Enabling Rapid State of Health Offline Estimation of a 48V Lithium-Ion Battery Pack

This paper investigates a model-based procedure for rapid SOH estimation of a 48V battery pack used in the automotive sector. The proposed approach is based on an artificial neural network, is off-line, and does not need the vehicle to be operational, thus reducing on-board computational cost and hardware complexity. Moreover, the algorithm discussed in this paper can be easily implemented in a prototype hardware and it allows estimating the SOH of a 48V battery pack in a matter of minutes with a 96.3% accuracy when compared to experimentally acquired data.