Collapse prediction on transmission networks

Predicting and preventing power system collapses is crucial for the economic prosperity of the world (Van Cutsem, 2000). Highly nonlinear power systems must be analysed in real time and if a system change might start a process that leads towards a voltage collapse then the operator should be alerted and corrective control initiated. Such changes might include a sudden increase in loading, the tripping of a generator, the loss of a synchronous condenser, reactive power limitations of static VAr control (SVC) systems, tap-changing on a transformer or the disconnection of a transmission feeder. The difficulty is trying to decide if such a change is a routine event that does not affect the integrity of the power system or a rare event that instigates a collapse. The possibility of predicting a collapse using a collapse prediction relay (Eberle, 2006) that compares and analyses the characteristic frequencies seen in the voltage signals during the precollapse period is described.