Real-Time Lightning Detection & Monitoring System

• Shubham Banwal

• June, 2022

AI-Enabled Lightning Detection System with 5 km Mesh Network for Early Storm Warning (IoT Platform)

Problem:

Timely detection of approaching lightning storms is critical for disaster preparedness, yet many systems lack real-time accuracy, scalable coverage, and integration with operational response frameworks.

Approach:

Designed and deployed an AI-enabled lightning detection system using a distributed mesh network with ~5 km coverage, enabling real-time storm proximity detection and early warning capabilities.

System Design:

• Electric field sensing units for atmospheric charge detection
• Mesh network architecture for distributed data transmission (~5 km range)
• Data aggregation and processing layer
• AI/ML models for pattern detection and storm proximity estimation
• IoT-enabled communication for real-time monitoring
• TFT display interface for on-device visualization

Key Contributions:

• Built and deployed a distributed lightning detection system with mesh networking for extended coverage
• Developed sensing and data pipelines to capture and process atmospheric electric field variations
• Applied AI/ML techniques to detect patterns and estimate approaching storm range
• Integrated IoT capabilities for real-time data transmission and monitoring
• Designed TFT-based interface for live visualization of system outputs
• Collaborated with State Disaster Management Authority, Lucknow for real-world application context
• Conducted field-level testing and optimization to ensure accuracy and reliability in live environments

Constraints & Tradeoffs:

• Environmental noise affecting sensor accuracy
• Network reliability across distributed mesh nodes
• Tradeoff between detection sensitivity and false positives
• Hardware and deployment limitations in field conditions