Health analysis results and 2nd life battery description

‍As electric vehicles continue to replace conventional fuel-powered cars, lithium-ion batteries (LIBs) are becoming central to the energy transition. However, once batteries reach the end of their automotive life, their capacity may no longer meet vehicle requirements, even though significant usable energy often remains.

Rather than immediately recycling these batteries, evaluating their State of Health (SOH) opens the possibility of second-life applications such as stationary energy storage, grid stabilization, or renewable energy backup systems. Within FREE4LIB, this deliverable focuses on identifying reliable and scalable methods to assess battery health for safe and efficient reuse.

‍

Context and Objective

Determining whether an end-of-life battery module can be reused requires accurate and practical health assessment methods. Traditional capacity testing — based on full charge and discharge cycles —provides reliable data but is time-consuming and difficult to scale for large volumes of batteries.

The objective of this work was therefore to:

• Compare different SOH assessment techniques,
• Evaluate their reliability, cost, speed, and safety,
• Explore options for faster, data-driven health estimation,
• Develop BMS technology to enable seamless transition from first to second life
• Assess the feasibility of integrating these methods into automated workflows.

Several approaches were considered, including internal resistance analysis, open-circuit voltage trends, incremental capacity analysis (ICA), differential voltage analysis (DVA), and entropy-based methods. Hybrid strategies combining traditional testing with data-driven models were also explored.

‍

Key Findings

The study highlights important insights for second-life battery evaluation:

• Traditional full charge/discharge testing remains reliable but is not ideal for large-scale applications due to time and operational constraints.
• Fast  health indicators (such as ICA or voltage-based analysis) show potential for accelerating SOH estimation, particularly when combined with     data-driven models.
• A hybrid approach — combining limited traditional testing with model-based estimation — appears to offer the best balance between accuracy and     scalability.
• Training  data availability is a key challenge, especially when evaluating multiple module types with different chemistries and histories.
• The  developed BMS prototype successfully tracked voltage, current, temperature, and State of Charge (SoC) in real time.
   
  • Minor measurement offsets were observed but did not affect overall trend accuracy for effective SoH assessment.
   
  • SoC estimation showed consistent and reliable behaviour under load conditions.
   
  • With calibration and model integration, measurement precision can be further improved with data driven approaches.

These results confirm that real-time monitoring combined with intelligent data processing can support reliable second-life battery classification.

‍

Recommendations and Next Steps‍

To strengthen second-life evaluation methodologies, the following actions are recommended:

• Further develop hybrid SOH assessment models combining traditional testing and     data-driven indicators.
• Expand training datasets using additional modules or open-source battery data to     improve model robustness.
• Refine BMS calibration to reduce measurement offsets and improve long-term     monitoring accuracy.
• Integrate health assessment tools into automated workflows for large-scale module     screening.
• Continue validating safety parameters to ensure reliable second-life deployment.

‍

By improving SOH evaluation methods andintegrating real-time monitoring systems, FREE4LIB contributes to safer,faster, and more scalable second-life battery solutions — supporting a morecircular and resource-efficient battery ecosystem.

‍You can read the full deliverable here