International Research Journal of Education and Technology

The transition to electric vehicles (EVs) necessitates efficient battery management systems to maximize battery lifespan and enhance energy efficiency. Traditional systems lack the capabilities for real-time monitoring and predictive maintenance, leading to issues like overcharging, deep discharge, and unexpected power drain. This study introduces an IoT-enabled, deep learning-based battery management solution for EVs, focusing on continuous monitoring and anomaly detection. Using sensors to track key battery parameters and an LSTM (Long Short-Term Memory) Auto-encoder model for anomaly detection, the system provides early warning alerts through a user-friendly dashboard. This approach offers users actionable insights to optimize battery performance, extending battery life while reducing costs. Key findings reveal improved energy efficiency and reduced failure rates, validating the system's positive impact on sustainable EV adoption.

Key Words: Electric Vehicle, Battery Management System, IoT, Deep Learning, Anomaly Detection, LSTM Auto-encoder, Real-Time Monitoring.