⚡ electrical engineering

Geometric control and multi-case crashworthiness optimization of functionally graded cross-section thin-walled conical tubes

This paper proposes and optimizes functionally graded cross-section thin-walled conical tubes with power-law diameter variations to achieve superior crashworthiness and peak-load control across multiple impact angles through finite element modeling and multi-objective NSGA-II optimization.

Original authors: Zhichao Xu, Xinrui Shao, Yuxin Chen, Jiangchen Li, Zhichao Li

Published 2026-10-22

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Zhichao Xu, Xinrui Shao, Yuxin Chen, Jiangchen Li, Zhichao Li

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

In the world of vehicle safety, engineers face a constant, delicate balancing act. They must design structures that are light enough to save fuel and reduce emissions, yet strong enough to protect passengers when a crash occurs. The key to this balance lies in how a structure absorbs energy. When a car hits an obstacle, thin metal tubes often crumple to soak up the force of the impact, preventing that energy from reaching the people inside. However, simply making a tube stronger or thicker is not the answer; a tube that is too strong might not crumple at all, transferring the full shock to the passengers, while one that is too weak might collapse too easily, offering little protection. The goal is to find a shape that collapses in a controlled, predictable way, absorbing as much energy as possible while keeping the initial jolt to a manageable level.

For decades, engineers have relied on tubes that taper in a straight, linear line, like a cone. But a new study suggests that the path to better safety might lie in changing the rate at which that tube widens. Researchers from universities in China have explored a concept called a "functionally graded cross-section." Instead of a straight taper, they designed tubes where the diameter changes along the length according to a specific, curved rule. Imagine a tube that stays narrow for a long stretch and then suddenly flares out, or one that flares out quickly at the start and then levels off. By testing thousands of these variations on a computer, the team discovered that the most effective designs are not the straight cones, nor the extreme curves, but a specific middle ground where the widening happens gradually but shifts toward the back of the tube.

The researchers focused on thin-walled conical tubes made of aluminum, a material common in automotive frames. They created a digital model of these tubes and subjected them to virtual crashes at different angles, simulating everything from a straight-on head-on collision to impacts where the car hits a barrier at a slant. In the real world, crashes rarely happen perfectly straight; a car might hit a pole or another vehicle at an angle, which changes how the metal bends and folds. The team wanted to see if a tube designed for a straight hit would still work well when hit from the side. They varied three main things in their simulations: how thick the metal walls were, how strong the metal was, and the specific curve of the tube's widening.

To understand how these tubes behave, the team looked at how they crumpled. When a standard straight cone is crushed, it tends to fold in a uniform pattern. However, the new tubes with curved widening profiles behaved differently. Some developed tight, tight wrinkles near the impact point, while others showed signs of bending sideways or folding unevenly. The researchers found that the shape of the tube dictated where these folds started and how the metal spread out. If the tube widened too quickly at the front, it created a different kind of stress than if it widened slowly at the start and then rapidly at the end. The goal was to find a shape that encouraged the tube to fold neatly, absorbing energy efficiently without creating a sudden, dangerous spike in force.

Using advanced computer algorithms, the team searched for the "sweet spot" among millions of possible combinations. They were looking for a design that would absorb the maximum amount of energy per unit of weight while keeping the initial force of the impact below a specific safety limit. They tested these designs under four different impact angles: straight on, and at ten, twenty, and thirty degrees off-center. A common mistake in safety design is to optimize a tube for a straight hit and assume it will perform just as well at an angle. The study showed this is not true. A tube that is excellent at absorbing energy in a straight crash might perform significantly worse when hit from the side, sometimes losing more than half its effectiveness.

The results pointed to a specific range of designs that outperformed the traditional straight cones. The best performers were those where the tube's diameter stayed relatively small for the first part of its length and then expanded more rapidly toward the back. Specifically, the most effective designs were those where the point at which the tube reached half of its total width increase was located between 70% and 83% of the way down the tube's length. In simpler terms, the tube should stay narrow for most of its journey and then flare out significantly near the end. This specific geometry allowed the tube to manage the crushing force more smoothly, offering a better trade-off between absorbing energy and controlling the peak force felt by the passengers.

Crucially, the study highlighted that a design optimized only for a straight crash does not guarantee safety in a real-world accident, which often involves angled impacts. When the researchers weighed the performance across all four angles, giving more importance to the straight and slightly angled hits, the same "rear-heavy" expansion pattern emerged as the winner. This suggests that by carefully controlling the curve of the tube's widening, engineers can create a structure that is robust enough to handle the unpredictability of real crashes. The findings do not claim to have solved every problem in crash safety, nor do they suggest that every tube with this shape will be perfect. Instead, they provide a clear, data-driven guide for engineers on how to shape these tubes to get the most protection out of the least amount of material, ensuring that when a crash happens, the structure works exactly as intended.

Technical Summary: Geometric Control and Multi-Case Crashworthiness Optimization of FGCS Thin-Walled Conical Tubes

Problem Statement
Thin-walled energy-absorbing structures face a fundamental design challenge: reconciling lightweight design, high energy absorption, and strict peak-load control. While existing designs often focus on material selection, wall thickness, or cross-sectional shape, the axial distribution of the cross-sectional dimension remains an underutilized design degree of freedom. Furthermore, performance evaluation is frequently limited to ideal axial loading, neglecting the reality of oblique impacts where energy-absorption advantages may not be preserved. This paper addresses the need to balance energy absorption efficiency, mass utilization, and peak load control by investigating Functionally Graded Cross-Section (FGCS) thin-walled conical tubes. The core objective is to determine how a candidate design can be obtained through the joint selection of geometric gradient, wall thickness, and material strength under multi-angle loading conditions.

