Australian technology that could prevent energy storage fires

Published: Updated: Estimated reading time: 3 minutes
niemcy-magazyn-energii (3)
Source: Shutterstock

UNSW Researchers Develop Passive Radio-Based AI System to Detect Battery Thermal Runaway. In laboratories at the University of New South Wales (UNSW), researchers are developing a technology that could fundamentally change how the energy industry approaches battery safety. Instead of relying on costly thermal sensors and complex cooling systems, the team proposes a passive monitoring solution. The concept uses existing radio signals — such as Wi-Fi or millimeter-wave radar — combined with artificial intelligence (AI) to detect early signs of overheating in lithium-ion cells.

Thermal Runaway – The Silent Enemy of Lithium-Ion Batteries

Thermal runaway is not a new phenomenon, but its importance is increasing rapidly with the global boom in energy storage systems (ESS). As battery installations scale up, so does the risk of cell overheating. Even a minor temperature anomaly can trigger a cascade of chemical reactions that, within seconds, can lead to catastrophic failure.

Traditionally, detecting such risks required expensive sensors, wiring, and continuous maintenance. The UNSW team, however, offers a different approach — using what is already present: Wi-Fi networks and millimeter-wave radars.

How Does the Contactless Early-Warning System Work?

The system developed at UNSW operates on a simple yet ingenious principle. Radio waves change their properties in response to environmental variations. While these subtle changes are imperceptible to the human eye or standard monitoring systems, they can be effectively analyzed by advanced AI algorithms.

These algorithms can detect abnormal thermal patterns with up to 97% accuracy, without any physical contact with the battery — the system only needs to be nearby.

“By combining state-of-the-art AI algorithms with low-cost, passive radio monitoring, we are creating a solution that is technically advanced yet easy to deploy in real-world environments,”
says Professor Aruna Seneviratne from UNSW.

Science–Industry Collaboration Accelerates Commercialization

The project is being developed in partnership with Trailblazer Recycling and Clean Energy, a $280 million research and commercialization initiative by UNSW and the University of Newcastle. Several Australian technology companies are also involved:

  • GinigAI – provides the patented embedded passive radio monitoring platform, forming the system’s core detection component.
  • Trantek MST – responsible for sensor connectivity and infrastructure management.
  • DeepNeural AI – develops and deploys real-time, scalable AI algorithms.

“Our research focuses on capturing subtle, hard-to-detect variations in Wi-Fi and mm-wave signals, enabling the prediction of thermal runaway before it occurs,”
explains Saksham Yadav, Managing Director of DeepNeural AI.
“We aim to establish a new global standard for proactive, contactless battery safety monitoring.”

Real-World Applications and Global Potential

The UNSW passive system has wide potential across energy and transport sectors, particularly in:

  • Large-scale energy storage systems, where traditional cooling and sensor networks are costly and complex.
  • Transport and logistics, for example in containers shipping batteries or electric vehicles.
  • Residential and commercial installations, as an affordable retrofit solution improving user safety.
  • Critical infrastructure, where battery fires or explosions could have severe economic and social impacts.

Because the system is entirely passive and does not require sensors embedded inside cells, it offers virtually unlimited scalability — opening the door to safer, smarter, and more efficient energy storage management worldwide.

Change consents