Finite-Time Fault Detection Against Censored Measurements: A Novel Memory-Adaptive Event-Triggered Approach

In this article, the problem of finite-time fault detection (FD) is investigated for Takagi–Sugeno (T–S) fuzzy systems with censored measurements under limited communication bandwidth. A novel memory-adaptive event-triggered mechanism (ETM) is developed to reduce the communication burden while improving buffer-memory utilization. In particular, the triggering threshold is adaptively adjusted based on historical data within a memory time window. Notably, the memory-window length is switched according to the FD feedback. Moreover, an improved$H_{\infty }$performance index that incorporates the censoring threshold information is introduced to evaluate the attenuation performance of the FD filter against both external disturbances and censoring effects. Subsequently, sufficient conditions are derived to guarantee that the residual dynamic system is finite-time bounded with a satisfactory$H_{\infty }$performance. On this basis, the desired FD filter parameters are obtained using the recursive linear matrix inequality (LMI) method. Finally, simulation studies on Chua’s circuit and a single-link robotic manipulator are conducted to demonstrate the applicability and effectiveness of the proposed FD scheme integrated with the memory-adaptive ETM.