Electric SUV Fire Simulation Reveals Critical Escape Window
In a groundbreaking study that could reshape how emergency responders and vehicle designers approach electric vehicle (EV) safety, researchers at Nanjing Forestry University have conducted a high-fidelity numerical simulation of a full-scale fire involving a pure electric sport utility vehicle (SUV) powered by ternary lithium-ion batteries. The findings, published in Energy Storage Science and Technology, underscore the rapid progression of thermal runaway-induced fires and highlight a perilously narrow window—just 15 seconds—for occupants to detect smoke and potentially escape before conditions become lethal.
The research, led by Qilin Guo under the supervision of Professor Zheshu Ma, offers one of the most detailed virtual reconstructions of an EV fire to date. Unlike previous studies that focused primarily on battery modules or simplified cabin environments, this work integrates real-world vehicle geometry, multiple combustible interior materials—including seats, door panels, dashboard components, and tires—and realistic thermal propagation dynamics. The result is a comprehensive model that mirrors actual fire behavior observed in field incidents, providing actionable insights for both public safety and automotive engineering.
The impetus for the study stems from a growing concern over the unique fire hazards posed by electric vehicles. While EVs represent a critical step toward decarbonizing transportation, their high-energy-density battery packs introduce new risks when compromised. Thermal runaway—a self-sustaining chain reaction triggered by mechanical damage, electrical faults, or overheating—can cause batteries to ignite violently, releasing intense heat, toxic fumes, and flames that spread with alarming speed. Unlike conventional gasoline fires, which often originate in the engine compartment and may provide visible warning signs, EV fires can erupt from the undercarriage with little external indication until it’s too late.
The team selected a real-world incident as their case study: a pure electric SUV that spontaneously combusted while parked at a repair shop. Prior to the fire, the vehicle had suffered underbody damage from a ground impact, deforming the battery pack’s rear-left section and compromising its cooling system. Although the casing remained intact and no immediate electrical faults were detected, internal cell deformation led to a latent short circuit. Hours later, the vehicle emitted white smoke, which quickly turned black as flames erupted from the chassis, engulfing the cabin and front power compartment while leaving the rear relatively less damaged—a pattern the simulation successfully replicated.
Using advanced computational fluid dynamics (CFD) and large eddy simulation techniques, the researchers built a full-vehicle digital twin based on the SUV’s actual CAD geometry. The model incorporated precise material properties for each combustible component: the battery’s electrolyte (with an ignition temperature of just 110°C), polyurethane foam seats, PVC door trims, rubber floor mats, and tire compounds. The ignition source was modeled as a single battery cell with a heat release rate of 1,535.82 kW/m²—a value validated against experimental data from prior studies.
The simulation revealed a three-phase fire evolution. In the first phase (0–60 seconds), thermal runaway propagated within the battery pack. Despite structural containment, heat and smoke began escaping through chassis gaps. By approximately 35 seconds, visible flames emerged from the underbody, spreading forward more rapidly than rearward due to the fire’s origin near the front portion of the battery pack. This directional bias was consistent with post-incident damage assessments showing greater destruction in the front of the vehicle.
The second phase (60–120 seconds) marked the transition to cabin involvement. As heat radiated upward, interior materials—particularly the seat foam and door panels—reached their ignition points. Smoke accumulated at the roof level, reducing visibility and oxygen levels. At around 70 seconds, thermal stress caused the driver-side window to shatter, introducing a surge of fresh air that intensified combustion in a phenomenon akin to backdraft. This sudden influx of oxygen transformed localized smoldering into open flaming, dramatically increasing temperatures inside the passenger compartment.
The third and final phase (120–150 seconds) saw full vehicle involvement. Flames extended into both front and rear power compartments, though the rear remained cooler due to distance from the ignition source. The peak heat release rate reached 5,100 kW—well within the documented range for large EV fires—and localized temperatures exceeded 800°C, with the battery zone peaking near 900°C.
