Long-term wear and rolling contact fatigue behaviour of a conformal wheel profile designed for large radius curves

There are many reasons to optimise the wheel–rail interface through redesign or maintenance. Minimising wear and rolling contact fatigue (RCF) initiation on wheels and/or rails is often at the forefront of such considerations. This paper covers the design of a conformal wheel profile and its long-term wear and RCF performance to optimise the wheel–rail interface and subsequently reduce the occurrence of surface-initiated RCF on South Africa’s iron ore export line. A comparative study is performed using multibody dynamics simulation together with numerical wheel wear and RCF predictions. The advantages of a conformal wheel profile design are illustrated by evaluating the worn shape and resulting contact conditions of the conformal design. The conformal design has a steadier equivalent conicity progression and a smaller conicity range compared with the current wheel profile design over the wheel’s wear life. The combination of a conformal wheel profile design with 2?mm hollow wear and inadequate adherence to grinding tolerances often result in two-point contact, thereby increasing the probability of RCF initiation. The conformal wheel profile design proved to have wear and potential RCF benefits compared with the current wheel profile design. However, implementation of such a conformal wheel profile must be accompanied by improved rail grinding practices to ensure rail profile compliance.     

Modelling of wear and fatigue defect formation in wheel–rail contact

The model for analysing wear and fatigue defect formation is developed based on the approaches of contact and fracture mechanics. The model includes the solution of the contact problem for the wheel and rail to find the shape, size and position of the contact zones and the contact stresses and calculation of the surface wear and the function of damage accumulation in the rail and wheel. The wear rate and the worn-profile evolution of the wheel surface are calculated using both statistic and deterministic approaches to modelling of vehicle dynamics (tribo-dynamic modelling). The influence of the evolution of the wheel–rail profiles due to wear on the damage accumulation process is analysed. It is shown that for some values of the wear rate coefficient, the wear process can prevent the crack initiation under the wheel surface.     

Carbody elastic vibrations of high-speed vehicles caused by bogie hunting instability

In particular locations of the high-speed track, the worn wheel profile matched up with the worn rail profile will lead to an extremely high-conicity wheel–rail contact. Consequently, the bogie hunting instability arises, which further results in the so-called carbody shaking phenomenon. In this paper, the carbody elastic vibrations of a high-speed vehicle in service are firstly introduced. Modal tests are conducted to identity the elastic modes of the carbody. The ride comfort and running safety indices for the tested vehicle are evaluated. The rigid–flexible coupling dynamic model for the high-speed passenger car is then developed by using the FE and MBS coupling approach. The rail profiles in those particular locations are measured and further integrated into the simulation model to reproduce the bogie hunting and carbody elastic vibrations. The effects of wheel and rail wear on the vehicle system response, e.g. wheelset bifurcation graph and carbody vibrations, are studied. Two improvement measures, including the wheel profile modification and rail grinding, are proposed to provide possible solutions. It is found that the wheel–rail contact conicity can be lowered by decreasing wheel flange thickness or grinding rail corner, which is expected to improve the bogie hunting stability under worn rail and worn wheel conditions. The carbody elastic vibrations caused by bogie hunting instability can be further restrained.     

Improvement of vehicle–turnout interaction by optimising the shape of crossing nose

Proper rail geometry in the crossing part is essential for reducing damage on the nose rail. To improve the dynamic behaviour of turnout crossings, a numerical optimisation approach to minimise rolling contact fatigue (RCF) damage and wear in the crossing panel by varying the nose rail shape is presented in the paper. The rail geometry is parameterised by defining several control cross-sections along the crossing. The dynamic vehicle–turnout interaction as a function of crossing geometry is analysed using the VI-Rail package. In formulation of the optimisation problem a combined weighted objective function is used consisting of the normal contact pressure and the energy dissipation along the crossing responsible for RCF and wear, respectively. The multi-objective optimisation problem is solved by adapting the multipoint approximation method and a number of compromised solutions have been found for various sets of weight coefficients. Dynamic behaviour of the crossing has been significantly improved after optimisations. Comparing with the reference design, the heights of the nose rail are notably increased in the beginning of the crossing; the nominal thicknesses of the nose rail are also changed. All the optimum designs work well under different track conditions.     

Sensitivity of the wheel–rail contact interactions and Dang Van Fatigue Index in the rail with respect to irregularities of the track geometry

This paper investigates the effects of the track geometry irregularities on the wheel–rail dynamic interactions and the rail fatigue initiation through the application of the Dang Van criterion, that supposes an elastic shakedown of the structure. The irregularities are modelled, using experimental data, as a stochastic field which is representative of the considered railway network. The tracks thus generated are introduced as the input of a railway dynamics software to characterise the stochastic contact patch and the parameters on which it depends: contact forces and wheelset–rail relative position. A variance-based global sensitivity analysis is performed on quantities of interest representative of the dynamic behaviour of the system, with respect to the stochastic geometry irregularities and for different curve radius classes and operating conditions. The estimation of the internal stresses and the fatigue index being more time-consuming than the dynamical simulations, the sensitivity analysis is performed through a metamodel, whose input parameters are the wheel–rail relative position and velocity. The coefficient of variation of the number of fatigue cycles, when the simulations are performed with random geometry irregularities, varies between 0.13 and 0.28. In a large radius curve, the most influent irregularity is the horizontal curvature, while, in a tight curve, the gauge becomes more important.     

