Effects of modelling simplifications in FEA of railway wheels

Cuperus, Jacobus Louwrens (2017-03)

Thesis (MEng)--Stellenbosch University, 2017.

Thesis

ENGLISH ABSTRACT: Typically, as part of the specifications outlining the requirements of railway wheels, the supplier must demonstrate through finite element analysis (FEA) that the design can withstand certain operational conditions and load cases. The specifications focused on in this study are the Transnet Freight Rail RS/ME/SP/008 specification for the supply of cast trailing stock wheels as well as the RS/ME/SP/021 specification for the supply of wrought wheels for tractive and trailing stock. These specifications contain a section pertaining to the finite element analysis of the wheel with respect to certain operational conditions. These conditions involve, however, complex interactions between loading and material non-linearities, superimposed onto a complex initial stress field. This stress field is laboriously obtained through a heat treatment schedule designed to induce compressive residual hoop stress in the rim of the wheel. These complexities are ostensibly routinely ignored in the application of these (and similar) standards. This study aimed to provide some engineering and scientific reasoning to the simplification of these complex analyses in the future by investigating the effects of various assumptions and simplifications on the FEA. The stress distribution of the normal contact loads are obtained through numerical simulation. The data is also used to derive a simple contact model for contact between two specific wheel and rail profiles. Residual stresses of the new wheel was determined by simulating the heat treatment process. These models were also used to investigate the modelling details required to get accurate results from such a simulation by selectively applying various simplifications to the model. The convective heat transfer to the atmosphere was investigated with computational fluid dynamic simulations and the data fitted to a non-dimensional heat transfer model. This data was then also compared to published heat transfer models of simplified flow scenarios in order to determine if any of these is a viable option to determine heat transfer behaviour in the future. Finally, all of the data gathered with the other analyses were used to determine the effect of various simplifications on the analysis as prescribed in the Transnet standards.

AFRIKAANSE OPSOMMING: As deel van die spesifikasies waarin die vereistes van spoorweg wiele uitgelê word, word die verskaffer gewoonlik ook vereis om deur middel van eindige element analise (EEA) te demonstreer dat die ontwerp sekere operasionele omstandighede en las gevalle kan weerstaan. Die spesifikasies waarop hierdie studie fokus is die Transnet Freight Rail RS/ME/SP/008 spesifikasie vir die verskaffing van gegote wiele vir gesleepte voertuie asook die RS/ME/SP/021 spesifikasie vir die verskaffing van gesmeede wiele vir aangedrewe sowel as gesleepte voertuie. Hierdie spesifikasies bevat 'n seksie wat die eindige element analise van die wiel beskryf met betrekking tot sekere operasionele omstandighede. Hierdie toestande behels komplekse interaksies tussen die belasting en materiaal nie-lineariteit wat inwerk op 'n komplekse aanvanklike spanningsveld. Hierdie spanningsveld is moeisaam verkry deur 'n hitte behandeling skedule, ontwerp om druk hoepel resspanning te skep in die rand van die wiel. Hierdie kompleksiteit word oënskynlik gereeld geïgnoreer in die toepassing van hierdie (en soortgelyke) standaarde. Hierdie studie is daarop gemik om ingenieurs- en wetenskaplike motivering te gee vir die toekomstige vereenvoudiging van hierdie komplekse analises deur die uitwerking van verskillende aannames en vereenvoudigings op die EEA te ondersoek. Die druk verspreiding van die normaal kontak laste is verkry deur numeriese simulasie. Die data word ook gebruik om 'n eenvoudige kontak model af te lei vir kontak tussen twee spesifieke wiel en spoor profiele. Resspanning van die nuwe wiel is bepaal deur die hitte behandeling proses te simuleer. Hierdie modelle is ook gebruik om die modellerings tegnieke wat nodig is om akkurate resultate van so 'n simulasie te kry, te ondersoek deur selektiewe vereenvoudigings op die model toe te pas. Die konvektiewe hitte-oordrag na die atmosfeer is ondersoek met numeriese vloei dinamika simulasies en die data was op 'n nie-dimensionele hitte-oordrag model gepas. Hierdie data is dan ook vergelyk met gepubliseerde hitte-oordrag modelle van eenvoudige vloei gevalle ten einde vas te stel of enige van hierdie 'n werkbare opsie bied om hitte-oordrag in die toekoms te bepaal. Ter afsluiting is al die versamelde data vanaf die ander analises gebruik om die effekte te bepaal van verskillende vereenvoudigings op die analises soos voorgeskryf in die Transnet standaarde.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/101107
This item appears in the following collections: