With growing global commitment to renewable energy generation, the role of energy storage systems has become a central issue in traction power applications, such as electric vehicles, trains, and elevators. To achieve the optimal integration of batteries in such applications, without unnecessary oversizing, improvements in the process of battery selection are needed. Specifically, it is necessary to develop models able to predict battery performance for each particular application. In this paper, a methodology for the parametrization of a battery equivalent circuit model (ECM) based on capacity and pulse tests is presented. The model can be extrapolated to different battery technologies, and was validated by comparing simulations and experimental tests with lead-acid and lithium-ion batteries.

This article presents a novel framework for onsite estimation of lithium-ion battery electrochemical parameters by combining physics-informed neural networks (PINNs) with transfer learning (TL). Unlike conventional physics-based models that require costly offline parameterization or extensive experimental tests, the proposed two-phase strategy first trains a PINN using only the governing equations of the single particle model, and then fine-tunes selected electrochemical parameters with field voltage measurements while keeping most network weights frozen. This enables efficient adaptation to ageing and cell-to-cell variability with minimal data requirements. The approach achieves high accuracy in estimating key parameters such as diffusion coefficients and active material volume fractions under different degradation conditions. Under a simulation environment, the method reconstructs voltage responses with a root-mean-square error of 8.94 mV and estimates active material fractions with an average relative error below 2.2% across 400 study cases, significantly outperforming classical optimization-based estimators. Experimental validation on an aged NMC cell (82.09% nominal capacity) implemented on a Raspberry Pi 5, demonstrates practical feasibility, yielding a relative error of 3.89% in estimating volume fractions with a computational cost of 106 ms per iteration. These results confirm that the proposed TL-based PINN framework offers an accurate, computationally efficient, and easily deployable solution for real-time battery management systems.