Fault Diagnosis for Electric Vehicle Battery Pack Interconnection System Using Real-World Driving Data

Electric vehicles (EVs) have gained prominence for addressing global challenges such as climate change and sustainability. With rising EV adoption, there is a growing need for efficient diagnostic methods adaptable to diverse conditions to ensure vehicle reliability and longevity. This study presents a generalized approach to diagnosing degradation and faults in EV battery packs, utilizing real-world driving data to enhance fault detection accuracy. Scaled-down experiments with cylindrical batteries simulated various fault conditions and operational states, enabling the development of a fault diagnosis methodology based on current–voltage profile analysis. This methodology accurately identifies faults, such as external wire harness issues, interconnect busbar anomalies, and individual cell defects, and is adaptable to different battery configurations and environments. Furthermore, a diagnostic technique for battery interconnect systems (BISs) was developed using temperature-compensated resistance calculations from real-world data. Empirical results demonstrate the approach's effectiveness in detecting and categorizing faults and age-related degradation within EV BISs across various conditions. Notably, we found that a twofold increase in BIS resistance reduces battery efficiency by 1.4% in terms of motor output and raises total energy consumption by 10%. This generalized methodology provides a framework for evaluating BIS performance, enhancing reliability, and optimizing maintenance for diverse EV applications.