Fault Isolation in Loops of Linear Dynamic Networks

This paper addresses the problem of fault isolation in networks containing loops where the dynamic behavior of the nodes can be modeled by finite gain linear systems. If the underlying graph of the network contains loops, isolating a faulty subsystem can be challenging. This is because the propagation of fault effects through the network interconnections is difficult to trace. In this study, it is shown that, if the gains of the subsystems satisfy certain assumptions, the fault can be isolated in the loops of the considered network. For each subsystem in the loop, a residual signal is generated. The isolation method examines the differences between the norms of these residual signals. Based on this information, the location of the faulty subsystem can be determined. Simulation measurements are provided to show the efficiency of the proposed fault isolation method.