Distributed fault detection for large‐scale interconnected systems

Abstract

AbstractThe main objective of this paper is to develop a distributed fault detection (FD) approach for large‐scale interconnected systems using sensor networks. Specifically, the one‐step prediction based on the measured data is implemented in a distributed fashion so that each node can receive corresponding estimations and innovation sequences in a real‐time manner. Then the innovation sequences are applied to improve the estimation result delivered from the one‐step prediction by filtering and smoothing. After filtering and smoothing, the residual signals are calculated to detect faults. Finally, a case study shows that the distributed approach can efficiently accomplish the FD task.