Across three micro-grid storage tasks, optimal-control baselines beat model-free RL in most settings, while PPO and TD3 match the baseline for lossy storage.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Comparing Traditional and Reinforcement-Learning Methods for Energy Storage Control
Across three micro-grid storage tasks, optimal-control baselines beat model-free RL in most settings, while PPO and TD3 match the baseline for lossy storage.