The wheel-rail contact is at the core of all research related to vehicletrackinteraction. This tiny interface governs the dynamic performanceof rail vehicles through the forces it transmits and, like any high stressconcentration zone, it is subjected to serious damage phenomena. Thus,a clear understanding of the rolling contact between wheel and rail is keyto realistic vehicle dynamics simulation and damage analysis. In a multi-body dynamics simulation, the demanding contact problemshould be evaluated at about every millisecond for several wheel-rail pairs.Hence, a rigorous treatment of the contact is highly time-consuming.Simplifying assumptions are therefore made to accelerate the simulationprocess. This gives rise to a trade-o between the accuracy and computationaleciency of the contact model in use. Conventionally, Hertz+FASTSIM is used for calculation of the contactforces thanks to its low computational cost. However, the elliptic patchand pressure distribution obtained by Hertz' theory is often not realisticin wheel-rail contact. Moreover, the use of parabolic traction bound inFASTSIM causes considerable error in the tangential stress estimation.This combination leads to inaccurate damage predictions. Fast non-elliptic contact models are proposed by others to tacklethis issue while avoiding the tedious numerical procedures. The studiesconducted in the present work show that the accuracy of these models iscase-dependent. To improve the accuracy of non-elliptic patch and pressure estimation,a new method is proposed. The method is implemented in an algorithmnamed ANALYN. Comparisons show improvements in patch and, particularly,pressure estimations using ANALYN. In addition, an alternative to the widely-used FASTSIM is developed, named FaStrip. Unlike FASTSIM, it employs an elliptic traction boundand is able to estimate the non-linear characteristic of tangential stressdistribution. Comparisons show more accurate estimation of tangentialstress and slip velocity distribution as well as creep forces with FaStrip. Ultimately, an ecient non-elliptic wheel-rail contact model consistingof ANALYN and FaStrip is proposed. The reasonable computationalcost of the model enables it to be used on-line in dynamics simulationand its accuracy can improve the damage predictions. / <p>QC 20160202</p>

Due to the increase in demand on freight transportation it becomes necessary to avoid delays to ensure that the goods reach its destination on time. The main factors causing disturbances in the traffic on the mainline is the breakdown of vehicles mainly due to damaged wheels. The damaged wheels are identified through the wheel-rail impact force measurements provided by the wheel impact load detectors (WILD). This calls for the optimal schedule of maintenance of wheelsets and wagons in general. During the maintenance, the officials manually check for defective wheels and the exchange of wheelsets is performed based on the type of damage. The classification of wheel damages plays a vital role in providing ease of damage identification and insights to deduce a strategy for wheelset exchange. In this study, an attempt to classify the damaged wheelsets is done by analysing the wheel-rail impact force data from the wayside detectors. The data from the detectors are acquired from PredgeAB, a Luleå based startup pioneering in providing decision support on optimal maintenance schedules and predictive maintenance of rail wheels. Through their detection and prediction solutions it was observed that of all the wheels marked as damaged by the detectors 10% were undamaged. The source of the deviation in the impact force readings could help Predge make better estimations in damage detection and prediction. In this study, the sources contributing to the deviation is studied using multi-body simulations in GENSYS. A new method for modelling wheel damage is developed to overcome the shortcomings of the software. The findings of this study can then be used appropriately to make classifications of wheel damages. / På grund av den ökade efterfrågan på godstransporter är det nödvändigt attundvika förseningar för att säkerställa att varorna når sin destination i tid. Deviktigaste faktorerna som orsakar störningar i trafiken på huvudlinjen är stopp ispår av fordon, främst på grund av skadade hjul. De skadade hjulen identifierasmed hjälp av mätningar av slagkraft mellan hjul och spår som tillhandahållsav hjulbelastningsdetektorer (WILD). Detta kräver ett optimerat schema förunderhåll av hjulsatser och vagnar i allmänhet. Under underhållet kontrollerartjänstemännen manuellt för defekta hjul och utbytet av hjulsatser utförs baserat på typen av skada. Klassificeringen av hjulskador spelar en viktig roll närdet gäller att underlätta identifiering av skador och ge insikt för att bedöma enstrategi för hjulbyte. I den här studien görs ett försök att klassificera de skadade hjulsatserna genom att analysera data från slagkraften mellan spår ochhjul från detektorer. Uppgifterna från detektorerna hämtas från PredgeAB, enLuleå-baserad uppstartsbolag som är pionjärer för att ge beslutsstöd om optimala underhållsscheman och prediktivt underhåll av järnvägshjul. Genom sinadetekterings- och prediktiva lösningar observerades att 10% av alla hjul märktasom skadade av detektorerna var oskadade. Källan till avvikelsen i slagkraftavläsningarna kan hjälpa Predge att göra bättre uppskattningar när det gällerupptäckning och förutsägelse av skador. I den här studien studeras de källorsom bidrar till avvikelsen med simuleringar av flera kroppar i GENSYS. En nymetod för modellering av hjulskador har utvecklats för att övervinna programvarans brister. Resultaten av denna studie skulle kunna användas på lämpligtsätt för att göra klassificeringar av hjulskador.