The proliferation of electric vehicles (EVs) has brought stringent safety requirements for their core energy storage component, the EV battery pack. Among various safety threats, a high-velocity impact to the underside of the vehicle by road debris poses a significant risk. Such an event can cause severe local deformation of the battery enclosure, potentially leading to internal short circuits within the lithium-ion cells, subsequent thermal runaway, fire, or even explosion. Therefore, enhancing the crashworthiness of the bottom structure of an EV battery pack and improving the predictive accuracy of simulation tools for these events are of paramount engineering importance. This study focuses on the explicit dynamics simulation of a standardized bottom crash test for a specific vehicle’s EV battery pack, aiming to accurately assess structural integrity and cell safety.
The bottom crash test for the EV battery pack is performed in accordance with the national standard GB 38031-2020. The test setup involves a hemispherical impactor with a radius (R) of 15 mm, made of 45# steel. The impactor is directed along the vehicle’s vertical (+Z) axis, perpendicular to both the longitudinal (X) and lateral (Y) directions. The impact energy is set at 150 J, and the target locations are predefined as high-risk points on the underside of the EV battery pack, as specified by the manufacturer.
The dynamic response of the EV battery pack under such an impact involves large deformations and high nonlinearities, making explicit dynamics the preferred numerical method. The governing equation of motion is given by:
$$ \mathbf{M}\ddot{\mathbf{U}} + \mathbf{C}\dot{\mathbf{U}} + \mathbf{K}\mathbf{U} = \mathbf{F}(t) $$
where \(\mathbf{M}\), \(\mathbf{C}\), and \(\mathbf{K}\) are the mass, damping, and stiffness matrices, respectively; \(\mathbf{U}\) is the displacement vector; and \(\mathbf{F}(t)\) is the external force vector. The Central Difference Method is applied for time integration, leading to the following update equations:
$$
\begin{aligned}
\hat{\mathbf{M}} \mathbf{U}^{(t+\Delta t)} &= \mathbf{R}^{(t)} \\
\hat{\mathbf{M}} &= \frac{1}{\Delta t^2} \mathbf{M} + \frac{1}{2\Delta t} \mathbf{C} \\
\mathbf{R}^{(t)} &= \mathbf{F}^{(t)} – \left(\mathbf{K} – \frac{2}{\Delta t^2}\mathbf{M}\right)\mathbf{U}^{(t)} – \left(\frac{1}{\Delta t^2}\mathbf{M} – \frac{1}{2\Delta t}\mathbf{C}\right)\mathbf{U}^{(t-\Delta t)}
\end{aligned}
$$
Furthermore, the principle of energy conservation is critical for validating the simulation. Based on the principle of virtual work and continuum mechanics, the energy balance for the system can be expressed as:
$$ \int_V \rho \ddot{x}_i \delta x_i dV + \int_V \sigma_{ij} \delta x_{i,j} dV – \int_V \rho f_i \delta x_i dV – \int_S p_i \delta x_i dS = 0 $$
Here, \(\rho\) is density, \(x_i\) are displacement components, \(\sigma_{ij}\) is the Cauchy stress tensor, \(f_i\) are body forces, and \(p_i\) are surface tractions. This equation ensures that the work done by inertial forces, internal stresses, body forces, and surface forces is balanced, allowing for the tracking of energy conversion between kinetic energy (KE) and internal strain energy (SE) within the EV battery pack system.
The EV battery pack under investigation features a Cell-to-Pack (CTP) architecture. Its primary structural components include an upper cover, a lower tray (or box), a liquid cold plate, cross-beams, thermal interface adhesive, battery modules, and a bottom protective plate (or under-shield). The large-format battery modules are adhered to the cold plate using thermally conductive structural adhesive. A crucial component for bottom protection is the sacrificial bottom guard plate, often coated with a protective layer like PVC.
A high-fidelity finite element (FE) model of the complete EV battery pack was developed in Abaqus/Explicit to simulate the bottom crash event accurately. The modeling strategy prioritized the accurate representation of connections, which are critical load paths. Welds, bolts, and rivets were modeled using appropriate connector elements or tied constraints. Bonded contacts, implemented via the “Tie” constraint in Abaqus, were used to simulate adhesive connections, such as those between the battery cells, the thermal interface material, and the cold plate, as well as the coating on the bottom plate. General contact with a penalty friction formulation was defined for all other potential interactions between components. Particular attention was paid to meshing the impact zone, where a fine mesh was used to capture local deformation accurately. In contrast, non-critical regions were meshed more coarsely to optimize computational efficiency. The final model comprised approximately 1.05 million elements. The boundary conditions replicated the test fixture: all degrees of freedom were constrained at the EV battery pack’s mounting points, while the impactor was constrained in all directions except for translation along the impact (+Z) axis.
