Skip to content

Latest commit

Β 

History

4 Commits

Folders and files

NameName
Last commit message
Last commit date
Β 
Β 
Β 
Β 

Repository files navigation

🌌 SolarGuard-X

Real-time Halo CME (Coronal Mass Ejection) Detection & Prediction ISRO Hack2Vision 2025

πŸ“– Overview

SolarGuard-X is a machine learning–driven solution to identify Halo CME events based on particle data from the SWIS-ASPEX Payload onboard Aditya-L1. Our system provides real-time detection, precursor analysis, and visualization to support heliophysics research and enhance space weather forecasting.

πŸš€ Key Features

⚑ Real-time CME Precursor Detection using contrastive learning.

πŸ“Š Streaming Data Pipeline for handling continuous solar wind & particle flux data.

🌍 Heliospheric Vector Field Modeling to analyze CME trajectories.

πŸ“ˆ Interactive Dashboard for live monitoring, anomaly detection, and historical comparisons.

πŸ›°οΈ Integration with ISRO’s Aditya-L1 SWIS-ASPEX Data (flux, density, speed, temperature).

πŸ“‚ Dataset

Primary Data: SWIS Level-2 particle data (Flux, Number Density, Temperature, Speed).

Secondary Data: CACTus CME Catalog (for validation & comparison).

Preprocessing: Cleaning, time-series alignment, normalization.

πŸ—οΈ Architecture

flowchart TD A[SWIS-ASPEX Data Stream] --> B[Preprocessing & Feature Engineering] B --> C[Contrastive Learning Model] C --> D[Real-time CME Detection] D --> E[Heliospheric Vector Modeling] E --> F[Visualization Dashboard]

πŸ› οΈ Tech Stack

Languages: Python, C++

ML/DL: PyTorch, Scikit-learn

Data Pipeline: Apache Kafka, Spark Streaming

Visualization: Plotly, Dash / React + Recharts

Deployment: Docker, FastAPI, Streamlit

πŸ“Š Dashboard Features

πŸ” Live CME precursor detection

⏳ Time-series analysis of solar wind data

🌐 Vector field maps for CME propagation

πŸ“œ Historical data comparison with CACTus catalog

πŸ“Œ Use Cases

πŸ›°οΈ Space Weather Forecasting: Early detection of hazardous solar events.

⚑ Satellite Safety: Protect communication & navigation satellites.

πŸ›‘οΈ Astronaut Protection: Radiation risk assessment.

🌍 Power Grid Management: Mitigate geomagnetic storm impacts.

πŸ“… Roadmap

βœ… Dataset preprocessing & validation

βœ… ML model prototyping (contrastive learning)

πŸ”„ Real-time streaming integration

πŸ”„ Dashboard deployment

πŸ“œ License

This project is developed exclusively for ISRO Hack2Vision 2025 and is intended for research & demonstration purposes.

✨ SolarGuard-X: Guarding Earth from Solar Storms πŸŒβ˜€οΈ

About

Developed an ML-based model to detect and classify solar events, specifically Halo Coronal Mass Ejections (CMEs), using particle data from the SWIS-ASPEX payload onboard Aditya-L1.

Resources

Stars

1 star

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages