基于扩展卡尔曼滤波器的航天器姿控飞轮故障诊断  

The Fault Diagnosis of Spacecraft Attitude Control Flywheel Based on Extended Kalman Filter

作　　者：吴红芳[1] 王志[2] 

机构地区：[1]中国空空导弹研究院,洛阳471009 [2]西安工业大学计算机学院,西安710032

出　　处：《航天控制》2012年第3期34-37,共4页Aerospace Control

摘　　要：将未知输入卡尔曼滤波器扩展到非线性系统,并将其应用于检测航天器飞轮出现的早期渐变型故障,以提高故障检测的及时性,避免重大事故的发生。通过构造一组结构化残差实现故障分离,采用Wald序列检测方法处理滤波残差,最后给出故障判断逻辑。仿真结果验证了结论的有效性。The unknown input Kalman filter is extended to nonlinear systems and then is applied to the detection of the early gradual faults of spacecraft attitude control flywheels to improve the timeliness of fault detection in order to avoid the occurrence of major accidents. A set of structured residuals is constructed to achieve fault isolation, and the Wald sequence detection method is employed to process filter residuals. Finally, the fault diagnosis logic is presented, and the numerical simulation results show that this method is effective and is able to quickly detect faults.

关 键 词：姿控飞轮 故障诊断 卡尔曼滤波 Wald序列 

分 类 号：TP277.2[自动化与计算机技术—检测技术与自动化装置]