Secure MARL Smart Grid Pricing

Overview

This project introduces a Secure Multi-Agent Reinforcement Learning (MARL) framework for smart grid pricing, where Consumers, Aggregators, and a Grid Controller collaborate to optimize energy consumption using hybrid price and incentive-based mechanisms.

Features

• Three-layer MARL architecture with Consumer, Aggregator, and Grid agents.
• Hybrid demand response combining price-based and incentive-based strategies.
• Reinforcement learning algorithms: DQN and DDPG.
• Reward optimization for cost, comfort, and grid stability.
• Performance evaluation on peak load reduction, cost efficiency, and fairness.
• Scalable and adaptive system robust to non-participation and edge conditions.
• Security with ChaCha20‑Poly1305 - a modern, fast, and authenticated encryption scheme.

Tech Stack

Category	Tools / Technologies
Programming Language	Python 3.10
Frameworks / Libraries	PyTorch – Deep learning and reinforcement learning models CityLearn – Urban building energy simulation NumPy, Pandas – Numerical computation and data processing scikit-learn – Data preprocessing and utilities Gym – Reinforcement learning environment interface tqdm – Progress visualization cryptography - Security
Tools & Environment	Virtual Environment (venv) Git & GitHub for version control Jupyter Notebook (optional, for analysis & visualization)

Installation Guide

Use Python 3.10 for best compatibility.

1) Clone the Repository

git clone https://github.com/yourusername/marl-smartgrid-pricing.git
cd marl-smartgrid-pricing

2) Create Python 3.10 Virtual Environment

Windows (PowerShell)

python -m venv marl_env310
marl_env310\Scripts\activate

macOS/Linux

python3 -m venv marl_env310
source marl_env310/bin/activate

3) Install Dependencies

Note:

• After activating the virtual environment, upgrade pip before installing dependencies:

  pip install --upgrade pip

• Install all the requirements:

  pip install -r requirements.txt

If you don’t have a requirements.txt file, create it with the following content:

citylearn==1.5.0
numpy==1.21.6
pandas==1.3.5
torch==1.12.0
scikit-learn==1.0.2
gym==0.25.1
tqdm==4.66.1
cryptography==41.0.3
matplotlib==3.5.3

Usage

Run the main script:

python main.py

Code of Conduct

Please read our Code of Conduct before contributing to this project.

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

Security

If you discover a vulnerability, please refer to our Security Policy for instructions on how to report it responsibly.

License

This project is licensed under the MIT LICENSE.

Contributors

• Aakaash M S
• S Karthik Ram
• Abishek K
• Ashwin T