A virtual sensor fusion approach for state of charge estimation of lithium-ion cells

2025-08-06

AI Summary

To address the low accuracy and poor robustness of state-of-charge (SOC) estimation for lithium-ion batteries, this paper proposes a hybrid virtual sensor method integrating machine learning (ML) with an extended Kalman filter (EKF). The approach features an affine parameter-varying equivalent circuit model, a linear observer array, and incorporation of ML-predicted outputs as auxiliary measurements in the EKF framework. Additionally, a data-driven online calibration mechanism for process and measurement noise covariances is introduced. This architecture synergistically combines model-based and data-driven paradigms. Experimental results under dynamic operating conditions demonstrate that the proposed method achieves a mean absolute SOC estimation error below 1.2%, substantially outperforming conventional EKF and purely data-driven approaches. Moreover, the estimated SOC trajectory exhibits improved smoothness, faster convergence, superior generalization across diverse battery aging states and temperatures, and strong practical applicability for real-world battery management systems.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Security and privacy of machine learning and AI applications

Abstract

This paper addresses the estimation of the State Of Charge (SOC) of lithium-ion cells via the combination of two widely used paradigms: Kalman Filters (KFs) equipped with Equivalent Circuit Models (ECMs) and machine-learning approaches. In particular, a recent Virtual Sensor (VS) synthesis technique is considered, which operates as follows: (i) learn an Affine Parameter-Varying (APV) model of the cell directly from data, (ii) derive a bank of linear observers from the APV model, (iii) train a machine-learning technique from features extracted from the observers together with input and output data to predict the SOC. The SOC predictions returned by the VS are supplied to an Extended KF (EKF) as output measurements along with the cell terminal voltage, combining the two paradigms. A data-driven calibration strategy for the noise covariance matrices of the EKF is proposed. Experimental results show that the designed approach is beneficial w.r.t. SOC estimation accuracy and smoothness.

Problem

• Estimating State of Charge in lithium-ion cells
• Combining Kalman Filters with machine-learning methods
• Improving SOC estimation accuracy and smoothness

Innovation

• Combines Kalman Filters with machine-learning for SOC
• Uses Affine Parameter-Varying model from data
• Data-driven calibration for noise covariance matrices