Methodology
The study employs a combined approach of geometric parameterization, finite element analysis (FEA), surrogate modeling, and multi-objective optimization:

Geometric Parameterization: The FGCS tube is defined by a power-law function describing the axial variation of the diameter $D(x) = D_s + (D_l - D_s)(x/L)^m$ , where $m$ is a dimensionless gradient parameter. Unlike linear conical tubes ( $m=1$ ), this formulation allows for tunable nonlinear expansion rates. The parameter $m$ is interpreted geometrically: $m < 1$ concentrates expansion near the impact end, while $m > 1$ shifts expansion toward the support end. A characteristic position $x(p)$ is derived to map abstract $m$ values to physical axial locations (e.g., the position where half the diameter increment is completed).
Finite Element Modeling: Dynamic crushing simulations were conducted using LS-DYNA with an elastic-plastic AL6060-T4 material model. The model was validated against literature benchmarks (Mamalis et al.), showing crush-distance errors within ~6% and total energy errors within ~17%. Four loading angles were analyzed: 0° (axial), 10°, 20°, and 30°.
Surrogate Modeling: To facilitate optimization, 16 partitioned polynomial response surfaces were constructed using wall thickness ( $t$ ), yield strength ( $\sigma_y$ ), and gradient parameter ( $m$ ) as design variables. These surfaces predict Specific Energy Absorption (SEA) and peak crushing forces ( $F_{init}$ and $F_{global}$ ) for each loading angle. The models demonstrated high fidelity, with coefficients of determination ( $R^2$ ) ranging from 0.957 to 0.995.
Multi-Case Optimization: The Non-dominated Sorting Genetic Algorithm II (NSGA-II) was used to solve a multi-objective problem. The objectives were to maximize weighted SEA and minimize the initial peak reaction, subject to a global peak-load constraint ( $F_{global} \le 90$ kN). The weighted SEA ( $SEAw$ ) incorporated case-specific weights (0.4 for 0°, 0.3 for 10°, 0.2 for 20°, 0.1 for 30°) to reflect design preferences, while the peak envelope ( $F_{env}$ ) ensured no single angle exceeded safety limits.

Key Contributions

Power-Law Parameterization: The paper establishes a continuous power-law expression for equal-wall-thickness thin-walled tubes, where a single gradient parameter ( $m$ ) adjusts the axial distribution of the cross-section. This treats the "unfolding law" of the cross-section as an independent design variable, distinct from end dimensions or wall thickness.
Geometric-Deformation Interpretation: An interpretive link is constructed between the geometric distribution and the crushing response. The study maps optimization-derived parameter intervals to physical axial positions, explaining how different $m$ values influence local folding, lateral spreading, and asymmetric deformation.
Multi-Angle Optimization Framework: A parameter-selection framework is formed that coordinates multi-angle energy absorption with peak load control. It demonstrates that optimizing solely for axial performance is insufficient, as higher axial SEA does not guarantee a superior ranking in a comprehensive multi-angle evaluation.

Results

Deformation Morphology: Different gradient parameters result in distinct deformation modes. Low $m$ values tend to form tight wrinkles, while high $m$ values can induce asymmetric deformation and local folding, though these modes are not mutually exclusive.
Optimization Outcomes: The Pareto fronts for the two gradient partitions ( $0 \le m \le 1$ and $1 < m \le 10$ ) reveal that the larger-gradient partition generally offers a more favorable trade-off between comprehensive SEA and peak load.
Optimal Parameter Range: Under the specified constraints and weights, the superior candidate parameters are located in the range $m \approx 1.95 - 3.85$ . Geometrically, this corresponds to a "half-increment position" (where the diameter reaches 75 mm) at approximately 0.701 to 0.835 of the tube length. This indicates a configuration where the front stage maintains a smaller cross-section, and the rear stage completes a larger proportion of the size expansion.
Ranking Discrepancy: Analysis of validation samples showed that axial SEA rankings do not always align with comprehensive rankings. For instance, a sample with higher axial SEA could rank lower in the comprehensive evaluation due to significant performance drops at 20° and 30°.

Significance and Claims
The paper claims to provide a design-analysis basis for selecting FGCS energy absorbers by integrating geometric derivation, morphological observation, and performance comparison. It emphasizes that the identified optimal range ( $m \approx 1.95 - 3.85$ ) is a candidate interval under specific numerical conditions (fixed dimensions, material, impact speed, and load constraints) rather than a universal rule for all configurations.

The authors explicitly state that their findings are limited to the defined design space and do not constitute a proof of stable axisymmetric folding for all structures within the interval. Furthermore, the weighted analysis of four deterministic angles is not equated with completed uncertainty robust optimization or engineering physical validation. The study serves to illustrate the necessity of multi-case optimization and provides a coherent analytical path from geometric description to candidate configuration selection, avoiding the mere listing of optimal indices without physical interpretation.

Technical Summary: Geometric Control and Multi-Case Crashworthiness Optimization of FGCS Thin-Walled Conical Tubes

More like this