Publication

Abstract : As the green movement gains momentum, all types of Electric Vehicles (EVs), such as electric automobiles, buses, trains, motorcycles, and bicycles, are becoming more and more prevalent. The way EVs manage their energy will be significantly influenced by future design, and developments in low-cost sensing and computation will be assessed to create more efficient systems that can be tailored to suit a range of battery types and vehicles with wildly disparate performance needs. Battery Management Systems (BMS) rely on the collection and transmission of data from battery sensors. Because inaccurate battery data from sensor failures, communication issues, or even cyberattacks can cause significant harm to BMS and lower the total reliability of applications based on BMS, such as electric cars, the battery sensors' lifespan and the BMS's transmission data must be evaluated. For a BMS to function properly, sensor data is required. For electric car battery systems to be secure and sustainable, effective detection of sensor failures is essential. This study proposes a deep learning-based method for battery data, specifically lithium-ion batteries, to detect and categorize abnormal battery sensor and gearbox data. The issue of optimizing real-time battery management systems is examined since this performance suggests increased battery thermal stability, efficiency, and durability.