Parametric study of wheel transitions at railway crossings

Vehicle–track interaction at railway crossings is complex due to the discontinuity of the crossings. In this study, the effect of the local crossing geometry, the track alignment, and the wheel profiles on the wheel transition behaviour is investigated using the multi-body system software package VI-Rail. The transition behaviour is evaluated based on the location of the transition point along the crossing (and the location of impact), the contact pressure and the energy dissipation during the wheel–rail contact. A detailed parametric study of the crossing geometry has been performed, through which the most effective parameters for defining the crossing geometry are identified. These parameters are the cross-sectional shape of the nose rail, which can be tuned by one variable, and the vertical distance between the top of the wing rail and the nose rail. Additionally, a parametric study on the interaction influence of the crossing geometry, the track alignment and the wheel profile is performed using the design of experiments method with a two-level full factorial design. The longitudinal height profile of the crossing and the wheel profile are the most significant factors.     

Stochastic analysis of dynamic interaction between train and railway turnout

The performance of a railway turnout (switch and crossing) is influenced by a large number of input parameters of the complex train–turnout system. To reach a robust design that performs well for different traffic situations, random distributions (scatter) of these inputs need to be accounted for in the design process. Stochastic analysis methods are integrated with a simulation model of the dynamic interaction between train and turnout. For a given nominal layout of the turnout, using design of experiments methodology and a two-level fractional factorial screening design, four parameters (axle load, wheel–rail friction coefficient, and wheel and rail profiles) are identified to be the most significant. These parameters are further investigated using a three-level full factorial design and stochastic analysis. The random distributions of transverse wheel profile and set of transverse rail profiles along the switch panel are accounted for by the Karhunen–Loève expansion technique. The influence of the random distributions of the input parameters on the statistical outputs of wheel–rail contact forces, wear and rolling contact fatigue is assessed using Latin hypercube sampling to generate a number of stochastic load realizations.     

Design optimisation of switch rails in railway turnouts

Inspired by a manufacturing process of switch rails for railway turnouts, a method for the optimisation of switch rail profile geometry is presented. The switch rail profile geometry is parameterised with four design variables to define a B-spline curve for the milling tool profile, and two design variables to prescribe the deviation from the nominal vertical path of the milling tool. The optimisation problem is formulated as a multi-objective minimisation problem with objective functions based on the contact pressure and the energy dissipation in the wheel–rail contact. The front of Pareto optimal solutions is determined by applying a genetic type optimisation algorithm. The switch rail profile designs are evaluated by simulations of dynamic train–turnout interaction. It is concluded that the obtained set of Pareto optimal solutions corresponds to a rather small variation in design variables where increased profile height and increased profile shoulder protuberance are preferred for both objectives. The improvement in the objectives comes at the cost of an earlier wheel transition to the switch rail and thus increased vertical loading at a thinner rail cross-section. The performance of the optimised geometry is evaluated using a set of 120 measured wheel profiles, and it is shown that the optimised geometry reduces damage also for this large load collective. It is concluded that accurate limits on switch rail loading need to be established to determine the feasible design space for switch rail geometry optimisation.     

A non-Hertzian method for solving wheel–rail normal contact problem taking into account the effect of yaw

A novel approach is proposed in this paper to deal with non-Hertzian normal contact in wheel–rail interface, extending the widely used Kik–Piotrowski method. The new approach is able to consider the effect of the yaw angle of the wheelset against the rail on the shape of the contact patch and on pressure distribution. Furthermore, the method considers the variation of profile curvature across the contact patch, enhancing the correspondence to CONTACT for highly non-Hertzian contact conditions. The simulation results show that the proposed method can provide more accurate estimation than the original algorithm compared to Kalker’s CONTACT, and that the influence of yaw on the contact results is significant under certain circumstances.     

Dynamic model for the wheel–rail contact friction

Accurately estimating the coefficient of friction (CoF) is essential in modelling railroad dynamics, reducing maintenance costs, and increasing safety in rail operations. The typical assumption of a constant CoF is widely used in theoretical studies; however, it has been noticed that the CoF is not constant, but rather depends on various dynamic parameters and instantaneous conditions. In this paper, we present a newly developed three-dimensional nonlinear CoF model for the dry rail condition and test the CoF variation using this model with estimated dynamic parameters. The wheel–rail is modelled as a mass–spring–damper system to simulate the basic wheel–rail dynamics. Although relatively simple, this model is considered sufficient for the purpose of this study. Simulations are performed at a train speed of 20 m/s using rail roughness as an excitation source. The model captures the CoF extremes and illustrates its nonlinear behaviour and instantaneous dependence on several structural and dynamic parameters.     

