A Comprehensive Review of Hybrid Battery State of Charge Estimation: Exploring Physics-Aware AI-Based Approaches

Accurate State of Charge (SOC) estimation in lithium-ion batteries (LiBs) poses significant challenges due to their nonlinear behavior over their lifetime. Establishing a balance between accuracy, robustness, and low implementation complexity remains critical. Over the past decade, numerous studies have aimed to analyze and compare various SOC estimation methods for commercial LiBs. However, there has been a lack of reviews focusing on SOC estimation from a physics-aware AI-based perspective for LiBs. This paper aims to bridge this gap by evaluating various SOC estimation methods, particularly highlighting hybrid approaches that integrate model-based (MB) and artificial intelligence-based (AIB) strategies. These hybrid methods are categorized into Physics-Aware AI-based (PAAIB) methods and AI-enhanced physical models. The research method involves a comprehensive review of existing literature, discussing the fundamental principles of MB and AIB methods and analyzing their strengths and limitations in capturing the complex dynamics of battery behavior. The paper then explores hybrid SOC estimation techniques in depth, examining each category based on various performance metrics. The paper's main contents include a detailed analysis of hybrid SOC estimation techniques, a comparative review of recent studies, and strategic recommendations for future research. The effects of this research highlight the importance of adopting hybrid SOC estimation methods to improve SOC estimation accuracy and robustness in practical applications. The paper concludes with a summary of key findings, emphasizing the necessity of adopting hybrid approaches for improved SOC estimation in LiBs and their critical role in enhancing the reliability and functionality of battery management systems in electric vehicles.