Accurate State of Charge Estimation for Electric Vehicle Lithium-ion Batteries Using Shapley Additive Explanations and Kolmogorov Arnold Network

Read full text Summarise with VIRAT
Authors: Deepesh S. Kanchan, K. Gopala Reddy
Cited by: 0
References: 23

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

In modern era, Electric Vehicles (EVs) continue to grow, increasing consumer needs for accurate and realtime estimation of the State of Charge (SoC) in lithium-ion (Li-ion) batteries.Traditional SoC estimation approaches such as Coulomb counting, equivalent circuit methods, sensor inaccuracies, and need for battery-specific parameters.Therefore, to resolve these issues, this research proposed SHapley Additive explanation with the Kolmogorov Arnold Network (SHAP-KAN) employs a Correlation Matrix (CM) and SHAP to enable interpretable, robust feature selection, followed by a KAN that effectively predicts the nonlinear SoC.Specifically, CM eliminates linearly redundant features, whereas SHAP produces a lower-order subset of meaningful features that incorporate the quantification of nonlinear contributions.Furthermore, KAN effectively models battery nonlinearity, temperature dependence, and electrochemical behavior using adaptive spline-based transformations with a low parameter burden.The novelty of this research lies in combination of SHAP after CM filtering to maximize feature interpretability and in enhancement of KAN through learnable knot positions and temperature-conditioned modulation, allowing accurate SoC estimation.Its performance was assessed using statistical performance metrics using Normalized Mean Squared Error (NMSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) values, demonstrated significant improvement over existing Deep Neural Networks-Mountain Gazelle Optimizer (DNN-MGO).

Fields of Study & Subjects

Battery Modeling and Management Advancements in Battery Materials Electric and Hybrid Vehicle Technologies

Publication details

Journal / source
International journal of intelligent engineering and systems
Volume
19
Issue
4
Pages
362–376
Publication date
23 March 2026
Language
EN