基于集员估计的卫星飞轮故障检测与预测方法  

Fault detection and prognosis approach for satellite flywheel based on set-membership estimation

作　　者：梁寒玉 张文瀚 王振华[2] 沈毅[2] LIANG Hanyu;ZHANG Wenhan;WANG Zhenhua;SHEN Yi(The Beijing Institute of Control Engineering,Beijing 100190,China;School of Astronautics,Harbin Institute of Technology,Harbin 150001,China)

机构地区：[1]北京控制工程研究所,北京100190 [2]哈尔滨工业大学航天学院,哈尔滨150001

出　　处：《航空学报》2022年第3期497-508,共12页Acta Aeronautica et Astronautica Sinica

基　　金：国家自然科学基金(61773145,61973098)。

摘　　要：为了对卫星飞轮故障进行提前预警,提出了一种基于集员估计的故障检测和预测方法。首先,基于飞轮的动态模型设计了一种基于中心对称多面体的故障检测方法。在检测出飞轮故障后,选取飞轮的电流数据作为故障特征,并采用线性模型或非线性模型来描述其退化规律。然后,提出了一种基于中心对称多面体的集员算法,对退化模型中的未知参数进行估计,并以此为基础预测未来时刻的故障,进而实现对飞轮故障的提前预警。最后,通过数值仿真,将所提出的方法和两种现有的故障预测技术进行了对比。仿真结果表明了所提方法的有效性和优越性。A fault detection and prognosis approach is proposed based on set-membership estimation to achieve early warning of the flywheel fault.Firstly,a zonotope-based fault detection method is designed based on the dynamic model of flywheel.After the flywheel fault is detected,the electric current of flywheel is selected as the fault feature,and the linear model or nonlinear model is used to describe its degradation law.Then,a set-membership algorithm via zonotope is proposed to estimate the unknown parameters of the degradation model.By using the parameter estimation results,fault magnitudes in the future instants are predicted to achieve early warning of the flywheel fault.Finally,the proposed method is compared with two existing fault prognosis techniques through numerical simulation.The simulation results demonstrate the viability and validity of the proposed approach.

关 键 词：卫星飞轮 故障检测 故障预测 集员估计 中心对称多面体 

分 类 号：V448[航空宇航科学与技术—飞行器设计] TP277[自动化与计算机技术—检测技术与自动化装置]