SDG-L: A Semiparametric Deep Gaussian Process based Framework for Battery Capacity Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Hanbing, Wu, Yanru, Li, Yang, Kuruoglu, Ercan E., Zhang, Xuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917006756806656
author Liu, Hanbing
Wu, Yanru
Li, Yang
Kuruoglu, Ercan E.
Zhang, Xuan
author_facet Liu, Hanbing
Wu, Yanru
Li, Yang
Kuruoglu, Ercan E.
Zhang, Xuan
contents Lithium-ion batteries are becoming increasingly omnipresent in energy supply. However, the durability of energy storage using lithium-ion batteries is threatened by their dropping capacity with the growing number of charging/discharging cycles. An accurate capacity prediction is the key to ensure system efficiency and reliability, where the exploitation of battery state information in each cycle has been largely undervalued. In this paper, we propose a semiparametric deep Gaussian process regression framework named SDG-L to give predictions based on the modeling of time series battery state data. By introducing an LSTM feature extractor, the SDG-L is specially designed to better utilize the auxiliary profiling information during charging/discharging process. In experimental studies based on NASA dataset, our proposed method obtains an average test MSE error of 1.2%. We also show that SDG-L achieves better performance compared to existing works and validate the framework using ablation studies.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10621
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SDG-L: A Semiparametric Deep Gaussian Process based Framework for Battery Capacity Prediction
Liu, Hanbing
Wu, Yanru
Li, Yang
Kuruoglu, Ercan E.
Zhang, Xuan
Machine Learning
Lithium-ion batteries are becoming increasingly omnipresent in energy supply. However, the durability of energy storage using lithium-ion batteries is threatened by their dropping capacity with the growing number of charging/discharging cycles. An accurate capacity prediction is the key to ensure system efficiency and reliability, where the exploitation of battery state information in each cycle has been largely undervalued. In this paper, we propose a semiparametric deep Gaussian process regression framework named SDG-L to give predictions based on the modeling of time series battery state data. By introducing an LSTM feature extractor, the SDG-L is specially designed to better utilize the auxiliary profiling information during charging/discharging process. In experimental studies based on NASA dataset, our proposed method obtains an average test MSE error of 1.2%. We also show that SDG-L achieves better performance compared to existing works and validate the framework using ablation studies.
title SDG-L: A Semiparametric Deep Gaussian Process based Framework for Battery Capacity Prediction
topic Machine Learning
url https://arxiv.org/abs/2510.10621