Machine-learning based prediction of crash response of tubular structures

Emmanouil Sakaridis, Nikolaos Karathanasopoulos, Dirk Mohr

Research output: Contribution to journalArticlepeer-review


This paper proposes a machine learning based methodology for predicting the buckling response of tubular structures. An extensive dataset of force-time curves is generated using a calibrated finite element model within a parametric space where buckling response is highly non-linear. Based on a fully connected neural network template, the machine learning hyper-parameters are determined and the resulting model is evaluated on a separate test set, with regard to maximum and average load and energy absorption errors. This evaluation shows a non-random error distribution which can be correlated with the physical properties of the structural collapse. To validate this assumption, a similar error analysis is conducted between finite element simulations with varying geometric imperfections. Evaluation of imperfection sensitivity reveals a similar error distribution and comparison of individual curves shows that errors made by the neural network model have a physical interpretation. These results indicate that the proposed machine learning based approach is capable of predicting the crushing response with a level of accuracy comparable to the errors that would be caused by a minor change in geometric imperfection.

Original languageEnglish (US)
Article number104240
JournalInternational Journal of Impact Engineering
StatePublished - Aug 2022


  • Artificial neural network
  • Buckling transition
  • Crashworthiness
  • Imperfection sensitivity
  • Machine learning

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Automotive Engineering
  • Aerospace Engineering
  • Safety, Risk, Reliability and Quality
  • Ocean Engineering
  • Mechanics of Materials
  • Mechanical Engineering


Dive into the research topics of 'Machine-learning based prediction of crash response of tubular structures'. Together they form a unique fingerprint.

Cite this