Cloud-Based Deep Learning for Co-Estimation of Battery State of Charge and State of Health

Rechargeable lithium-ion batteries are currently the most viable option for energy storage systems in electric vehicle (EV) applications due to their high specific energy, falling costs, and acceptable cycle life. However, accurately predicting parameters of complex, nonlinear battery remains challenging, given diverse aging mechanisms, cell-to-cell variations, dynamic operating conditions. The states becoming increasingly important ubiquitous application scenarios, yet our ability predict cell performance under realistic conditions limited. To address challenge modelling evolution multiphysics multiscale systems, this study proposes a cloud-based AI-enhanced framework. framework aims achieve practical success co-estimation state charge (SOC) health (SOH) during system’s operational lifetime. Self-supervised transformer neural networks offer new opportunities learn representations observational data with multiple levels abstraction attention mechanisms. Coupling cloud-edge computing versatility deep learning can leverage predictive exploiting long-range spatio-temporal dependencies across scales.