Artificial intelligence-based lifetime estimation for battery in electric vehicles

• Main Author: Manya, Miglani
• Other Authors: Hung Dinh Nguyen
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering
• Online Access: https://hdl.handle.net/10356/176592
• Institution: Nanyang Technological University

This report presents a data-driven method to estimate the lifetime of Electric Vehicles (EVs) batteries using simulations based on a Li-Ion battery model. Accurate estimation of the lifetime of EV batteries is essential for improving the efficiency of EVs. It focuses on understanding the factors that contribute to battery degradation, such as Resistance, State of Charge (SOC), Capacity, and Current, and how they affect the End of Life (EOL) of these batteries. The research highlights the importance of SOC and Current range, which are studied through simulations in constant current (CC) mode under two main scenarios: changing SOC and fixed SOC while varying charging/discharging currents for every simulation. The project uses neural networks and regression models such as support Vector Regression (SVR), Random Forest, Bagging Regressor, and Long Short-Term Memory (LSTM) networks with data from MATLAB simulations. These predictions help to know when the EV batteries would require maintenance or if they could be reused for secondary applications like energy storage systems, commercial buildings and EV fast charging stations. This research will help create a stronger battery management system( BMS) and contribute to the sustainability of EVs.