C2L-Net: Efficient SOC Estimation for Lithium-Ion Batteries
A new model called C2L-Net has been introduced by researchers for estimating the state-of-charge (SOC) of lithium-ion batteries while they discharge. This data-driven approach relies on a mere 20-second historical window, which minimizes computational expenses and eliminates positional bias caused by padding. By distinguishing contextual encoding from the updating of the latest measurements, the model allows for effective temporal analysis and quick adjustments to changing battery conditions. Additionally, it features a chunk-based mechanism for extracting characteristics, utilizing Theta Attention. The findings are published on arXiv.
Key facts
- C2L-Net is a data-driven model for SOC estimation.
- It uses a short historical window of 20 seconds.
- The model reduces computational cost compared to long-history methods.
- It avoids padding-induced positional bias at the start of drive cycles.
- The framework separates contextual encoding from latest-measurement updating.
- It enables efficient temporal modeling and rapid adaptation.
- The model incorporates chunk-based feature extraction with Theta Attention.
- The paper is published on arXiv with ID 2605.08653.
Entities
Institutions
- arXiv