Perhaps the most alarming finding concerned smoke propagation. A virtual smoke detector placed at the driver’s head position recorded the first traces of smoke just 15 seconds after ignition. Within 10 seconds—by the 25-second mark—the entire cabin was filled to near-total opacity, with smoke concentration exceeding 60% almost immediately and approaching 100% by 40 seconds. Critically, this occurred while the vehicle remained sealed; windows had not yet broken, and ventilation systems were inactive. This means occupants would be exposed to dense, toxic smoke long before flames become visible or heat becomes unbearable.
The implications for occupant safety are profound. In traditional vehicle fires, occupants often have minutes to react—time to pull over, exit the vehicle, and call for help. In contrast, this study suggests that in an EV fire initiated by battery thermal runaway, the critical escape window may be measured in seconds. By the time most drivers notice smoke or odor, the cabin could already be uninhabitable. Carbon monoxide, hydrogen fluoride, and other toxic gases released during lithium battery combustion further compound the danger, potentially causing disorientation or unconsciousness before evacuation is possible.
From a design perspective, the research highlights several areas for improvement. First, battery pack enclosures must not only resist penetration but also manage internal pressure and vent gases safely away from the cabin. Second, vehicle underbody sealing and fire barriers between the battery and passenger compartment need reevaluation—especially around cable penetrations, cooling lines, and structural apertures that serve as smoke pathways. Third, early detection systems capable of identifying off-gassing or temperature anomalies within the battery module could provide crucial advance warning, potentially triggering alarms or automatic shutdowns before thermal runaway becomes irreversible.
For emergency responders, the study reinforces the need for specialized EV firefighting protocols. Standard approaches—such as attacking a fire from the front or using limited water—may be ineffective against deep-seated battery fires that reignite hours or even days later. The simulation’s confirmation of forward-biased flame spread also suggests that approach angles and cooling strategies should prioritize the front undercarriage and battery midsection.
Moreover, the research challenges assumptions about fire behavior in different vehicle classes. Most prior EV fire studies have focused on sedans, which have lower ground clearance and different weight distributions. SUVs, with their higher ride height, larger battery packs, and more spacious interiors, present distinct thermal dynamics. The inclusion of realistic interior materials—often omitted in simplified models—proved essential to capturing accurate smoke development and heat feedback loops that accelerate cabin ignition.
The team’s methodology also sets a new benchmark for virtual fire testing. By calibrating their single-cell model against empirical heat release data and then scaling up to a full vehicle with heterogeneous combustibles, they demonstrated that high-fidelity simulations can reliably replicate real-world fire progression without the cost, danger, and logistical complexity of full-scale burn tests. This paves the way for virtual prototyping of fire safety features during the design phase, potentially reducing reliance on destructive physical testing.
Looking ahead, the researchers suggest several avenues for future work. These include simulating fires in confined spaces like underground parking garages—where smoke accumulation and limited egress amplify risks—as well as exploring the effects of different battery chemistries (e.g., lithium iron phosphate vs. ternary lithium) on fire severity. Additionally, integrating human factors—such as occupant reaction time, egress speed, and the impact of impaired visibility—could yield even more realistic safety assessments.
In an era when electric vehicles are rapidly entering mainstream markets, understanding their failure modes is not just an engineering concern but a public safety imperative. This study by Guo, Tao, Ma, Gu, and Wang provides a sobering yet invaluable look at how quickly an EV fire can escalate from a hidden internal fault to a life-threatening inferno. It serves as a clarion call for automakers, regulators, and first responders to prioritize thermal runaway mitigation, early warning systems, and occupant protection strategies tailored to the unique physics of electric mobility.
As Professor Zheshu Ma noted, “The goal isn’t to discourage EV adoption, but to make it safer. Every second counts in a fire—our models show that with better design and awareness, we can give people those seconds back.”
The full study, titled “Numerical simulation analysis of combustion of electric sport utility vehicles,” appears in Energy Storage Science and Technology, Volume 13, Issue 3, March 2024, with DOI: 10.19799/j.cnki.2095-4239.2023.0762. The research team includes Qilin Guo, Liangyu Tao, Zheshu Ma, Yongming Gu, and Yuting Wang from the College of Automobile and Traffic Engineering, Nanjing Forestry University, Nanjing 210037, Jiangsu, China.