Thermal propagation is one of the major challenges when batteries will be used in dwellings in large scale. It means the exothermic reactions in the cell are out of control and can lead to a fast release of flammable and toxic gases. In a system involving a large number of cells, thermal runaway can rapidly propagate from one battery cell to the whole system, which means substantial fire and explosion risks, an event that is important to mitigate and prevent. Multi-physics simulations together with full-scale testing is a cost-effective method for designing safer batteries. This project aims at simulating thermal runaway initiation and propagation using a multi-physics commercial software GT-Suite. 

A battery thermal runaway model containing 12 prismatic cells based on 3-D Finite Element approach was built using GT-Suite. The computed thermal runaway time instants versus thermal runaway cell number were compared with full-scale experimental data with reasonable agreement. Quantitative sensitivity study on the model input parameters and model space and time resolutions on the computed start time instant and time duration of thermal runaway were performed. The thermal runaway model was then extended with an electric equivalent sub-model to simulate the short circuit. With the electrical model acting as the input to the thermal model, the most interesting output of the simulation is the change in temperature of the cells, dependent on the current in the cells, with respect to time. The current is determined by the value of the external resistance through which the short takes place and the voltage level of the battery pack. The obtained results from the above short circuit simulations can only be used as a starting point and not as absolute values for neither triggering the thermal model nor for accurately simulating a battery under an electrical load. Furthermore, GT-Suite was applied to simulate the gas dispersion inside a room. A comparative study of the dispersion of toxic gases during thermal runaway, utilising an arbitrary release of HCN to represent the battery gases, in a small compartment with natural ventilation was investigated and the results compared the same situation simulated in FDS. The pipe based modelling supported by GT-Suite has limited applicability and overestimated the concentrations close to the ceiling whereas the lateral concentrations where underestimated. 

The multi-physics model for battery thermal runaway process is promising and worth to be applied with care for designing safer batteries in combination with full-scale testing.