.png){kind=icon}
.png){kind=icon}
Return
Edge-Optimized Battery SOH Estimation Using LLM-Based Feature Engineering and TinyELM
.png){kind=icon}
PRE
AI
DOI：10.1109/JIOT.2026.3676996.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This article presents a computationally efficient approach for state-of-health (SOH) estimation. This approach has two novel components: large language model (LLM)-based prompt engineering for feature extraction and an edge-optimized extreme learning machine (TinyELM) for efficient SOH estimation. In our LLM-based feature engineering, we leverage GPT-5 to automatically generate potential health indicators (HIs) from raw voltage and current data, then select two to four features that exhibit strong correlations with the target SOH. Building upon these features, we develop TinyELM as a streamlined and optimized variant of ELM designed for improved estimation accuracy and portability on resource-constrained edge devices. The performance of TinyELM is compared against several deep learning (DL)-based models as well as the conventional ELM model and linear-based models on three edge devices, namely, the STM32, Raspberry Pi 4, and Jetson Nano. Experimental results demonstrate that TinyELM achieves state-of-the-art performance in SOH estimation while delivering up to $28.4\times $ faster inference than the conventional ELM and three times faster inference than the other linear models. Furthermore, the results highlight the effectiveness of our LLM-based feature engineering, showing approximately five times lower root mean square error (RMSE) on average compared to traditional and automated feature engineering approaches.
Keywords:
Extreme learning machine (ELM)
feature engineering
model optimization
state-of-health (SOH) estimation
Journal
IF:
8.9
Papers:
1.4W
Citations:
7.8W