39 / 2025-05-12 18:20:32

Accurate Online Capacity Estimation of Lithium-Ion Batteries: A Multi-Source Feature Fusion Approach with Interpretability Analysis

lithium-ion battery, multi-source feature fusion, XGBoost, capacity estimation, interpretability

Special Sessions > SSL 06 The reliability of intelligent equipment

Accurate online capacity estimation of lithium-ion batteries represents a critical technological challenge for ensuring the safe operation of Battery Management Systems. Conventional data-driven methods often suffer from model overfitting and ambiguous decision-making when characterizing the complex nonlinear and time-varying dynamics of battery degradation processes. To overcome these limitations, this study proposes an interpretable capacity estimation framework based on multi-source heterogeneous feature fusion. The methodology first extracts multi-scale health features from three aspects: charge-discharge cycle profiles, incremental capacity curves and differential thermal voltammetry curves. Pearson correlation analysis and principal component analysis are then applied sequentially to eliminate redundant features and compress dimensionality. A robust XGBoost regression model is subsequently trained on the optimized feature set for capacity estimation, with SHapley Additive exPlanations values systematically quantifying feature contributions to model decisions. Experimental validation on NASA's lithium-ion battery dataset demonstrates the framework's superiority. The results demonstrate that the proposed method achieves significant improvements in capacity estimation accuracy, with the root mean square error and mean absolute error under leave-one-out cross-validation both maintained below 2%.