摘要
针对卫星姿态控制系统的执行机构故障,提出了一种基于T-S模糊方法和滑模观测器的故障诊断方法.首先,建立了基于T-S模糊模型的卫星姿态控制系统模型,并在此基础上,设计了一种用于状态检测的模糊滑模观测器.该观测器在执行机构发生故障的情况下,仍然保持滑模模态,可以利用等效输出注入实现故障的直接估计.在故障和干扰上界未知的情况下,采用迭代学习算法在线计算并调节滑模项的切换增益大小,避免了不必要的高频抖振现象;同时利用H∞思想,通过求解LMI获得观测器参数,将干扰对故障估计的影响最小化.计算机仿真结果表明,该故障诊断方法能够对执行机构故障进行较为精确的估计,从而验证了方法的有效性.
A fault diagnosis approach based on the Takagi-Sugeno( T-S) fuzzy and the sliding mode observer is proposed for actuator faults in the attitude control systems of a satellite. First,an attitude control system model of the satellite is constructed based on the T-S fuzzy and a sliding mode observer is proposed for state detection. The observer can maintain a sliding motion even in the presence of actuator faults; thus,the estimation of the faults can be directly achieved by equivalent output injection. When the upper bounds of the faults and disturbances are unknown,an iterative learning algorithm is used to calculate and update the switching gain of the sliding mode term,avoiding unnecessary chattering. And by using the concepts of H∞ ,a linear matrix inequality( LMI) is calculated to minimize the effects of disturbances on the estimation of the faults. The results of the numerical simulation show that the accurate estimation of the actuator faults can be obtained,which proves the effectiveness of this approach.
出处
《东南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2010年第S1期238-243,共6页
Journal of Southeast University：Natural Science Edition
基金
江苏省自然科学基金重点资助项目(BK2010072)
南京航空航天大学研究生创新实验室基金资助项目
南京航空航天大学大学生创新基金资助项目
关键词
卫星
姿态控制
故障诊断
T-S模糊模型
滑模观测器
satellite
attitude control
fault diagnosis
Takagi-Sugeno( T-S) fuzzy model
sliding mode observer