Robust optimisation of railway crossing geometry

This paper presents a methodology for improving the crossing (frog) geometry through the robust optimisation approach, wherein the variability of the design parameters within a prescribed tolerance is included in the optimisation problem. Here, the crossing geometry is defined by parameterising the B-spline represented cross-sectional shape and the longitudinal height profile of the nose rail. The dynamic performance of the crossing is evaluated considering the variation of wheel profiles and track alignment. A multipoint approximation method (MAM) is applied in solving the optimisation problem of minimising the contact pressure during the wheel–rail contact and constraining the location of wheel transition at the crossing. To clarify the difference between the robust optimisation and the normal deterministic optimisation approaches, the optimisation problems are solved in both approaches. The results show that the deterministic optimum fails under slight change of the design variables; the robust optimum, however, has improved and robust performance.     

An efficient approach to the analysis of rail surface irregularities accounting for dynamic train–track interaction and inelastic deformations

A two-dimensional computational model for assessment of rolling contact fatigue induced by discrete rail surface irregularities, especially in the context of so-called squats, is presented. Dynamic excitation in a wide frequency range is considered in computationally efficient time-domain simulations of high-frequency dynamic vehicle–track interaction accounting for transient non-Hertzian wheel–rail contact. Results from dynamic simulations are mapped onto a finite element model to resolve the cyclic, elastoplastic stress response in the rail. Ratcheting under multiple wheel passages is quantified. In addition, low cycle fatigue impact is quantified using the Jiang–Sehitoglu fatigue parameter. The functionality of the model is demonstrated by numerical examples.     

Deterioration mechanisms in the wheel–rail interface with focus on wear prediction: a literature review

Wheel–rail interface management is imperative to railway operation and its maintenance represents a major share of the total maintenance cost. In general, the course of events usually called wear is a complicated process involving several modes of material deterioration and contact surface alteration. Thus material removal or relocation, plastic flow and phase transformation may take place at, just below, or in-between the contacting surfaces. A higher degree of predictability of deterioration mechanisms and a firm basis for optimisation of the wheel–rail system are anticipated to reveal a great potential for cost savings. Wear in the sense of material loss and related wheel–rail profile evolution represents one of several modes of damage. The purpose of this survey is to explore research on wear simulation, to some degree extended to neighbouring disciplines. It is believed that a cross-disciplinary approach involving, for instance, adhesive and abrasive wear, surface plasticity, and rolling contact fatigue opens new perspectives to improved damage prediction procedures.     

Dynamic performance of high-speed train passing windbreak breach under unsteady crosswind

Field test and computational fluid dynamics (CFD) method are conducted to investigate the safety of high-speed train under unsteady crosswind. Wheel–rail forces of high-speed train passing a breach between two windbreaks under strong crosswind are measured in a field test. The derailment coefficient of first wheelset of front car at the windward side reaches the allowable value. Meanwhile, the left and right of lateral wheel–rail force are in the opposite direction. This kind of phenomenon has not been tested before. Therefore, CFD and multi-body simulations are performed in order to study the phenomena. Good agreement is obtained between the simulation results and the experimental data. It is concluded that the sudden increase of transient aerodynamic loads, when the train passing the breach, is the root of this phenomenon; after running along the same direction as carbody and bogie run along the opposite direction during the high-speed train passing the windbreak breach; larger opposite longitudinal creeping forces of first wheelset compel the first wheelset to yaw toward the windward side; meanwhile, larger lateral wheel–rail forces compel the first wheelset to run toward the windward side rail; the left and right lateral wheel–rail forces become opposite because the right wheel impacts the windward side rail.     

Indirect method for wheel–rail force measurement and derailment evaluation

Wheel set flange derailment criteria for railway vehicles are derived and the influence of wheel–rail contact parameters is studied. An indirect method for wheel–rail force measurement based on these derailment evaluation criteria is proposed. Laboratory tests for the calibration of strain–force devices on the bearing box are carried out to determine the relationship between the applied force and the measured strain. The simulation package, SIMPACK, is used to develop a passenger car model to generate wheel–rail forces and vibration signals. Different cases are considered in this model to provide an accurate validation of the identified wheel–rail forces. A feasibility test is conducted in the Beijing Loop test line using a passenger car equipped with a set of strain gauges on the wheel set. The comparison of the force time history applied to the instrumented wheel set and that obtained using the indirect method is presented.     

