Welcome to Solar Sense! Join India's renewable energy revolution today and explore our cutting-edge prediction tool to discover how much clean energy your location can generate throughout the year.
This advanced platform uses machine learning and real-time weather data to provide accurate, location-specific forecasts, making solar energy planning simple and efficient.
We are here to revolutionize solar energy planning in India.
To provide accurate, location-specific predictions of solar power generation using cutting-edge machine learning and weather data analysis.
- Provide precise solar generation forecasts to optimize energy planning.
- Integrate real-time weather data with robust machine learning models.
- Create an intuitive interface for both professionals and homeowners.
- Support India's transition to renewable energy sources.
| Icon | Feature | Description |
|---|---|---|
| 📍 | Precision Mapping | Accurate location selection with a clickable map of India that fetches detailed Latitude and Longitude coordinates. |
| 🌤️ | Real-time Weather | Integration of live weather data (Temperature, Solar Radiation, Humidity, etc.) from Visual Crossing API. |
| 🧠 | Advanced ML | A sophisticated Support Vector Regression (SVR) model, trained on historical data, provides highly accurate predictions. |
| 💻 | User-friendly UI | Intuitive design built with React + Vite JS for a seamless experience. |
| Component | Technology | Role |
|---|---|---|
| Frontend | React + Vite JS | The fast, modern, and interactive user interface. |
| Backend | Django (Python) | Handles API requests, coordinates data flow, and executes the ML model. |
| ML Model | Python (SVR) | The core prediction engine, stored as a pickle file. |
| API Endpoint | ngrok | Used to temporarily host the backend ML prediction API as a public endpoint. |
| Data Source | Visual Crossing API | Provides real-time weather parameters. |
Generating a prediction is easy and takes just a few clicks!
Click anywhere on the map of India. The system instantly detects the coordinates and populates the Location, Latitude, and Longitude fields.
The system automatically fetches and displays up-to-the-second weather data for your selected location:
- 🌡️ Temperature (e.g.,
$0^{\circ}C$ ) - 💧 Humidity (e.g.,
$0%$ ) - 🌧️ Precipitation (e.g.,
$0\ mm$ ) - ☀️ Solar Radiation (e.g.,
$0\ W/m^2$ ) - ⚡ Solar Energy (e.g.,
$0\ kWh/m^2$ ) - ☁️ Cloud Cover (e.g.,
$0%$ ) - 💦 Dew Point (e.g.,
$0^{\circ}C$ )
Click the green "Predict Solar Generation" button to feed the real-time weather data into our advanced SVR machine learning model.
The predicted value will appear in the Predicted Solar Power (kWh) field, providing you with a precise energy forecast.
For running or deploying this project, please ensure you handle the API keys and endpoints correctly.
The application must have a valid API key to fetch weather data.
- Action: Obtain a key from visualcrossing.com and make sure to replace the placeholder/default key in your backend code (where the API is called) with your own valid key.
The frontend calls the backend API to get the prediction.
- Setup: The current setup uses ngrok to create a temporary public URL for the Django backend hosting the SVR model.
- Crucial Step: Since ngrok URLs expire and change, if you are running this project locally, you must:
- Start your Django backend.
- Start ngrok to expose your backend port (e.g.,
ngrok http 8000). - Update the API endpoint URL in your React frontend code to use the current, public ngrok URL so the website can communicate with the ML model.
- Production Note: For a permanent, production deployment, replace the ngrok setup with a stable public cloud hosting solution.