Accurate material modeling is essential for predicting plasticity, hardening, and energy absorption. Uniaxial tensile test data at different strain rates were obtained for the key metallic materials: the aluminum alloy of the cold plate (AL3003-H14), the high-strength steel of the structural members (HC340-590DP), and the aluminum of the bottom guard (AL3003-O). The true stress vs. plastic strain data were input into the simulation. Given the non-destructive nature of the standard test, sophisticated damage and failure models were not employed; the focus was on capturing plastic hardening up to the necking point. The strain-rate sensitivity was incorporated using the Cowper-Symonds model or by inputting multiple stress-strain curves. The key material parameters are summarized in the table below.
| Material Component | Material Designation | Density (kg/m³) | Young’s Modulus (GPa) | Poisson’s Ratio | Yield Stress (MPa) @ 0.2% offset | Ultimate Tensile Strength (MPa) | Failure Strain (Approx.) |
|---|---|---|---|---|---|---|---|
| Cold Plate | AL3003-H14 | 2730 | 69 | 0.33 | 145 | 155 | 0.24 |
| Structural Beams/Tray | HC340-590DP | 7850 | 210 | 0.3 | 340 | 590 | 0.22 |
| Bottom Guard Plate | AL3003-O (Annealed) | 2730 | 69 | 0.33 | 40 | 110 | >0.30 |
| Battery Cell Can | Aluminum Alloy | 2700 | 71 | 0.33 | 280 | 310 | 0.01 |
The impactor, with a mass \( m = 10 \, \text{kg} \), was assigned an initial velocity \( v_0 \) to achieve the prescribed kinetic energy of 150 J:
$$ KE_0 = \frac{1}{2} m v_0^2 = 150 \, \text{J} \quad \Rightarrow \quad v_0 = \sqrt{\frac{2 \times 150}{10}} \approx 5.477 \, \text{m/s} $$
This velocity was applied as an initial condition to the impactor node set. The simulation output included time histories of displacement, velocity, acceleration, contact forces, stress, strain, and various energy components (Kinetic Energy, Strain Energy, Artificial Strain Energy) for the entire EV battery pack system and its key components.
The dynamic response of the impactor provides fundamental insight into the crash pulse. The simulation results show that the impactor’s velocity decreases linearly to zero at approximately \( t = 2.94 \, \text{ms} \). The acceleration, and consequently the contact force, peaks earlier at around \( t = 2.17 \, \text{ms} \). The maximum contact force recorded between the impactor and the EV battery pack bottom structure was \( F_{max} = 27,299 \, \text{N} \). After reaching zero velocity, the impactor rebounds due to the elastic recovery of the deformed structure, resulting in negative velocity. The contact force decays to zero at about \( t = 4.25 \, \text{ms} \), signifying separation. The area under the force-displacement curve equals the initial kinetic energy of 150 J, validating the global energy balance of the simulation:
$$ W = \int F(x) \, dx \approx 150,000 \, \text{N·mm} = 150 \, \text{J} $$
The structural integrity of the impacted components is assessed based on equivalent plastic strain (PEEQ). A conservative criterion is used: the maximum PEEQ in any component should be less than the material’s plastic strain at the onset of necking (approximated from the failure strain in the table). The simulation results confirmed that the maximum PEEQ in the bottom guard plate, the cold plate, and the battery cell casing remained below their respective limits of 0.22, 0.24, and 0.01, indicating no risk of structural fracture or tear in this specific test scenario for the EV battery pack.
The intrusion depth is a direct metric for evaluating the protective performance of the EV battery pack’s bottom structure and the resulting threat to the battery cells. The simulation predicted the following maximum dynamic intrusions:
- Bottom Guard Plate: 8.20 mm
- Liquid Cold Plate: 6.15 mm
- Battery Cell Housing (Casing): 2.38 mm
The safety threshold for cell intrusion, determined by cell-level abuse testing, was 4.5 mm for this cell type. The predicted 2.38 mm intrusion suggests the cell safety margin was maintained in this test setup.
