Safety risk assessment method for thermal abuse of lithium-ion battery pack based on multiphysics simulation and improved bisection method

The thermal safety of lithium-ion (Li-ion) batteries is of great importance for their further development and application. The accurate evaluation of the thermal safety boundary has always been the key issue. Due to the coupling effect of the internal, external, and randomness factors, the risk probability is difficult to evaluate using existing deterministic analyses. Therefore, a safety risk assessment method for the thermal abuse of Li-ion battery packs is proposed, and an improved bisection-method-based analysis algorithm for the thermal safety boundary is established. Moreover, a multiphysics model is developed considering an equivalent circuit, thermal abuse, and a fluid dynamics model. Furthermore, stochastic models of the battery parameters and loading are con-structed to describe the randomness. The temperature and power stress-strength interference models are inte-grated to evaluate the thermal safety risk of the battery pack. Then, the models are validated by the temperature and analyzing the stochastic parameters. Finally, several case studies are implemented, including the analyses of thermal safety boundary, effect of stochastic parameters, safety risk probability, and thermal runaway propa-gation in different scenarios. The results denote that the effect of internal resistance dispersion is dominant and that the risk probability will increase with degradation.

Automated summary

The thermal safety of lithium-ion batteries is crucial for their further development and application. Existing deterministic analyses struggle to accurately evaluate the risk probability due to the coupling effect of various factors. To address this, a safety risk assessment method and an improved bisection-method-based analysis algorithm for the thermal safety boundary are proposed. By considering an equivalent circuit, thermal abuse, and fluid dynamics, a multiphysics model is developed, along with stochastic models for battery parameters and loading to account for randomness. The integration of temperature and power stress-strength interference models enables the evaluation of thermal safety risk for battery packs. Results from case studies show that internal resistance dispersion has a dominant effect and the risk probability increases with degradation.