AI-BASED DIGITAL TWIN FOR EV BATTERY HEALTH MONITORING

Electric vehicles (EVs) rely on lithium-ion batteries whose capacity, power capability, and performance change with temperature, charging patterns, operating conditions, and aging. Conventional battery management systems (BMS) monitor voltage, current, and temperature, but important battery states such as state of charge (SOC), state of health (SOH), and remaining useful life (RUL) must be estimated. An artificial-intelligence-enabled digital twin provides a virtual representation of a physical battery by combining real-time measurements, mathematical models, historical data, and predictive algorithms. This review examines digital-twin architecture, battery degradation, model-based estimation, machine-learning techniques, data preparation, and validation strategies for EV battery health monitoring. It discusses the use of extended Kalman filters, gradient-boosting algorithms, and deep-learning methods for estimating battery states and predicting degradation. The review also identifies challenges related to data quality, model generalization, uncertainty, computational requirements, and cybersecurity. A conceptual architecture and implementation workflow are proposed for developing a data-driven battery health monitoring prototype. The study establishes a foundation for future implementation and experimental evaluation using publicly available battery datasets.