In order to solve the problem that the battery model parameters can only be identified off-line in the SOC estimation, a strategy of dynamically identifying the model parameters using the least squares method with forgetting factor is proposed, and the particle filter (PF) algorithm is applied to the SOC estimation. Based on the combined model and PF algorithm, the simulation verification is performed. By comparing the SOC curve estimated by the PF algorithm proposed based on dynamic parameter identification proposed in this paper with the SOC curve obtained under the real NDEC condition, the estimation error of the proposed algorithm is maintained at about 1%. It proves that the algorithm can reflect the battery characteristics and accurately estimate the battery SOC.