A Region-based Framework for Cyberattacks Leading to Undervoltage in Smart Distribution Systems

One of the technical challenges in power systems operation is to prevent/tackle undervoltage, which can result in overheating of the equipment or even cascading failures under severe circumstances. Hence, it can be considered as an intention for cyberattackers to target smart distribution systems leading to undervoltage(s). This paper proposes a bi-level framework to attack a distribution network such that: 1) load centers and renewable-based distributed generation units are simultaneously targeted, 2) the attack bypasses the AC state estimation-based bad data detection, and most importantly, 3) the attacker targets the most vulnerable region of the system at the optimal time instead of attacking random sections/entire system at a random time. The effectiveness of the framework is validated on an IEEE standard distribution system modified to contain photovoltaic arrays.