Artificial intelligence algorithms optimize immersion boiling heat transfer strategies to mitigate thermal runaway of lithium-ion batteries

Abstract

Thermal runaway (TR) of lithium-ion batteries is the main cause of fire accidents in Electric Vehicles (EVs) and Energy Storage Stations (ESSs). Mitigating the TR is crucial for keeping safety of EVs and ESSs. The immersion boiling heat transfer technology is a promising candidate for mitigating TR of lithium-ion batteries. In this paper, to address the TR issue induced by tab-overheating at the positive tab of pouch-type lithium-ion batteries, a coupled model, considering electro-thermal model, lumped TR model and boiling heat transfer model, was applied to investigated the mechanism of mitigating TR for pouch-type lithium-ion batteries. Besides, the artificial intelligence (AI) algorithms were applied to analyze the importance of parameters, predict the optimum surface heat flux of batteries and then optimize the key parameters of coolants to reinforce immersion boiling heat transfer performance. The results exhibit that the immersion boiling technology can mitigate TR issue of pouch-type lithium-ion batteries induced by tab overheating. Besides, the importance analysis of parameters of coolants shows that the density, viscosity, and specific heat capacity are the top three parameters that affect the mitigating performance. The AI algorithms behaved a good performance in evaluating and optimizing the mitigating performance for TR of lithium-ion batteries. Hence, this work can provide a refence for improving the safety of EVs and ESSs.