The energy distribution analysis offers profound insight into the crash energy management strategy of the EV battery pack. The initial 150 J of kinetic energy from the impactor is transformed and dissipated through several pathways throughout the event. The following table details the final energy distribution at the end of the impact event (\( t = 5 \, \text{ms} \)).
| Energy Component | Symbol | Energy Value (J) | Percentage of Total Input (150 J) | Description |
|---|---|---|---|---|
| Total Strain Energy (Pack) | SEtotal | 123.56 | 82.37% | Internal energy from elastic and plastic deformation of all pack components. |
| Kinetic Energy (Pack) | KEpack | 20.91 | 13.94% | Residual kinetic energy of the vibrating/deforming pack structure. |
| Strain Energy – Bottom Guard | SEguard | 48.55 | 32.36% | Energy absorbed by the plastic deformation of the sacrificial guard plate. |
| Strain Energy – Cold Plate | SEcold | 42.75 | 28.50% | Energy absorbed by the deformation of the liquid cold plate. |
| Strain Energy – Battery Cell Casing | SEcell | 30.15 | 20.10% | Energy absorbed by the deformation of the cell housings. |
| Artificial Strain Energy | ASE | 2.95 | 1.97% | Numerical energy used to control hourglass modes in reduced-integration elements. Kept below 10% of SEtotal. |
| Other Dissipated Energy | – | 2.58 | 1.72% | Energy dissipated through friction, damping, and other minor mechanisms. |
The key findings from the energy analysis are:
1. Energy Conservation: The sum of all final energy components is approximately 150 J, confirming the simulation’s numerical stability and adherence to physical laws.
2. Primary Energy Absorption: The three directly impacted components—the bottom guard, cold plate, and cell casings—collectively absorb \( 48.55 + 42.75 + 30.15 = 121.45 \, \text{J} \), which is 80.96% of the total input energy. This underscores their role as the primary energy-absorbing structure for the EV battery pack in this bottom impact scenario.
3. Energy Absorption Proportion vs. Intrusion: A strong correlation is observed. The component with the highest intrusion (bottom guard, 8.20 mm) also has the highest energy absorption proportion (32.36%). The cell casing, with the lowest intrusion (2.38 mm), absorbs 20.10% of the energy. This relationship suggests a design principle: increasing the energy absorption share of the external protective layers (guard plate, cold plate) can effectively reduce the energy transmitted to, and thus the intrusion into, the critical battery cells of the EV battery pack.
The temporal evolution of energy shows two distinct phases. In the initial phase (0-1.75 ms), the soft CR foam between the guard and cold plate is compressed, and the guard plate deforms plastically, absorbing most of the energy. In the second phase (1.75-2.94 ms), after the foam is fully compressed, the stiff cold plate and cell pack engage, leading to a sharp rise in their strain energy absorption and an increase in the global kinetic energy of the vibrating EV battery pack system.
A physical bottom crash test was conducted on the EV battery pack prototype following the standard. Post-test inspection revealed no cracks, coolant leaks, or electrolyte leakage. The intrusion depths were measured using calipers. The comparison between simulation predictions and experimental measurements is summarized below, demonstrating excellent correlation.
| Component | Simulated Intrusion (mm) | Experimental Intrusion (mm) | Simulation Accuracy |
|---|---|---|---|
| Bottom Guard Plate | 8.20 | 8.69 | 94.36% |
| Liquid Cold Plate | 6.15 | 6.08 | 98.80% |
| Battery Cell Housing | 2.38 | 2.46 | 96.78% |
The minor discrepancy for the bottom guard plate is likely attributable to simplifications in modeling the bonded PVC coating and the foam interlayer, which were approximated using tied contacts and linear elastic material models, respectively. The deformation shapes and patterns observed in the test specimens showed high visual agreement with the simulation contour plots, further validating the model’s ability to capture the complex dynamic response and failure mechanisms of the EV battery pack structure.
This study successfully demonstrates a high-fidelity finite element methodology for simulating the bottom crash test of an EV battery pack. The key conclusions are as follows:
- The developed explicit dynamics model accurately predicts the structural response, with intrusion depth accuracy exceeding 94% for all key components. It provides a reliable virtual tool for assessing the strength of the EV battery pack enclosure and evaluating the safety margin for battery cells against crushing.
- The analysis of the impactor’s dynamic response quantifies the severe load environment, identifying a maximum impact force of 27,299 N. This value serves as a critical reference for setting strength and stiffness targets in the mechanical design of the EV battery pack’s bottom protection system.
- The energy-based analysis reveals that 80.96% of the input crash energy is absorbed as strain energy by the three directly impacted layers (guard plate, cold plate, cell casing). The proportional relationship between a component’s energy absorption share and its intrusion depth establishes a clear design guideline: enhancing the energy absorption capability and proportion of the outer protective structures (e.g., using high-strength, high-ductility materials, sandwich panels, or multi-cell aluminum extrusions for the guard/cold plate) is an effective strategy to divert energy away from the battery cells, thereby reducing cell intrusion and improving the overall safety of the EV battery pack.
The findings underscore the importance of integrating detailed finite element analysis and energy management principles in the design and validation of crashworthy EV battery pack systems.