Simulation of Li-Ion Pouch Cell Venting during Thermal Runaway
Please login to view abstract download link
As the development of Lithium-ion batteries for electrical vehicles goes towards a maximization of battery energy density, the risk of personal and hardware damage arising from thermal runaway events increases significantly. Therefore, predictive battery simulation models are a very important tool as they allow for a safe, accurate, and cost-efficient design of a battery system. The venting process during thermal runaway plays an important role in the safety design of a battery system. However, it is a multi-physical phenomenon involving a complex interaction of reaction kinetics and reactive gas flow, particle flow, turbulence, and heat transfer. Due to the complex physics of thermal runaways, several simulation approaches with different levels of detail and computational effort are mentioned in literature [1,2]. The model developed in this work differs from other publications as it focuses on the venting behavior of a pouch cell, for which the absence of a defined opening increases the complexity of the venting process. In this work, a thermal runaway simulation model of a current high-energy density pouch cell with NMC chemistry [3] is further developed and validated based on an in-house experimental method. The experimental set-up consists of two battery cells assembled in a closed fire-proof test box with a mounting that ensures close to reality heat transfer and loading conditions. The first cell is triggered into thermal runaway by an attached heater, while the second cell is triggered by the heat release from the first cell. Using measured pressure, temperature, gas species and flow velocity, a battery source-term is defined based on a mass conservation and ideal gas law approach without the necessity of detailed reaction kinetics. The resulting battery source-term is validated and analyzed in a detailed 3D-CFD simulation of the test set-up (Figure 1) applying a conjugate heat transfer (CHT) approach and detailed species model. As the simulation results show that the model reproduces the measurements in the cell tests (Figure 2), it can be further applied in the safety design process of battery packs.
