Leveraging IoT-enabled machine learning techniques to enhance electric vehicle battery state-of-health prediction

Abstract

State of Health (SOH) is a critical parameter for evaluating and managing electric vehicle (EV) battery systems, directly impacting battery reliability, safety, and lifespan, and accurate SOH estimation is essential for optimizing EV performance and efficiency. Machine learning (ML) has recently become a promising approach for SOH prediction, utilizing large datasets, enhanced storage, and advanced computing power to develop sophisticated predictive models. This study introduces an IoT-Fog-Cloud system to enhance SOH prediction by providing a structured approach for selecting and customizing ML models tailored to specific SOH challenges. By analyzing the unique characteristics of EV battery data and SOH estimation accuracy requirements, this research identifies optimal ML algorithms that improve prediction performance. Furthermore, the study emphasizes the need for advancements in battery management systems (BMS) for IoT-integrated EVs, including exploration of novel ML architectures for enhanced predictive accuracy, integration of diverse sensor data for improved real-time monitoring, and development of robust evaluation metrics to ensure model reliability. These contributions are crucial for advancing SOH estimation methods and ultimately extending the efficiency, performance, and longevity of EV batteries.