TLDR: Researchers have developed a novel, non-invasive method to monitor the State of Health (SoH) of lithium-ion batteries in real-time. By analyzing a brief discharge pulse at the end of a charge cycle, their technique uses an equivalent electrical model to accurately predict battery degradation with high precision, making it suitable for integration into battery management systems.
Monitoring the health of lithium-ion batteries in real-time is crucial, especially in systems like micro-grids where continuous operation is essential. Traditional methods often require a complete discharge of the battery, which is impractical for systems that need to be constantly available. This challenge has led researchers to seek more innovative and less disruptive solutions.
A new study, presented at the SYMPOSIUM DE GENIE ELECTRIQUE (SGE 2025), introduces a promising method for real-time State of Health (SoH) monitoring of lithium-ion batteries. Developed as part of the 4BLife project, this approach focuses on analyzing a brief discharge pulse applied at the end of a battery’s charging cycle. The core idea is to use the parameters from an equivalent electrical model, which describes how the battery’s voltage changes during this short pulse, to estimate its SoH.
The uniqueness of this method lies in its non-invasive nature. Unlike conventional techniques, it doesn’t require a full discharge, allowing for continuous monitoring without interrupting the battery’s operation. This makes it particularly suitable for integration into Battery Management Systems (BMS), which are vital for optimizing battery performance and lifespan.
How the Method Works
The research team applied a short discharge pulse to the battery when it was fully charged and its State of Charge (SoC) was stable. By observing the voltage response over a brief period (1 to 10 seconds), they could identify specific dynamic parameters within the battery’s electrical model. These parameters, particularly one called Ï„1 (tau-one), showed a strong correlation with the battery’s aging process.
Initial experiments, which began in April 2023, involved four LiFePO4 lithium-ion batteries. These batteries were subjected to controlled charge-discharge cycles at a consistent temperature of 21°C. The data collected from these tests, covering a SoH range from 100% down to 85%, demonstrated the effectiveness of the proposed method.
Also Read:
- AI-Powered Energy Management for Healthcare Facilities: A Case Study in Demand Forecasting and Load Balancing
- AI Uncovers Hidden Insights in Crash Reports: Enhancing Road Safety Data
Promising Results and Future Potential
The researchers trained their SoH estimator using data from two batteries that showed significant degradation. They then successfully predicted the degradation of two other batteries, achieving an impressive average absolute error of approximately 1%. This level of accuracy, combined with the simplicity of the approach, suggests it could be easily implemented in existing BMS.
While the current study was conducted under controlled conditions (stable temperature, full discharges, and a limited SoH range), the results are highly encouraging. The simplicity of the linear regression model, which performed best in their tests, makes it ideal for embedded systems where computational resources might be limited. The team acknowledges that further testing under varying temperatures and with partial discharge cycles is necessary to fully validate the method’s robustness in diverse real-world scenarios.
This innovative technique represents a significant step forward in battery management, offering a reliable and non-intrusive way to track battery health. For more detailed information, you can refer to the full research paper available here.
