Battery calendar aging prediction is of extreme importance for developing durable electric vehicles. This paper derives machine learning-enabled calendar aging prediction for lithium-ion batteries. Specifically, the Gaussian process regression (GPR) technique is employed to capture the underlying mapping among capacity, storage temperature, and SOC. By modifying the isotropic kernel function with an automatic relevance determination (ARD) structure, high relevant input features can be effectively extracted to improve prediction accuracy and robustness. Experimental battery calendar aging data from nine storage cases are utilized for model training, validation, and comparison, which is more meaningful and practical than using the data from a single condition. Illustrative results demonstrate that the proposed GPR model with ARD Matern32 (M32) kernel outperforms other counterparts and can achieve reliable prediction results for all storage cases. Even for the partial-data training test, multi-step prediction test and accelerated aging training test, the proposed ARDbased GPR model is still capable of excavating the useful features, therefore offering good generalization ability and accurate prediction results for calendar aging under various storage conditions. This is the first known data-driven application that utilizes the GPR with ARD kernel to perform battery calendar aging prognosis.

Lithium-ion (Li-ion) batteries are still the best technology to power the Electric Vehicle (EV), due to their high power and energy density. However, the use of these batteries can be limited in cars with a high demand for peak power and very high energy density. One way to improve the performance of the Li-ion battery and reduce its weight is to associate this battery with another technology of higher specific energy as a second energy source, e.g. Lithium-sulfur (Li-S). The development of Hybrid Energy Storage Systems (HESSs) is a promising solution optimizing the energy management of EVs. In this paper, we present experimental results obtained with a high specific energy and power capability HESS prototype, composed of i) a Lithium-Titanate-Oxide battery to ensure high power capabilities, ii) a Li-S battery to improve specific energy, and iii) a power converter based on Gallium Nitride (GaN) devices to link both battery modules, minimizing at the same time system weight, volume and power losses. The developed GaN-based power converter achieves high efficiency (96.5%) operating at 300 kHz with a reduced size (0.4 L). Besides, the behavior of the developed HESS prototype is experimentally evaluated under standard automotive profiles, for different driving scenarios. The HESS prototype achieves an increase of energy density and specific energy of +5.56% and +28.21%, respectively, compared to a battery system composed only of Li-ion cells. The limitation of the developed system for automotive application are highlighted, and the critical research needs are clearly identified to accelerate the implementation of such systems on commercial EV solutions.