提出一种适用于线性系统的故障可诊断性分析方法,并将其应用于卫星姿态确定系统,从而实现在地面设计阶段提高系统故障诊断能力的目标。为有效解决K-L散度应用于卫星姿态确定系统故障可诊断性分析中存在的问题,基于方向相似度原理设计新的可检测性和可隔离性量化分析方法。通过数学仿真,验证了该方法的正确性和有效性。
This paper proposes a method for fault diagnosability analysis of linear systems. The method is applied to satellite attitude determination systems and can improve the FDI ( Fault Detection and Isolation) performance during the system design process. In order to solve the problems in quantifying fault diagnosability of satellite attitude determination systems based on K - L divergence, the principle of directional simi- larity are adopted to design a new approach to analyze quantitatively fault detectability and isolability. Finally, the correctness and effectiveness of the proposed method is verified through a mathematical simulation.