An efficient battery management system for electric vehicles using IoT & Blockchain

• Proposes an integrated EV-BMS using IoT, LightGBM-based ML, and Blockchain. • Achieves 96.52% accuracy in charging station identification. • Introduces a Power Scheduling Algorithm (PSA) for reduced charging delays and congestion. • Utilizes Hyperledger Blockchain to ensure secure, decentralized EV-grid transactions. • Demonstrates a 12% reduction in communication overhead compared to existing methods. • Ensures real-time EV status tracking using Swindon Silicon System (SSS) sensors. • Offers a scalable and secure framework for smart mobility systems. Research and development in the vehicle industry have emphasized the potential for advancing electric transportation that is highly efficient, secure, and sustainable. The electric vehicle (EV), powered by renewable energy sources and equipped with high-efficiency electric motors and controls, offers a practical, dependable, and ecologically friendly urban transportation system. EVs operate using a battery that is equipped onboard. Practical and dependable system operation relies heavily on managing and monitoring batteries. Nevertheless, the market for electric vehicles has experienced a decline in growth due to their limited lifespan and high price. To enhance the system's efficiency and lifespan, substantially improving the battery management aspect is imperative. In this research, the Internet of Things (IoT), machine learning (ML), and Blockchain (BC) technologies are used to develop an energy-efficient EV battery management system (BMS). The IoT sensors are attached to the electric vehicles to collect data such as the charging level, the distance that must be driven, and the position of the electric vehicles. This information was saved and processed by a database, then inputted to the LightGBM classifier to determine the cost of charging. After that, it was processed by the power scheduling approach (PSA) to determine the space and time of charging that is closest to a particular electric vehicle and the charging site. At last, this information is saved in blocks to prevent electric vehicles from being misrouted and ensure that pricing transactions between users and charging stations are conducted securely using BC. The results demonstrate that the research model provided enhanced EV-BMS with an accuracy rate of 96.52% and that it retains a communication overhead that is 12% lower compared to the other models.