New sensor system provides early warning for battery overheating

A multi-model framework uses fiber optic sensors to detect thermal runaway risks.

A new multi-model collaborative early-warning framework helps identify thermal-runaway risks. This system uses distributed temperature sensing to monitor battery packs.

The integrated sensing scheme increases temperature sensing points from 32 to 540. This is done with minimal wiring complexity. The system was validated in real-world vehicle operation.

The framework uses temperature Shannon entropy (SE) and average temperature variation. Temperature Shannon entropy (SE) is used to derive a dynamic warning boundary. This boundary includes error tolerance.

The system combines thresholds for temperature, temperature change rate, and SE change rate. This enables collaborative early warning. It provides maximum warning lead times of 706 s in laboratory modules. It provides 554 s in real-world vehicle operation.

Conventional battery management systems (BMS) face constraints. These constraints include sparse temperature measurements and monolithic diagnostic frameworks.

Key facts

  • Sensing points in a battery pack increased from 32 to 540.
  • The framework provides a maximum warning lead time of 706 s in laboratory modules.
  • The framework provides a maximum warning lead time of 554 s in real-world vehicle operation.

All sections

Edition