A Comparative Analysis of Thermal Runaway Predictions Across Lithium-Ion Battery Chemistries Used in Electric Vehicles

Thermal runaway remains a critical safety challenge for lithium-ion batteries (LiBs) used in electric vehicles (EVs), with varying characteristics across different chemistries. This study presents a comparative analysis of thermal runaway predictions for five widely used LiB chemistries: Nickel Manganese Cobalt (NMC), Nickel Cobalt Aluminum (NCA), Lithium Iron Phosphate (LFP), Lithium Manganese Oxide (LMO), and Lithium Titanate Oxide (LTO). A hybrid methodology combining controlled experimental abuse tests and advanced physics-based and machine learning models was employed to predict onset temperatures and propagation behavior. Results reveal significant differences in thermal stability and prediction accuracy among chemistries, with LFP and LTO exhibiting higher thermal stability and more reliable model predictions, whereas NMC and NCA showed earlier onset and rapid temperature escalation. These findings have direct implications for battery management system (BMS) design and safety protocols in EVs, emphasizing chemistry-specific thresholds and response strategies.