Critical Review of State of Charge Estimation Methods for Lithium-Ion Batteries in Electric Vehicle Battery Management Systems

The global shift toward electric vehicles (EVs) to reduce CO۲ emissions underscores the critical role of Battery Management Systems (BMS) in ensuring the safe and efficient operation of lithium-ion batteries (LIBS). Accurate State of Charge (SOC) estimation is essential for optimizing battery performance, extending lifespan, and enhancing EV safety. This paper critically reviews SOC estimation methods for LIBS in EVs, categorizing them into direct, model-based, observer-based, filter-based, data-driven, and hybrid approaches. Each method is evaluated for accuracy, computational complexity, noise robustness, and hardware implementation feasibility. Key challenges, including nonlinear battery dynamics, temperature variations, aging effects, and computational constraints, are analyzed, highlighting trade-offs between model fidelity and real-time applicability. The review also examines hardware platforms, such as microcontrollers, DSPs, FPGAs, and GPUs, and their suitability for SOC estimation in BMS. Recent advancements, such as physics-informed machine learning and uncertainty-aware hybrid models, are discussed alongside future research directions, including standardization, cross-chemistry generalization, and edge-friendly implementations. This study provides a comprehensive framework for developing robust, accurate, and real-time SOC estimation strategies to support the widespread adoption of sustainable EV technologies.

Keywords:

Electric Vehicle (EV) , State of Charge (SOC) Estimation , Battery Management System (BMS) , Artificial intelligence , Hybrid Models