RL-ADN 0.1.3

RL-ADN: A Benchmark Framework for DRL-based Battery Energy Arbitrage in Distribution Networks

RL-ADN: A Benchmark Framework for DRL-based Battery Energy Arbitrage in Distribution Networks

RL-ADN is the first open-source framework designed to revolutionize DRL-based battery energy arbitrage in distribution networks. By abstracting distribution network dynamics and offering a modular structure, RL-ADN paves the way for modeling diverse energy arbitrage tasks.

Features

• Versatile Benchmarking: Model diverse energy arbitrage tasks with full flexibility.
• Laurent Power Flow: Over 10 times faster computational speed compared to traditional methods.
• Seamless Transition: Designed for both simulated environments and real-world applications.
• Open-source: Easily accessible for modifications, customizations, and further research.

Outline

• Overview
• Documentation
• File Structure
• Installation
• Tutorials
• Publications
• Citing FinRL
• LICENSE

documentation

under developing

File Structure

The main folder RL-ADN is shown below

└─power_network_rl
    │  README.md
    │  requirements.txt
    │  setup.py
    │  __init__.py
    │
    ├─benckmark_algorithms
    │      Optimality_pyomo.py
    │      __init__.py
    │
    ├─data_manager
    │  │  data_manager.py
    │  │  __init__.py
    │
    ├─data_sources
    │  ├─network_data
    │  │  │  __init__.py
    │  │  │
    │  │  ├─node_123
    │  │  │      Lines_123.csv
    │  │  │      Nodes_123.csv
    │  │  │
    │  │  ├─node_25
    │  │  │      Lines_25.csv
    │  │  │      Nodes_25.csv
    │  │  │
    │  │  ├─node_34
    │  │  │      Lines_34.csv
    │  │  │      Nodes_34.csv
    │  │  │
    │  │  └─node_69
    │  │          Lines_69.csv
    │  │          Nodes_69.csv
    │  │
    │  └─time_series_data
    │          123_node_time_series.csv
    │          25_node_time_series.csv
    │          34_node_time_series.csv
    │          69_node_time_series.csv
    │
    ├─docs
    ├─DRL_algorithms
    │  │  Agent.py
    │  │  DDPG.py
    │  │  PPO.py
    │  │  SAC.py
    │  │  TD3.py
    │  │  utility.py
    │  │  __init__.py
    │
    │
    ├─environments
    │  │  Component_Battery.py
    │  │  Environment_Integrated.py
    │  │  env_config.json
    │  │  gym_env_create_t.py
    │  │  __init__.py
    │
    ├─example
    │      customize_env.py
    │      training_DDPG.py
    │
    ├─tests
    │      123_node_network_powerflow_test.py
    │      25_node_network_powerflow_test.py
    │      69_node_network_powerflow_test.py
    │      test_comparison_power_flow.py
    │
    ├─utility
    │  │  gpu_interface.py
    │  │  grid.py
    │  │  Not_converge_Power_Flow.py
    │  │  numbarize.py
    │  │  Power_Flow.py
    │  │  utils.py
    │  │  __init__.py
    │  │

Installation

To install RL-ADN, simply run:

pip install RL-ADN

Or install from git

git clone https://github.com/shengrenhou/RL-ADN.git
cd your repository
pip install -e .

Or install from local, if you have download the source code

cd to the path contains setup.py
python setup.py install

Status Update

Version History [click to expand]

• 2023-09-27 0.1: Beta version

Tutorials

In example folder, training_DDPG.ipynb shows a tutorial for training DDPG agent using RL-ADN step by step. customize_env.py shows a simple tutorial for users customize their own environment by using RL-ADN

Future work

Develop a data augmentation functin by implementing a data generator for time-series data used for training algorithms.

Publications

Citing RL-ADN

Contributing

LICENSE

MIT License

Disclaimer: We are sharing codes for academic purpose under the MIT education license. Nothing herein is financial advice, and NOT a recommendation to trade real money. Please use common sense and always first consult a professional before trading or investing.