Rail Wear Simulation and Validation

Transportation Technology Center, Inc. (TTCI), a wholly owned subsidiary of the Association of American Railroads (AAR), has developed an iterative rail wear prediction model in the NUCARS^® vehicle/track interaction multibody simulation program through internal research and development efforts and with funding from Network Rail (NR) in the UK. The rail wear model was built upon the NUCARS^® penetration model¹ to take advantage of the wheel/rail (W/R) contact calculation methodology for conformal W/R profiles. In addition to the advantages of NUCARS vehicle and track modeling capabilities, it modifies the rail profile online based on the Wear Indices (Tγ) and penetrated W/R profile shapes in the multipoint contact patches, and automatically updates the rail profile for the next run. The penetrated wheel profile segments or “wheel footprints” are blended into the modified rail profile. The worn rail shape eventually resembles the wheel shapes in the wheel database, and the wear process results in conformal W/R profile shapes. Rail wear prediction was validated using rail wear test results based on 515 million gross tons (MGT) of heavy axle load (39-ton (35-tonne) axle loads) freight traffic accumulated from 2003 to 2007 at the Facility for Accelerated Service Testing (FAST) on the nonground test zone (Section 25, 6-degree (291-meter (m)) curve with 5 inches (127 millimeter (mm)) superelevation). A wheel database, consisting of 50 measured new, mildly worn and heavily worn FAST train wheel profiles, was used to reflect the wheel shape effects during the wear process. This model has been used to predict rail relative wear trends of ground rail profiles for NR.^2,3,4 A quantified W/R gap loss function has been implemented in NR’s Track-Ex© program⁵ for prioritizing rail grinding.