信息检测技术与仪器系,上海200030

出　　处：《航天控制》1999年第4期63-66,共4页Aerospace Control

摘　　要：针对卫星姿态测量系统的硬件故障, 应用人工神经网络进行故障检测与诊断, 得到了较为满意的结果。研究表明, 人工神经网络结构简单, 学习速度快, 适应性广, 对于卫星姿态测量系统中的姿态传感器故障,Artificial neural networks are proposed in this paper, which are used to detect and diagnose faults of hardware for the satellite attitude measurement system. The result is satisfied. It is shown by research that the networks are simple in structure and good in speed of learning and adaptability. The ANN can be used to detect and diagnose faults of sensors in measurement system of satellite attitude, and its detection rate and correct rate of diagnosis are 100 percent for assigned faults.

关 键 词：人工神经元网络 故障诊断 卫星 姿态测量 

分 类 号：V448.22[航空宇航科学与技术—飞行器设计]