Optimization of heavy haul railway wheel profile based on rolling contact fatigue and wear performance

Abstract

This paper presents a methodology to develop optimal railway wheel profiles based on their rolling contact fatigue (RCF) and wear performance. The NSGA-II multi-objective optimization algorithm was employed to minimize two objective functions: the wear index and the fatigue index while satisfying objective constraints. The commercial multibody software SIMPACK® was used for modeling the railway vehicle's dynamics and macrogeometry of a Brazilian heavy haul railway. The objective functions and constraints are obtained from the dynamic simulation and used as inputs to the optimization algorithm which then provides new wheel profiles as the output. To verify the long-term performance of the new profiles, wear and fatigue analysis were performed. Wear simulation using Archard's law generated worn profiles that were used to estimate the wheel's theoretical service life and degradation rate. Fatigue simulation employing the Dang Van fatigue criterion was done to obtain the stress histories for fatigue analysis using a Finite Element Model (FEM) inside ANSYS® software. The study resulted in three optimized profiles. The first was optimal for wear, the second was optimal for fatigue, and the last presented an intermediate performance in terms of wear and fatigue. The findings show that all three optimized wear profiles reduced the wear index by at least 47% when compared to the currently used profile. Furthermore, an improvement of at least 159% was observed in the fatigue index for all new profiles. The service life increased 12% according to the current hollow wheel criterion, 3 mm.