VALIDATED MULTIBODY MODEL FOR TRAIN CRASH ANALYSIS

The development of numerical simulation tools for train crashworthiness design requires their validation with reference crash scenarios similar, in nature, to the eventual collision conditions in which the new train designs have to be used. The modeling assumptions and the suitability of these tools can be verified using data feedback from experimental testing. In this work, a validated multibody-based model is presented for the design of train crashworthy components. In the proposed methodology, the moving components of a vehicle are described as sets of rigid bodies, with their relative motion constrained by kinematic joints. Rigid bodies connected by nonlinear force elements, which represent the lumped flexibility of the structural components, model the sub-structures that deform as a result of the train collisions. The characteristics of these nonlinear force elements represent the force-deformation curves of individual car-bodies extremities, couplers between car-bodies and stiffness of the suspensions springs. The wheel-rail contact is also represented by a model in which the normal and friction forces are present. The friction forces are described by Coulomb friction, which includes their dependency on static and dynamic friction coefficients. The contact forces between the end extremities of the colliding car-bodies, that model the longitudinal impact, include the action of anti-climber devices, which are designed to prevent sliding between the contacting buffers. The validated model is applied to the collision of two different trains, which have distinct specifications for the nonlinear force elements that represent the end extremities of the colliding car-bodies and their couplers. These two force-deformation curves correspond to the design specifications and to the experimental data acquired in a crash test. The validation of the model is discussed considering the deviations between the results of the test and the numerical tool with both design and experimental specifications. It is shown that the simulation of the model with the design specifications, characterized by elastic-perfectly plastic deformation curves for the structural elements, leads to results similar to those observed in the experimental test. When the force-deformation curves obtained experimentally are used to represent the structural elements the correlation between simulation and experimental test results increases significantly. (A)

  • Availability:
  • Corporate Authors:

    Woodhead Publishing, Limited

    Abington Hall, Abington
    Cambridge,   United Kingdom  CB1 6AH
  • Authors:
    • MILHO, J F
    • AMBROSIO, JAC
    • PEREIRA, MFOS
  • Publication Date: 2003

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 00962725
  • Record Type: Publication
  • Source Agency: Transport Research Laboratory
  • Files: ITRD
  • Created Date: Sep 3 2003 12:00AM