AI-powered energy analytics platform for predicting household electricity consumption and solar energy generation potential — empowering smarter, sustainable energy decisions for Indian households.
- Electricity Consumption Prediction — Enter appliance counts, usage hours, and tariff rate to get monthly kWh & bill estimates (XGBoost model with formula-based fallback for low-usage households)
- Solar Generation Prediction — Input location coordinates, cloud coverage, system size, cost per kW, and geographic zone for annual solar output estimates (Gradient Boosting model)
- Auto-Detect Location — One-click geolocation + Open-Meteo API for automatic latitude, longitude, and cloud coverage fill
- Insights Dashboard — Financial returns (annual savings, payback period) and environmental impact (CO₂ reduction, trees equivalent)
- ROI / Payback Calculator — Year-by-year investment breakdown with inflation, degradation, and maintenance factored in
- User Authentication — JWT-based signup/login with personalized prediction history stored in SQLite
- Dark / Light Theme — Toggle between themes with preference persistence
- Animated UI — Floating particles, gradient orbs, animated counters, toast notifications, skeleton loaders
| Layer | Technology |
|---|---|
| Frontend | React 19, React Router 6, React Icons (Feather + Tabler) |
| Backend | Flask (Python), Flask-CORS |
| ML Models | XGBoost (electricity), scikit-learn Gradient Boosting (solar) |
| Database | SQLite (users + prediction history) |
| Auth | JWT (PyJWT) + bcrypt password hashing |
| APIs | Open-Meteo (weather data for auto-detect) |
React-App/
├── frontend/
│ ├── src/
│ │ ├── components/ # Reusable UI components
│ │ │ ├── common/ # InputField, ResultCard, Toast, Skeleton, AnimatedBackground
│ │ │ ├── layout/ # Navbar
│ │ │ └── index.js # Barrel exports
│ │ ├── context/ # React contexts
│ │ │ ├── AppContext.js # Shared state (solar results, tariff, toasts)
│ │ │ ├── AuthContext.js # Auth state (user, token, login/signup/logout)
│ │ │ └── ThemeContext.js# Dark/light theme toggle
│ │ ├── pages/ # Route pages
│ │ │ ├── Home.jsx
│ │ │ ├── Electricity.jsx
│ │ │ ├── Solar.jsx
│ │ │ ├── Insights.jsx
│ │ │ ├── Dashboard.jsx
│ │ │ ├── RoiCalculator.jsx
│ │ │ ├── Login.jsx
│ │ │ └── Signup.jsx
│ │ ├── App.js # Root component with routes
│ │ └── App.css # All styles (CSS variables, glass morphism)
│ └── package.json
├── backend/
│ ├── app.py # Flask server (API + auth + ML inference)
│ ├── electricity_prediction_model.joblib
│ ├── solar_prediction_model.joblib
│ ├── solar_scaler.joblib
│ ├── suryaveda.db # SQLite database (auto-created)
│ └── requirements.txt
├── Datasets/
│ ├── electricity_bill_dataset.csv
│ ├── Solar_dataset.csv
│ └── Solar_prediction.csv
├── SAMPLE_INPUTS.md # Test scenarios with real-world inputs
└── README.md
- Node.js ≥ 18
- Python ≥ 3.10
cd backend
python -m venv .venv
# Windows
.\.venv\Scripts\Activate.ps1
# macOS/Linux
source .venv/bin/activate
pip install -r requirements.txt
cp .env.example .env
python app.pyBackend runs at http://127.0.0.1:5000
Set a strong SECRET_KEY in backend/.env before deploying.
cd frontend
npm install
cp .env.example .env
npm startFrontend runs at http://localhost:3000
| Method | Endpoint | Auth | Description |
|---|---|---|---|
| POST | /api/signup |
No | Create account |
| POST | /api/login |
No | Get JWT token |
| GET | /api/me |
Yes | Get current user info |
| GET | /api/predictions |
Yes | Get prediction history |
| POST | /predict_electricity |
Optional | Run electricity prediction |
| POST | /predict_solar |
Optional | Run solar prediction |
- Sign up for an account at
/signup - Navigate to Electricity — enter your appliance details and get consumption prediction
- Navigate to Solar — use Auto-Detect or manually enter location and system details, then get a full-system solar generation estimate
- View Insights for combined financial and environmental metrics
- Use ROI Calculator to model long-term investment returns
- Check Dashboard to review all your past predictions
 |
 |
 |
This project is an enhanced version of SuryaVeda, an AI-powered energy analytics platform that I originally developed as part of the Edunet Energy program.
Building on the original implementation, this version was developed collaboratively to transform the project into a more feature-rich and production-ready application. Key enhancements include:
- Modernized React-based user interface with improved user experience
- JWT-based authentication and personalized user dashboards
- Prediction history management with SQLite
- Enhanced machine learning workflows for electricity and solar predictions
- Financial and environmental insights, including ROI and payback analysis
- Automatic location detection using the Open-Meteo API
- Improved backend architecture, API design, and project organization
- Better deployment support, documentation, and maintainability
The original repository serves as the foundation for this project, while the current version significantly expands its functionality and overall user experience.