Early warning signals for critical transitions in power systems

Abstract Several episodes of sudden large scale disruptions in electrical service deeply impacted both the social stability and economic development in affected countries. The prevention of such catastrophic incidents poses huge challenges for reliability study and operational practices in power systems. Studies in other scientific fields show that, upon reaching a tipping point, complex dynamical systems can experience sudden transitions into a contrasting state. These transitions may be predicted through behavioral changes in some statistical measures of the system state. Inspired by these studies, this paper proposes an analysis of the critical transition in power systems from a long-term perspective. The evolution of the operational “stress” and its cyclical variation due to a slowly increasing demand and system expansions is simulated on a test system. The disturbances and the resulting failures under different stress levels are studied. Our analysis identifies the statistical trends known as flickering and critical slowing down in the operational and the recorded outage data along the simulation. These statistical changes can be used as early warning signals of the upcoming operational state which is more prone to a catastrophic blackout. The development of such early warning signals is the key to reaching higher levels of reliability in the energy supply infrastructure that society requires today.