Thermomechanical and Vibrational Fatigue Life Assessment of Railway Wheels

Rail wheel failures can lead to derailment of trains, cost lives and billions of dollars to the railway industry. Therefore, it is important to understand the failure behavior of critical components of the train, especially the ones which are exposed to extreme weather conditions. This understanding is much needed to develop effective risk management procedures. Most research studies in wheel fatigue life estimation did not account for structural, thermal and prestressed random vibrational loading as a coupled field. The present study develops an advanced numerical analysis methodology that incorporates real service conditions to calculate fatigue life. Firstly, a validation study was carried out before the fatigue life estimation. The prime objective of this validation study is to verify that the contact pressure observed in the Ansys FEA analysis is consistent with the Hertzian contact theory. The contact pressure estimated by Ansys simulation is 656 MPa and the mathematical calculation is 680 MPa showing therefore a variation of only 1.25%. Having verified the FEA model with a theoretical calculation, a full three-dimensional finite element analysis is developed to estimate fatigue life considering static, thermal and vibrational loadings. Mesh sensitivity study, selection of the appropriate contact formulation and detection method aided further to increase the accuracy of the FEA results. Secondly, having achieved a good validation between FEA and Hertzian theory, the major findings of this preliminary study such as second-order hexahedron with 10 mm element size and contact settings of normal Lagrange formulation and nodal normal from contact detection method to be adopted for the thermo-mechanical and pre-stressed vibration analyses. The fatigue life of the wheel was estimated using the stress life approach for thermo-mechanical loading and the Steinberg approach was utilized for the dynamic loading cases. This robust methodology was finalized using a typical rail-track subassembly comprised of a wheel, track, axle, ballast and sleepers. Finally, the developed methodology was used to estimate the fatigue life of the train wheel which failed and caused a derailment in Manitoba (Canada). This case study was performed to verify the proposed FEA methodology and to understand the level of applicability and accuracy in terms of failure location and failure mode prediction. The contact pressure estimated by Ansys simulation was 1028 MPa and from theory was 1188 MPa. This shows that a variation of 15% was observed. The simulation shows a good validation with the real case as the location of the major failure was accurately predicted by the thermomechanical FEA analysis. Subsurface failures that occurred in the circumference of the wheel are predicted by the vibrational FEA fatigue analysis. Hence, this FEA methodology can be employed for any wheel to estimate the fatigue life in the design phase in a better way. Estimating the fatigue life during the design phase has many advantages and having a robust methodology to aid this process is very important. This research aimed at developing an FEA methodology that helps fatigue life estimation process in the design phase. With this research approach, a robust design can be built with minimal cost. Not only the approach has the advantage of minimal product development cost but it also aims at reducing many life-threatening disasters due to derailments caused by wheel failures.

Name thesis:degree_name

Master of Applied Science (MASc)

Level thesis:degree_level

Master's

Discipline thesis:degree_discipline

Engineering - Industrial Systems

Grantor dc:publisher

Faculty of Graduate Studies and Research, University of Regina

Year dc:date.issued

2021

Author dc:creator

• Sekar, Saravanakumar

Advisors dc:contributor.advisor

• Henni, Amr
• Muthu, Jacob

Committee members dc:contributor.committeemember

• Ibrahim, Hussameldin
• El-Darieby, Mohamed

Language dc:language.iso

en

DOI

https://doi.org/10.82465/4579

OAI identifier oai:identifier

oai:uregina.scholaris.ca:10294/14983