In the rapidly evolving automotive industry, the shift towards electrification has placed immense focus on the development of efficient and safe Electric Vehicle (EV) battery packs. The structural integrity of the EV battery pack, particularly its underbody, is paramount for vehicle safety and performance. This study focuses on the simulation and optimization of an aluminum alloy EV battery pack subjected to a bottom support or “stone strike” condition, a critical load case that mimics impact from road debris. The primary objective is to achieve lightweight design while ensuring that the battery modules remain protected from deformation-induced contact, thereby preventing potential thermal runaway or short-circuit risks. Through detailed finite element analysis, Design of Experiments (DOE), and experimental validation, this work establishes a methodology for optimizing the thickness parameters of the aluminum alloy bottom plate, leading to a significant mass reduction without compromising safety.
The structural configuration of the EV battery pack analyzed herein consists of three main components: an upper cover, multiple battery modules, and a lower housing. The lower housing is further divided into a perimeter frame and a bottom plate. The frame is constructed from welded aluminum extrusions (AA6061-T6), while the bottom plate is a single-double layer aluminum extruded panel made from AA6005-T6. The upper cover utilizes Sheet Molding Compound (SMC). The battery modules comprise lithium-ion cells and aluminum end plates (AA6063-T6). The material properties critical for the simulation are summarized in Table 1. The finite element model was developed using Altair HyperMesh, where shell elements (5 mm base size) were used for the housing and cover, and hexahedral solid elements (10 mm base size) for the battery modules. Connections, such as friction stir welds between the frame and bottom plate, were modeled using shared nodes or appropriate contact definitions.
| Component | Material | Density (kg/m³) | Poisson’s Ratio | Elastic Modulus (MPa) | Tensile Strength (MPa) | Elongation (%) |
|---|---|---|---|---|---|---|
| Lower Housing Frame | AA6061-T6 | 2700 | 0.3 | 70,000 | 320 | 8 |
| Upper Cover | SMC | 1820 | 0.3 | 10,500 | 90 | 1 |
| Bottom Plate | AA6005-T6 | 2700 | 0.3 | 70,000 | 292 | 9.5 |
| Module End Plate | AA6063-T6 | 2700 | 0.3 | 69,000 | 240 | 12 |
The bottom support simulation replicates a scenario where the EV battery pack encounters a spherical road obstacle. A rigid sphere with a diameter of 150 mm is positioned at the most critical location on the pack’s underside, as determined from packaging studies. A vertical upward force of 25 kN is applied to the sphere to simulate the peak impact load. The boundary conditions constrain all degrees of freedom at the EV battery pack’s mounting points to the vehicle body. Self-contact is defined for the pack itself, and a surface-to-surface contact (TYPE7 in RADIOSS) is established between the pack and the rigid sphere. Gravitational acceleration (9.81 m/s²) is applied to the entire model. The nonlinear quasi-static analysis is performed using the RADIOSS solver to capture plastic deformation and contact behavior.
The performance metric for this EV battery pack under bottom support is the clearance between the deforming bottom plate and the battery modules. The design requirement mandates that under the 25 kN load, the bottom plate must not contact any module. The initial design featured a uniform bottom plate thickness of 3.0 mm. Simulation results indicated a post-impact clearance of 2.08 mm, which satisfied the requirement but with a considerable safety margin, suggesting potential for lightweighting. A preliminary reduction of the thickness to 2.5 mm resulted in a clearance of 1.13 mm, which was less than half the plate thickness (1.25 mm), indicating contact and thus failure of the requirement. This highlighted the need for a systematic optimization approach rather than a uniform thickness reduction.
To efficiently explore the design space and identify the most influential parameters on the EV battery pack bottom plate’s stiffness, a Design of Experiments (DOE) study was conducted using Altair HyperStudy. The bottom plate’s cross-sectional geometry, with fixed rib spacing, is primarily defined by three thickness parameters: the upper plate thickness ($t_u$), the lower plate thickness ($t_l$), and the rib thickness ($t_r$). A Resolution IV fractional factorial design was selected to screen these factors, allowing for the estimation of main effects while confounding two-factor interactions with higher-order interactions, which is a computationally efficient approach for engineering screening. Each factor was varied across five levels: ±20% and ±10% around a nominal value of 3.0 mm, plus the nominal itself. This created a matrix of 25 design runs. The responses of interest were the maximum displacement at the impact point ($\delta_{max}$) and the energy absorbed ($E_{abs}$) by the bottom plate structure up to the specified load. The mathematical relationship for stiffness in a simplified sense can be expressed as the ratio of force to displacement: $$K \approx \frac{P}{\delta}$$ where $K$ is the local structural stiffness, $P$ is the applied force (25 kN), and $\delta$ is the displacement. The goal is to maximize $K$ (minimize $\delta$) for a given mass.
