Araliya Mosleh - CONSTRUCT – LESE, Faculty of Engineering, University of Porto, Porto, Portugal
Mohammadreza Mohammadi - CONSTRUCT – LESE, Faculty of Engineering, University of Porto, Porto, Portugal
Cecilia Vale - CONSTRUCT – LESE, Faculty of Engineering, University of Porto, Porto, Portugal
Diogo Ribeiro - CONSTRUCT – LESE, School of Engineering, Polytechnic of Porto, Porto, Portugal
Pedro Montenegro - CONSTRUCT – LESE, Faculty of Engineering, University of Porto, Porto, Portugal
Andreia Meixedo - CONSTRUCT – LESE, Faculty of Engineering, University of Porto, Porto, Portugal

As the rail sector plays a significant role in society, traffic and maintenance costs are critical aspects for railway managers and operators. Although current track wayside monitoring systems detect geometric defects in wheels, such as flats, they do not categorize them in terms of severity. To overcome this limitation, the WAY۴SafeRail project aims to develop techniques to improve rail safety by determining the damage conditions of train wheels. As a preliminary step in applying Artificial Intelligence Techniques, the present paper presents part of the research developed in the WAY۴SafeRail project, particularly the numerical simulations of wheel defects. The proposed methodology has shown to be a reliable and cost-effective method for identifying wheel defects.

Artificial intelligence; damage detection; wheel flat; train-track dynamic interaction; wayside condition monitoring