Lithium-ion batteries offer high energy density and low self-discharge, making them a key energy storage system for transportation electrification. However, the coupled electrochemical dynamics of lithium-ion batteries result in nonlinear characteristics and a multimodal optimization landscape, making accurate parameter identification challenging. To address the limitation, we propose a novel uniform puma optimizer (UPO) for battery model parameter optimization. UPO incorporates a uniform mutation strategy into the puma optimizer (PO) to maintain population diversity and enhance global search capability. In the identification stage, the hybrid pulse power characterization test was conducted to establish an accurate relationship between open-circuit voltage and the state of charge, while UPO was employed to identify battery model parameters. In the validation stage, capacity tests at 0.5 and 1 C are performed to evaluate the model accuracy. The results show that the UPO-based model achieves mean absolute error values of 26.93 and 44.01 mV at 0.5 and 1 C, respectively, representing reductions of 13.96 and 15.04% compared with the PO-based model. These results confirm that UPO provides an effective and robust parameter identification method, supporting the development of accurate battery models for advanced battery management systems.