|
Reference | System | Method | Attack | Recovery action | Aim | Limitations |
|
[16] | Modified 9-bus system | Deep deterministic policy gradient (DDPG) | Multiswitch attacks and false data injection (FDI) attacks | Reclose the transmission lines lost in the cyber-attack by optimizing the reclosing time. | Reach the asynchrony in the power system by applying power blocking which will accelerate/decelerate the rotors of the generators | Owing to its continuous action space, it will not be suitable for topological resilience studies |
[17] | IEEE 9, 14 and 30-bus systems | Deep Q-network (DQN) | Data integrity attacks | No recovery action | Evaluate the delay-alarm error rates, false-alarm error rates, and detect-failure rates for the systems | DQN suffers from overestimation |
[18] | IEEE 30-bus system | Deep Q-network (DQN) | Coordinated cyber physical topology (CCPT) attacks | Control center can detect the line outage by using phasor measurement units (PMU) data | Investigate the coordinated topology attacks in smart grid which combine a physical topology attack and a cyber-topology attack |
[19] | Wood & Wollenberg 6-bus system and IEEE 30-bus system | Q-learning | Sequential attacks | Automatic generation control (AGC) | Identify the minimum number of attacks/actions to reach blackout threshold | Q-learning and SARSA techniques are limited to systems with small state-action space |
[20] | IEEE 14-bus system | SARSA | False data injection (FDI), jamming, and denial of service (DoS) attacks | No recovery action | Formulation an online cyber-attack detection as a POMDP problem and propose a solution based on the model-free RL for POMDPs |
Our work | IEEE 6-bus system | Deep Q-network (DQN), double DQN, REINFORCE, and REINFORCE with baseline | Sequential attacks | Disconnecting the faulted transmission lines | Evaluating the resiliency of power systems against faults/attacks using DRL | Needs to investigate tabular methods such as Q-learning and SARSA to compare their performance with DRL methods |
|