Over the past decade, considerable advances have been made on battery safety models but achieving predictive accuracy across a wide range of conditions continues to be extremely difficult. From a numerical perspective, the obstacles are numerous. Multiple physics can potentially be involved and interact with one another, electrochemistry, thermal, mechanical, fluid dynamics and so forth. The question of modeling scale also invariably arises. Is it reasonable to imagine a numerical model resolved at the micro scale being later used in a macro model such as car crash simulation?
In-depth characterization of the mechanical and electrical response of an LR61 alkaline battery is performed, for incorporation into a coupled electro-mechanical battery model using LS-DYNA. Tension tests are performed on the outer metal casing to develop a plasticity model, and a Bayesian Model Calibration analysis is performed to determine crushable foam model parameters for the interior anode, cathode, and separator battery components. Electrochemical impedance spectroscopy, distribution of relaxation times, and Kramers-Kronig analysis are used to determine the electrical response of the battery during incremental crush tests. X-ray imaging is also utilized to determine the dimensions of the inner battery components, and to gain insight into the geometric changes to these individual components that occurs during crushing and the corresponding changes to the electrical behavior. This data is used to determine the required order of a Randles circuit necessary to accurately model the electrical behavior. This experimental data is then incorporated into a solid element LS-DYNA model of the battery utilizing the MAT_24 and MAT_63 material models and the EM_RANDLES_SOLID electrical circuit model.
Exploring Ansys LS-DYNA's Battery Modeling Capabilities