sukanyamandal/fuzzy-dollop

Smart Grid DApp

Overview

This decentralized application (DApp) leverages the power of federated learning and blockchain technology to enable secure and collaborative data analysis for smart grids.

Key Features

• Federated Learning: Train a shared machine learning model across multiple smart grids without directly sharing raw data.
• Differential Privacy: Enhance data privacy during model training using PySyft's DPAdam optimizer.
• Homomorphic Encryption: Explore secure data aggregation techniques using homomorphic encryption.
• Blockchain Integration: Record model updates and data usage agreements on the Energy Web Chain for transparency and auditability.

Technologies Used

• PySyft: A framework for secure and private federated learning.
• PyTorch: A deep learning library for model development.
• Flask: A Python web framework for building the backend API.
• Web3.py: A library for interacting with the Energy Web Chain.
• Solidity: The programming language for writing the smart contract.
• Truffle: A development environment for Ethereum and EVM-compatible blockchains.
• Ganache: A local blockchain emulator for development and testing.

Project Structure

📦 fuzzy-dollop
├─ Dockerfile
├─ LICENSE
├─ app.py
├─ contracts
│  └─ DataSharing.sol
├─ docker-compose.yml
├─ federated_learning.py
├─ migrations
│  ├─ 1_initial_migration.js
│  └─ 2_deploy_contracts.js
├─ models
│  ├─ handler
│  │  └─ custom_handler.py
│  └─ model.py
├─ readme.md
├─ requirements.txt
├─ setup
│  ├─ setup_node.sh
│  ├─ setup_python.sh
│  └─ setup_torchserve.sh
├─ static
│  ├─ css
│  │  └─ styles.css
│  └─ js
│     └─ scripts.js
├─ templates
│  └─ index.html
├─ tests
│  ├─ test_blockchain.py
│  └─ test_prediction.py
└─ utils
   ├─ blockchain.py
   └─ prediction.py

Getting Started

Prerequisites

• Node.js: https://nodejs.org/
• Python 3.8+: https://www.python.org/
• Docker: https://www.docker.com/
• Energy Web Chain Account: https://www.energyweb.org/

Installation

1. Clone the Repository:

   git clone https://github.com/your-username/smart-grid-dapp.git
   cd smart-grid-dapp
   

2. Install Dependencies:

Install Node.js dependencies

bash setup/setup_node.sh

Create a Python virtual environment and activate it

python3 -m venv venv source venv/bin/activate

Install Python dependencies

pip install -r requirements.txt

Install TorchServe

bash setup/setup_torchserve.sh

Configuration

Blockchain Configuration:

1. Open truffle-config.js.
2. Configure the networks section to connect to the Energy Web Chain.
3. Provide your EWC account credentials and an API key for an EWC node provider (if necessary).

Private Key:

Important: Replace "YOUR_PRIVATE_KEY" in app.py and test_blockchain.py with a valid private key for interacting with the blockchain. For security reasons, use a test account and a testnet for development.

Running the Application

1. Start Ganache (for local development):

   ganache-cli

2. Deploy the Smart Contract:

   truffle compile
   truffle migrate --network ewc 

3. Train the Model:

   python federated_learning.py

4. Archive and Serve the Model:

   torch-model-archiver --model-name smartgridmodel --version 1.0 \
   --model-file models/model.py --serialized-file global_model.pth \
   --handler models/handler/custom_handler.py --export-path model_store
   
   torchserve --start --ncs --model-store model_store --models smartgridmodel=smartgridmodel.mar

5. Run the Flask Application:

   python app.py

6. Access the UI: Open the provided URL (from Codespaces or your local server) in your web browser.

Running Tests

• Run All Tests:

  python -m unittest discover

• Run Specific Tests:

  python -m unittest tests/test_blockchain.py
  python -m unittest tests/test_prediction.py

Future Work

• Integrate real smart grid data sources.
• Develop a more sophisticated user interface.
• Implement additional blockchain functionalities (e.g., data usage agreements, tokenized rewards).
• Explore more advanced privacy-preserving techniques.
• Deploy to a production environment.

Contributing

Contributions are welcome! Please fork the repository and submit pull requests.

License

This project is licensed under the MIT License.