Safety Study of High-Speed Collisions between Trains and Live Intruder

Hai Zhang, Gengzhe Fu, Yongzhang Su, Yixin Yue, Wei Zhu, Chenyu Zhang, Yuxiang LuSensors 2022 · DOI

TL;DR

This paper introduces a novel 3D finite element model of live intruders (specifically pigs and cows) to simulate high-speed train collisions. By validating this model against existing cadaver test data and then simulating train-intruder impacts, the study demonstrates that current train safety standards (EN15227) are met even at 110 km/h, with critical safety indicators remaining within acceptable limits.

What Problem Does This Paper Address?

As train speeds increase, the likelihood and severity of collisions with unexpected obstacles, including live animals, also rise. While inter-train and train-building collisions have been extensively studied, collisions with live intruders pose unique challenges due to the complex biomechanics involved. Existing research often focuses on passenger injury or uses simplified models. This paper addresses the critical need to understand and quantify the safety implications of high-speed train collisions with live animals, ensuring the structural integrity of the train and the safety of its operation, rather than just the fate of the intruder.

Methods

The core of this research lies in the development and validation of a sophisticated 3D finite element (FE) model of live intruders.

  • Intruder Modeling: A novel approach was used to create a "laminated" FE model of a live intruder. This involved:
  • 3D Scanning: Physical 3D scanning of animal models (pig and cow) to capture their external geometry.
  • Layered Structure: The model was constructed with multiple layers simulating skin, bone, and internal organs, each with distinct material properties.
  • Internal Filling: Internal organs were represented by mass points and viscoelastic material properties to better simulate their behavior under impact.
  • Ligament Simulation: Beam elements were used to simulate ligaments, preventing limb detachment during high-speed impacts.
  • Model Validation: The accuracy of the intruder FE model was verified through a pendulum side-impact simulation. The force-time and force-displacement curves generated by this simulation were compared to experimental data from Viano's 1989 cadaver pendulum collision tests.
  • Train Collision Simulation: The validated intruder models were then integrated with a detailed FE model of a train. High-speed collision simulations were performed using LS-DYNA software, considering both frontal and offset impact scenarios at 110 km/h.
  • Safety Analysis: Key safety indicators were analyzed, including:
  • Maximum train acceleration.
  • Maximum wheel pair lift.
  • Cowcatcher deformation and force.
  • Coupler and anti-creeper forces. These results were then compared against the EN15227 standard for train crashworthiness.

Datasets

  • Cadaver Pendulum Collision Test Data: The study references and compares its simulation results against the experimental data from Viano's 1989 cadaver pendulum collision test. This serves as a crucial benchmark for validating the accuracy of the developed intruder FE model.
  • No new benchmark datasets were introduced. The study relies on existing standards and experimental data for validation and comparison.

Key Insights and Ideas

  • Biomechanical Fidelity Matters: The paper highlights that a detailed, layered FE model of the intruder, accounting for different tissue types and their material properties, is crucial for accurately simulating collision dynamics. Simply using a "mass block" model is insufficient for capturing energy absorption and deformation characteristics.
  • Validation is Key: The successful validation of the intruder model against cadaver test data provides strong confidence in its ability to represent real-world impacts. This is a critical step before applying it to more complex train collision scenarios.
  • Frontal vs. Offset Collisions: Frontal collisions generally pose a higher risk, leading to greater train acceleration and forces, compared to offset collisions. However, the specific impact location and the intruder's mass can significantly influence these outcomes.
  • Cowcatcher Resilience: The cowcatcher, a critical component for deflecting obstacles, demonstrated sufficient resilience. Its deformation and the forces it experienced were well within the EN15227 standard limits, even in more severe frontal collisions.
  • Train Safety Margins: Despite the high-speed impacts (110 km/h), all analyzed safety indicators for the train (acceleration, wheel lift, cowcatcher performance) remained below the stringent EN15227 requirements, suggesting that current train designs offer a robust safety margin against such events.

Key Results

  • Model Validation: The FE model's pendulum impact simulations showed good agreement with Viano's cadaver test results, verifying its accuracy.
  • Train Acceleration: Maximum train acceleration was observed in frontal collisions with the heavier cow model (0.117 g), which is significantly lower than the EN15227 limit of 5 g.
  • Wheel Pair Lift: The maximum wheel pair lift was minimal (0.4 mm for the pig frontal collision), far below the EN15227 requirement of 21 mm.
  • Cowcatcher Performance:
    • Maximum cowcatcher deformation was 9.2 mm (pig offset collision), with a reduction of about 37.3 mm per 5m length, well within standards.
  • Cowcatcher forces reached a maximum of 140 kN (pig frontal collision), which is lower than the EN15227 requirement of 150 kN.
  • Coupler and Anti-Creeper: Forces on the coupler and anti-creeper were also within acceptable limits and did not cause deformation or damage to these components.

Limitations and Future Work

  • Limited Intruder Types: The study focused on pigs and cows. The authors acknowledge that other living intruders could lead to different collision consequences, suggesting this as a direction for future research.
  • Collision Location Uncertainty: While frontal and offset collisions were simulated, the initial position of the intruder on the track can vary. The paper notes that animals might sprint laterally rather than directly onto the track, and future work could explore these varied collision locations.
  • Animal Behavior: The study assumes a stationary intruder. Future research could incorporate more dynamic animal behavior during the collision.
  • Further Refinement: While the model is considered accurate, further improvements in building multiple organism models and considering other collision scenarios are suggested.

Summary updated

Abstract

To investigate the safety of train collisions with live intruders under high-speed operation, a new 3D finite element laminated model of live intruder filling was constructed based on reconstruction using physical 3D scanning, with three outer layers of the model simulating the skin, three inner layers simulating bone, and internal filling simulating internal organs. The model was simulated in LS-DYNA with pendulum side collision, and the force-time and force-displacement curves of the collision between the pendulum and the living intruder were obtained, which were consistent with the curve trend of the results of the cadaver pendulum collision test by Viano in 1989, and the accuracy of the finite element model of the intruder was verified. Through the simulation calculation of high-speed collision between the train and two kinds of living intrusions, the maximum acceleration of the train body, the maximum lifting of the wheel pair, the deformation of the cowcatcher, and the maximum central load on the cowcatcher during the collision can be obtained. The results of the study show that at a collision speed of 110 km/h and different collision positions, the collision risk factor between the train and heavier organisms is relatively high, and the risk arising from frontal collisions is generally greater than that of offset collisions; despite this, all the indicators such as the maximum acceleration of the train, the maximum lift of the wheel pairs, the reduction in the length of the cowcatcher discharge per 5 m of space, and the maximum central load borne by the cowcatcher discharge are lower than the EN15227 standard. Additionally, the safety of the train is not affected and the components can work reliably.
BibTeX
@inproceedings{zhang2022safety,
  author = {Hai Zhang and Gengzhe Fu and Yongzhang Su and Yixin Yue and Wei Zhu and Chenyu Zhang and Yuxiang Lu},
  title = {{Safety Study of High-Speed Collisions between Trains and Live Intruder}},
  booktitle = {Sensors},
  year = {2022},
  doi = {10.3390/s22228824},
}
Figure