Avaliação da Condição da Via Permanente usando Dados de Dinâmica de Veículos Ferroviários: Uma Abordagem de Aprendizado de Máquina

  • Pedro H. O. Silva Programa de Pós-Graduação em Engenharia Elétrica, Universidade Federal de Juiz de Fora, MG
  • Raphael D. Marotta VLI Multimodal e Programa de Pós-Graduação em Engenharia de Transportes, Instituto Militar de Engenharia (IME), RJ
  • Augusto S. Cerqueira Programa de Pós-Graduação em Engenharia Elétrica, Universidade Federal de Juiz de Fora, MG
  • Erivelton G. Nepomuceno Centre for Ocean Energy Research Department of Electronic Engineering Maynooth University
  • Luiz A. S. Lopes Programa de Pós-Graduação em Engenharia de Transportes, Instituto Militar de Engenharia (IME), RJ
Keywords: rail defects, track quality, machine learning, acceleration data, wheel–rail contact dynamic forces, railroad dynamics

Abstract

The safety and stability of freight train operations are closely related to the wheel–rail contact dynamic forces. To evaluate and monitor wheel-rail dynamic forces based on the accelerations of the wagon components, a methodology based on data processing and machine learning methods was developed. The experimental results show adequate performance, identifying critical values that relate to the condition of the railway. The model obtained helps railway specialists in the assessment of the quality of the railway, showing georeferenced indicators.
Published
2022-10-19
Section
Articles