Fault detection in large-scale stochastic dynamic systems

A new approach is developed for fault detection in linear, stochastic, interconnected dynamic systems, based on designing a set of decentralised Kalman filters for the subsystems resulting from overlapping decomposition of the overall large-scale system. The malfunctioning sensors can be detected and isolated by comparing the estimated values of a single state from different Kalman filters. The new method is applied to a sixth-order system represented by a ladder network.