The DOE results, including the input parameters and output responses for all 25 runs, are consolidated in Table 2. Subsequent analysis of the main effects plots (Figure 1) revealed the sensitivity of each thickness parameter on the responses. The slope of the effect line indicates the degree of influence. For the maximum displacement $\delta_{max}$, the lower plate thickness ($t_l$) exhibited the steepest negative slope, meaning that increasing $t_l$ most effectively reduces displacement, thereby increasing stiffness. The relationship can be qualitatively described by the bending stiffness of a plate, which is proportional to the cube of its thickness for a given width and material: $$D \propto \frac{E t^3}{12(1-\nu^2)}$$ where $D$ is the flexural rigidity, $E$ is Young’s modulus, $t$ is thickness, and $\nu$ is Poisson’s ratio. Although the bottom plate is a complex extruded profile, this principle explains why thickness variations have a nonlinear, powerful impact. The upper plate thickness ($t_u$) showed a moderate effect, while the rib thickness ($t_r$) had the least influence on displacement. For energy absorption, $t_u$ had a slightly more pronounced effect, but overall, $t_l$ remained the dominant factor for structural rigidity under this specific loading condition for the EV battery pack.
| Run # | $t_u$ (mm) | $t_l$ (mm) | $t_r$ (mm) | $\delta_{max}$ (mm) | $E_{abs}$ (mJ) ×10³ |
|---|---|---|---|---|---|
| 1 | 2.00 | 2.00 | 2.00 | 18.69 | 171.38 |
| 2 | 2.00 | 2.25 | 2.25 | 17.50 | 163.99 |
| 3 | 2.00 | 2.50 | 2.50 | 16.42 | 157.02 |
| 4 | 2.00 | 2.75 | 2.75 | 15.49 | 152.33 |
| 5 | 2.00 | 3.00 | 3.00 | 14.65 | 148.87 |
| 6 | 2.25 | 2.00 | 2.25 | 18.04 | 165.08 |
| 7 | 2.25 | 2.25 | 2.50 | 16.88 | 157.76 |
| 8 | 2.25 | 2.50 | 3.00 | 15.62 | 148.08 |
| 9 | 2.25 | 2.75 | 2.00 | 15.98 | 156.34 |
| 10 | 2.25 | 3.00 | 2.75 | 14.49 | 146.74 |
| 11 | 2.50 | 2.00 | 2.50 | 17.44 | 158.80 |
| 12 | 2.50 | 2.25 | 3.00 | 16.16 | 149.72 |
| 13 | 2.50 | 2.50 | 2.75 | 15.43 | 145.54 |
| 14 | 2.50 | 2.75 | 2.25 | 15.35 | 150.04 |
| 15 | 2.50 | 3.00 | 2.00 | 15.15 | 153.33 |
| 16 | 2.75 | 2.00 | 2.75 | 16.89 | 152.41 |
| 17 | 2.75 | 2.25 | 2.00 | 16.84 | 155.95 |
| 18 | 2.75 | 2.50 | 2.25 | 15.71 | 148.47 |
| 19 | 2.75 | 2.75 | 3.00 | 14.22 | 136.74 |
| 20 | 2.75 | 3.00 | 2.50 | 14.18 | 142.87 |
| 21 | 3.00 | 2.00 | 3.00 | 16.37 | 146.02 |
| 22 | 3.00 | 2.25 | 2.75 | 15.73 | 144.44 |
| 23 | 3.00 | 2.50 | 2.00 | 15.84 | 149.28 |
| 24 | 3.00 | 2.75 | 2.50 | 14.49 | 140.15 |
| 25 | 3.00 | 3.00 | 2.25 | 14.27 | 143.47 |
The accuracy of the finite element modeling approach for the aluminum alloy used in the EV battery pack was verified through physical testing. A section of the bottom plate extruded profile was subjected to a three-point bending test, which simulates the local bending induced by the spherical impactor. The test setup involved supporting the plate at two ends and applying a quasi-static load via a semi-cylindrical indenter at the mid-span, corresponding to a critical rib cavity. The same material model, specifically a piecewise linear plasticity model (MAT36 in RADIOSS) defined by the true stress-strain curve of AA6005-T6 (Figure 2), was used in the simulation of this isolated test. The force-displacement curves from two physical tests and the simulation are compared in Figure 3. The curves show good agreement, with the simulation slightly underestimating the force in the mid-displacement range (7-13 mm) and overestimating it later. The maximum force from the tests was 11.08 kN and 11.23 kN at displacements of ~10.9 mm, while the simulation predicted a maximum force of 10.90 kN at 11.70 mm. The relative errors in peak force (2.94%) and corresponding displacement (7.3%) are both below 10%, validating the fidelity of the material model and the finite element approach for the EV battery pack component. Furthermore, the failure location—cracking at the mid-span of the rib cavity—was accurately predicted by the simulation (Figure 4).