Mathematical modelling of train–turnout interaction

The paper proposes a mathematical model of train–turnout interaction in the mid-frequency range (0–500 Hz). The model accounts for the effects of rail profile variation along the track and of local variation of track flexibility. The proposed approach is able to represent the condition of one wheel being simultaneously in contact with more than one rail, allowing the accurate prediction of the effect of wheels being transferred from one rail to another when passing over the switch toe and the crossing nose. Comprehensive results of train–turnout interaction during the negotiation of the main and the branch lines are presented, including the effect of wear of wheel/rail profiles and presence of track misalignment. In the final part of the paper, comparisons are performed between the results of numerical simulations and line measurements performed on two different turnouts for urban railway lines, showing a good agreement between experimental and numerical results.     

Development and validation of a wear model for the analysis of the wheel profile evolution in railway vehicles

The numerical wheel wear prediction in railway applications is of great importance for different aspects, such as the safety against vehicle instability and derailment, the planning of wheelset maintenance interventions and the design of an optimal wheel profile from the wear point of view. For these reasons, this paper presents a complete model aimed at the evaluation of the wheel wear and the wheel profile evolution by means of dynamic simulations, organised in two parts which interact with each other mutually: a vehicle's dynamic model and a model for the wear estimation. The first is a 3D multibody model of a railway vehicle implemented in SIMPACK?, a commercial software for the analysis of mechanical systems, where the wheel–rail interaction is entrusted to a C/C++user routine external to SIMPACK, in which the global contact model is implemented. In this regard, the research on the contact points between the wheel and the rail is based on an innovative algorithm developed by the authors in previous works, while normal and tangential forces in the contact patches are calculated according to Hertz's theory and Kalker's global theory, respectively. Due to the numerical efficiency of the global contact model, the multibody vehicle and the contact model interact directly online during the dynamic simulations.

The second is the wear model, written in the MATLAB® environment, mainly based on an experimental relationship between the frictional power developed at the wheel–rail interface and the amount of material removed by wear. Starting from a few outputs of the multibody simulations (position of contact points, contact forces and rigid creepages), it evaluates the local variables, such as the contact pressures and local creepages, using a local contact model (Kalker's FASTSIM algorithm). These data are then passed to another subsystem which evaluates, by means of the considered experimental relationship, both the material to be removed and its distribution along the wheel profile, obtaining the correspondent worn wheel geometry.

The wheel wear evolution is reproduced by dividing the overall chosen mileage to be simulated in discrete spatial steps: at each step, the dynamic simulations are performed by means of the 3D multibody model keeping the wheel profile constant, while the wheel geometry is updated through the wear model only at the end of the discrete step. Thus, the two parts of the whole model work alternately until the completion of the whole established mileage. Clearly, the choice of an appropriate step length is one of the most important aspects of the procedure and it directly affects the result accuracy and the required computational time to complete the analysis.

The whole model has been validated using experimental data relative to tests performed with the ALn 501 ‘Minuetto’ vehicle in service on the Aosta–Pre Saint Didier track; this work has been carried out thanks to a collaboration with Trenitalia S.p.A and Rete Ferroviaria Italiana, which have provided the necessary technical data and experimental results.     

An inverse design method for rail grinding profiles

Inspired by the optimisation design method for restoration of worn wheel profiles, an inverse design method based on optimal rail grinding profiles is presented in this paper. To improve grinding quality, vehicle dynamic performance is chosen as the main criteria and rolling radii difference function is selected as the key factor (also main target function) determining dynamic performance. Grinding material to be removed is chosen as the auxiliary target aimed at extending rail service life. Besides that, wheel–rail contact distribution is also taken into consideration as an auxiliary target preventing stress concentration and fatigue growth. By introducing certain presuppositions, all the design targets will form an inverse design problem. This problem can be solved using hybrid discrete numerical methods. Considering different grinding requirements, two examples of grinding profile design for straight and curved track will be discussed. Results show that the presented method is efficient and effective. Practical implementation has been carried out at several grinding sites in China.     

Wheel life prediction model – an alternative to the FASTSIM algorithm for RCF

In this article, a wheel life prediction model considering wear and rolling contact fatigue (RCF) is developed and applied to a heavy-haul locomotive. For wear calculations, a methodology based on Archard's wear calculation theory is used. The simulated wear depth is compared with profile measurements within 100,000?km. For RCF, a shakedown-based theory is applied locally, using the FaStrip algorithm to estimate the tangential stresses instead of FASTSIM. The differences between the two algorithms on damage prediction models are studied. The running distance between the two reprofiling due to RCF is estimated based on a Wöhler-like relationship developed from laboratory test results from the literature and the Palmgren-Miner rule. The simulated crack locations and their angles are compared with a five-year field study. Calculations to study the effects of electro-dynamic braking, track gauge, harder wheel material and the increase of axle load on the wheel life are also carried out.