Hybrid Compression for Accurate End of Life Prediction on Edge Battery Management System
Date:
Presented research on hybrid compression techniques for accurate End of Life (EOL) prediction on edge Battery Management Systems at the 2025 IEEE Global Reliability & PHM Conference in Xi’an, China.
Abstract
This research proposes a hybrid compression technique that enables accurate End of Life (EOL) prediction while maintaining computational efficiency on edge Battery Management Systems. The method combines knowledge distillation and model compression to achieve real-time performance on resource-constrained devices.
Key Contributions
- Hybrid compression framework for edge deployment
- Knowledge distillation techniques for model efficiency
- Accurate EOL prediction with minimal computational overhead
- Real-time capable implementation on edge devices
Applications
The proposed method is particularly suitable for:
- Electric vehicle battery management systems
- Energy storage systems
- IoT-enabled battery monitoring applications
Authors: Kim T., Seo Y., Barde S.