BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//EURODYN//Schedule//DE
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Berlin
X-LIC-LOCATION:Europe/Berlin
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:394@eurodyn2026.org
DTSTAMP:20261026T144037Z
DTSTART;TZID=Europe/Berlin:20260929T150000
DTEND;TZID=Europe/Berlin:20260929T152000
SUMMARY:Running safety assessment of trains crossing bridges using hybrid
  machine-learning based vehicle-structure algorithm
LOCATION:Room B305
DESCRIPTION:MS 13\n\nPresenter: Granato\, Eduardo\n\nAuthors: Granato\, E
 duardo\n\nChair: Ribeiro\, Diogo\n\nAbstract: Dynamic vehicle–structure 
 interaction analyses for train running safety assessment require accurat
 e modelling of wheel–rail contact behaviour\, as contact forces strongly
  influence the dynamic response of both the vehicle and the structure. R
 ealistic contact formulations involve highly nonlinear algorithms capabl
 e of capturing the complex spatial and temporal behaviour of wheel–rail 
 interaction\, although at significant computational cost and implementat
 ion complexity. To improve the efficiency of these analyses\, this paper
  presents a hybrid machine-learning-based vehicle–structure interaction 
 framework employing deep neural networks for the prediction of wheel–rai
 l contact parameters within a three-dimensional nonlinear dynamic model.
  In contrast to conventional feedforward approaches\, the proposed metho
 dology investigates convolutional and long short-term memory neural netw
 ork architectures to capture the sequential and nonlinear characteristic
 s of the contact behaviour during train crossing of railway bridges. The
  models are trained using numerically generated datasets obtained from a
  nonlinear three-dimensional vehicle–structure interaction algorithm con
 sidering multiple track irregularity profiles and train speeds represent
 ative of operational and running safety conditions. The developed networ
 ks are designed to learn the complex spatial-temporal behaviour of the c
 oupled dynamic system while maintaining reduced model complexity and imp
 roved computational efficiency. Numerical results demonstrate that the p
 roposed CNN–LSTM framework is capable of accurately reproducing the nonl
 inear wheel–rail contact behaviour and contact parameters throughout the
  dynamic interaction process. The study highlights the potential of adva
 nced deep-learning architectures for efficient running safety assessment
  of trains crossing bridges using hybrid vehicle–structure interaction m
 ethodologies.
URL:https://eurodyn2026.org/submission/394
END:VEVENT
END:VCALENDAR