Deep Reinforcement Learning-Based Battery Management System for Electric Vehicles

Electric vehicles (EV) are the future. So what does the future hold for EV batteries? To make these batteries less complicated and easy to charge, it’s important to improve the battery management system (BMS) that monitors, controls and protects the EV lithium-ion battery packs. Considering that EV batteries are subjected to harsh operating cycles and varying environmental conditions leading to very complicated interactions, the EU-funded DeepBMS project will develop a new model to improve the BMS functionality in EVs. Efficient deep reinforcement learning-based algorithms will capture the convoluted time-varying behaviour of battery. DeepBMS will also boost reliability and extend battery lifetime by improving the estimation accuracy in a wide temperature range and over the full life span of the batteries.