Based on the DOE sensitivity analysis, which identified the lower plate thickness ($t_l$) as the most critical parameter for the EV battery pack’s bottom plate stiffness, an optimized thickness set was proposed. To achieve lightweighting while meeting the no-contact requirement, the strategy was to prioritize $t_l$, slightly reduce $t_u$, and adjust $t_r$ considering manufacturability. Aluminum extrusion processes can face challenges with maintaining flatness as wall thickness decreases; therefore, a balanced approach is necessary. The selected optimized thicknesses are: $t_u = 2.50$ mm, $t_l = 2.75$ mm, and $t_r = 2.50$ mm. This configuration was then analyzed in the full EV battery pack bottom support simulation. The results showed a maximum displacement that brought the mid-plane of the bottom plate to within 1.32 mm of the module. Accounting for the upper plate thickness, the final clearance between the physical bottom plate surface and the module was 0.07 mm, as calculated by: $$\text{Clearance} = \text{Mid-plane Gap} – \frac{t_u}{2} = 1.32 – \frac{2.50}{2} = 0.07 \text{ mm}$$ This minimal clearance satisfies the requirement of no contact under the 25 kN load. The mass comparison between the initial and optimized EV battery pack bottom plate is presented in Table 3. The optimization yielded a mass reduction of approximately 10%, a significant achievement for EV battery pack lightweighting, which directly contributes to extended vehicle range. The relationship between ball displacement and clearance is nearly linear in the elastic-plastic regime studied, confirming that controlled parameter adjustment via DOE is effective.
| Design Scheme | Bottom Plate Mass (kg) | Mass Reduction |
|---|---|---|
| Initial (all 3.0 mm) | 36.55 | Baseline |
| Optimized ($t_u$=2.5, $t_l$=2.75, $t_r$=2.5 mm) | 32.87 | ~10.1% |
The success of this optimization hinges on understanding the complex structural behavior of the EV battery pack under localized impact. The stress distribution follows a pattern where the maximum principal stress concentrates in the lower plate and the rib webs directly under the impactor. The yield criterion for the aluminum alloy can be expressed using the von Mises stress ($\sigma_{vm}$), which for a state of plane stress is: $$\sigma_{vm} = \sqrt{\sigma_{xx}^2 + \sigma_{yy}^2 – \sigma_{xx}\sigma_{yy} + 3\tau_{xy}^2}$$ When $\sigma_{vm}$ exceeds the material’s yield strength (292 MPa for AA6005-T6), plastic deformation begins, permanently deforming the EV battery pack’s protective structure. The energy absorbed during this process, a key response in the DOE, is the area under the force-displacement curve and is integral to assessing crashworthiness: $$E_{abs} = \int_{0}^{\delta_{max}} P(\delta) \, d\delta$$ The optimized design ensures that even after yielding, the deformation is contained within limits that preserve the vital clearance.
In conclusion, this comprehensive study demonstrates a robust methodology for the simulation-driven design and lightweight optimization of an aluminum alloy EV battery pack subjected to bottom support loads. The integration of detailed FEA, systematic Design of Experiments for factor screening, and physical validation testing provides a high-confidence engineering framework. The findings clearly establish that for this specific extruded profile and loading condition, the thickness of the lower plate is the most significant design variable influencing stiffness. By strategically adjusting the thickness parameters—increasing the critical lower plate thickness while reducing others—a 10% mass reduction was achieved while maintaining a positive, albeit minimal, clearance between the deforming bottom plate and the battery modules. This outcome underscores the importance of targeted, performance-based optimization over uniform material removal in the pursuit of lightweight EV battery packs. The methodologies and insights presented here can be directly applied to the design and development of future generations of EV battery packs, contributing to safer, lighter, and more efficient electric vehicles. Future work could explore more complex multi-objective optimization including additional load cases like side impact or crush, and incorporate advanced materials such as carbon fiber composites or hybrid structures to further push the boundaries of EV battery pack performance and lightweighting.