Research on Battery Remaining Capacity Estimation Based on Radial Basis Kernel Function-Support Vector Regression Model
Bin Xiao
School of Information Engineering, Guangzhou Vocational College of Technology & Business, Guangzhou, China
ABSTRACT:
Battery remaining capacity estimation is a critical indicator in battery management systems. Accurate estimation of battery remaining capacity, i.e., State of Health (SoH), can guide the timely recycling and cascade utilization of LiFePO4 batteries, contributing to economic savings and environmental protection. This paper proposes a Support Vector Regression (SVR) model using the Radial Basis Kernel Function (RBF) to estimate battery SoH. Sample data is rapidly obtained from aged batteries through the Hybrid Pulse Power Characterization (HPPC) test, and features are constructed using the minimum, maximum, and average values of the hysteresis curve. Hyperparameters of the RBF-SVR model are determined through literature review and empirical analysis. To validate the proposed method, the RBF-SVR model is trained and tested using LiFePO4 battery samples with varying degrees of aging, demonstrating the method’s accuracy and effectiveness.
Published in: International Journal of Research in Interdisciplinary Studies (Volume 2, Issue 12, December 2024)
Page(s): 1-5
Date of Publication: 06/12/2024
Publisher: IJRIS
Cite as: Bin Xiao, “Research on Battery Remaining Capacity Estimation Based on Radial Basis Kernel Function-Support Vector Regression Model,” in International Journal of Research in Interdisciplinary Studies, vol. 2, no. 12, pp. 1-5, December 2024.