Energy grid state estimation under random and structured bad data

The problems of state recovery and bad data detection in energy grids, while being strongly interconnected, have been treated independently. Furthermore, while state recovery has been studied intensively, it has been less well studied when the measurements are deemed to be contaminated by random bad data (due to sensor failures) or structured bad data (due cyber attacks). This paper provides a unifying framework that takes into account the inherent connection between state recovery and bad data detection in order to accomplish the combined tasks of detecting the presence of random and structured bad data, and providing reliable estimates for the state of the grid and injected bad data. Optimal detectors and estimators are characterized.