At present, the widely used battery models mainly include the equivalent circuit model (ECM), data-driven model (DDM) and electrochemical model [41, 42]. ECM uses different combinations of electrical components such as resistance, capacitance, voltage source to simulate the charge and discharge characteristics of lithium-ion batteries [43, 44]. Bec. In order to make electric vehicles operate safely and efficiently, the battery SOC, SOH estimation and RUL prediction are important, which can provide a basis for the energy management and safety management of electric vehicles. The electrochemical reaction process and reaction stage of the battery are complex and difficult to determine, and the co. Although researchers have done a lot of research work on the parameterization of electrochemical models, they still face another challenges. The model parameters can be obtained by using different methods. It is difficult to accurately estimate the variation of battery internal parameters during battery aging. Understanding the regularity of electr. DT models require multi-scale physical models of all aspects of the battery and real-time updates of the models. It is worth noting that the focus of the DT model is the reduction of the model. Instead of coupling multiple physical models together, multiple individual models represent the various properties of the battery. The internal behavior of. Battery modeling is the basis of DT, and battery charging and balancing strategies need to consider state and aging tracking of the battery, so real-time state estimation is also the basis of battery feedback control. The biggest advantage of the DT is that it can realize the online state estimation and then realize dynamic management of the batter.