# Reinforcement Learning on OPF with Battery Energy Storage

Source: [Reinforcement Learning on OPF with Battery Energy Storage](<https://www.sota2.com/research/sota/reinforcement-learning-on-opf-with-battery-energy-storage>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:42:07.576Z

Evaluation of Reinforcement Learning agents on the OPF (Optimal Power Flow) environment specifically tailored for Battery Energy Storage systems.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Reinforcement Learning |
| Dataset (including any stated split) | OPF with Battery Energy Storage |
| Metric | Episode Reward |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Episode Reward](<https://www.sota2.com/research/sota/reinforcement-learning-on-opf-with-battery-energy-storage.md?metric=Episode%20Reward>): Higher is better; unit: SCALAR.

- [Max In-Episode Inequality Constraint Violation](<https://www.sota2.com/research/sota/reinforcement-learning-on-opf-with-battery-energy-storage.md?metric=Max%20In-Episode%20Inequality%20Constraint%20Violation>): Lower is better; unit: SCALAR.

- [Max In-Episode Equality Constraint Violation](<https://www.sota2.com/research/sota/reinforcement-learning-on-opf-with-battery-energy-storage.md?metric=Max%20In-Episode%20Equality%20Constraint%20Violation>): Lower is better; unit: SCALAR.

- [Max Episode Inequality Constraint Violation](<https://www.sota2.com/research/sota/reinforcement-learning-on-opf-with-battery-energy-storage.md?metric=Max%20Episode%20Inequality%20Constraint%20Violation>): Lower is better; unit: SCALAR.

- [Max Episode Equality Constraint Violation](<https://www.sota2.com/research/sota/reinforcement-learning-on-opf-with-battery-energy-storage.md?metric=Max%20Episode%20Equality%20Constraint%20Violation>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Episode Reward | Result date | Report paper |
| --- | --- | --- | --- |
| RPO-DDPG | 42.971 | Not recorded | 68fa71a91e209131cea5263a |
| RPO-SAC | 42.0764 | Not recorded | 68fa71a91e209131cea5263a |
| CUP | -5.8377 | Not recorded | 68fa71a91e209131cea5263a |
| CPO | -10.3469 | Not recorded | 68fa71a91e209131cea5263a |
| SAC-L | -19.4738 | Not recorded | 68fa71a91e209131cea5263a |
| DDPG-L | -22.0649 | Not recorded | 68fa71a91e209131cea5263a |
| Safety Layer | -24.1147 | Not recorded | 68fa71a91e209131cea5263a |

### RPO-DDPG

Episode Reward: 42.971. Result ID: 6991d9a23a6c3148490afa9e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Implementation](<https://github.com/wadx2019/rpo>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### RPO-SAC

Episode Reward: 42.0764. Result ID: 6991d9a23a6c3148490afa9f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Implementation](<https://github.com/wadx2019/rpo>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CUP

Episode Reward: -5.8377. Result ID: 6991d9a23a6c3148490afa9a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6dd91e209131cea3c676.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### CPO

Episode Reward: -10.3469. Result ID: 6991d9a23a6c3148490afa99.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63691e209131ce9fe8ff.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### SAC-L

Episode Reward: -19.4738. Result ID: 6991d9a23a6c3148490afa9d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Implementation](<https://github.com/wadx2019/rpo>)

Recorded evaluation configuration: {"type":"abridged version"}. This record does not assert independent reproduction.

### DDPG-L

Episode Reward: -22.0649. Result ID: 6991d9a23a6c3148490afa9c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Implementation](<https://github.com/wadx2019/rpo>)

Recorded evaluation configuration: {"type":"abridged version"}. This record does not assert independent reproduction.

### Safety Layer

Episode Reward: -24.1147. Result ID: 6991d9a23a6c3148490afa9b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa71a91e209131cea5263a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63f31e209131cea01be6.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.