Power System Topology Control on CASE118 modified IEEE bus (evaluation)
137,420.6Average RewardLJN (topology-only)
Evaluation results
| Method | Links | ||
|---|---|---|---|
| LJN (topology-only)Decision time (ms/step)=81.4952026.04 | 137,420.6 | 717.6 | |
| Physics-Informed Reinforcement Learning with Gibbs PriorsDecision time (ms/step)=32.4052026.04 | 84,643.5 | 435.4 | |
| PPODecision time (ms/step)=18.4772026.04 | 23,830.9 | 113.4 | |
| RandomDecision time (ms/step)=0.0062026.04 | 1,038.1 | 13.2 |