Multi-Physics Battery Modelling and Calibration for Electric Vehicles
Please login to view abstract download link
Lithium-ion batteries play an important role in the electrification of transportation systems. Accurate prediction of their performance and safety under operational and abuse conditions requires physics-based multi-physics modeling. In this work, a fully coupled thermal–electrical–mechanical battery cell model was developed and calibrated for electric vehicle applications. A commercial pouch-type EV battery cell is physically disassembled, and the mechanical, electrical, and thermal properties of all components were tested and characterized into material models. The multilayer sandwich structure of the real cell is represented using a reduced number of equivalent layers, preserving the realistic material properties of each component, including the pouch bag, cathode current collector, cathode coating, separator, anode coating, and anode current collector. Cell electrical behavior is modeled using a Randles Circuit with generalized voltage source in a resistive heating solver for efficiency considerations. A combined thermal, electrical and mechanical contact model is used to simulate separator failure, electrodes short circuits, joule heat generation, and heat dissipation, which can lead to thermal runaway. Cell-level punch tests are used to validate the mechanical response of the model. The HPPC and EIS tests are used to calibrate the equivalent circuit model. Cell-level thermal chamber tests are used to validate the ignition point of thermal runaway predicted by simulation. The proposed multi-physics battery model has the following capabilities: (1) simulating thermal runaway under mechanical, electrical, and thermal abuse conditions [1], (2) connect multiple cells in series and parallel to represent realistic high-voltage battery circuits, and (3) performing coupled multi-physics crash simulations at the module, pack, and full vehicle levels [2]. The model provides a computationally efficient and physics-consistent framework suitable for EV safety and system-level analysis.
