Fig. 1

Framework of the battery digital twin technology"

Fig. 2

(a)The schematic and photograph of integrated functional electrode[25]; (b) The configuration of the optical fiber sensor implanted inside a commercial 18650 cell[29]; (c) The schematic illustration of wireless signal transmission[41]; (d) The schematic diagram of a full cell with a single optical fiber[52]; (e) The schematic of the in-situ strain measurement inside a cylindrical Li-ion battery[55]"

Fig. 3

(a) The gas pressure sensor on the top of the prismatic lithium-ion battery[57]; (b) The relative pressure sensor attached onto the perforated negative electrode of the cylindrical lithium-ion battery[58]; (c) The commercial lithium-ion battery with an implanted MEMS fiber-optic pressure sensor[59]; (d) Rapid measurement of gases inside a lithium-ion battery based on multiple NDIR sensors in a sealed tank[61]"

Fig. 4

(a) Schematic illustration of P2D model during discharge[79]; (b) Schematic illustration of SPM model during discharge[79]; (c) 2D simulation of temperature distribution inside a 18650 cell at the end of 1C (left-hand side) and 2C (right-hand side) discharge[71]; (d) 3D simulation of total heat generation (left-hand side) and temperature distribution (right-hand side) of a pouch battery at the end of 3C discharge[76]"

Fig. 5

(a) 2D simulation of von Mises stress in the fully charged state for a cylindrical cell[83]; (b) 3D simulation of displacement distribution of a pouch battery at the end of charge[85]; (c) Displacement distribution (left-hand side) and von Mises stress distribution (right-hand side) of a pouch cell at the end of charge[86]; (d) Temperature (left-hand side) and the projected stress fields (right-hand side) inside the jellyroll at the end of discharge under the forced convection condition[88]; (e) Temperature (left-hand side) and the projected stress fields (right-hand side) in jellyrolls for the nonisothermal case after being charged for 50 min[89]"

Fig. 6

(a) Digital twin framework for real-time battery SOC estimation[97]; (b) Digital twin framework for real-time lithium-ion battery monitoring[101]; (c) The explainable battery digital twins framework[104]"

Fig. 7

(a) ETNN model structure (left-hand side) and NN detail (right-hand side) for battery SOC estimation[110]; (b) Framework of the PIML-based multi-fidelity model for battery SOH estimation[116]; (c) Deep learning framework for battery SOH estimation[122]"

Fig. 8

The five-dimensional battery digital twin framework[106]"

Fig. 9

(a) Framework of the PI-LSTM for battery RUL prediction[123]; (b) The adaptive evolution enhanced PINN-based time-variant health prognosis framework[117]"

Fig. 10

The cloud–edge collaborated online adaptive battery ROM framework[108]"

Fig. 11

(a) Schematic diagram of the mechanism and data-driven fusion model[128]; (b) Framework of the PIML for battery degradation diagnostics[130]"