Determination method of wheel-rail rolling contact fatigue damage function parameters

Abstract

The invention discloses a determination method for wheel-rail rolling contact fatigue damage function parameters, which considers an intra-spot mesoscopic contact solution, can be suitable for different railway systems and has a good prediction effect. The determination method comprises the following steps: solving transient wheel-rail contact state parameters; calculating a macroscopic rolling contact solution and an intra-spot mesoscopic contact solution; solving a transient abrasion number T gamma, a total transient damage value Dt in a contact spot and a transient damage peak value hm at any moment; solving a local abrasion number tau [gamma] L in the contact spot; hm is used as dL of tau gamma L, and the corresponding relation between tau gamma L and hm is obtained; obtaining a relation between T gamma and tau gamma L covering the whole domain of the macroscopic damage function model under different traction coefficients through working condition changes; the tau gamma L at the critical point is reduced according to the abrasion reduction amount until the local damage total value dt in the corresponding contact spot is equal to the transient damage total value Dt in the contact spot, and the relation between the reduced tau gamma L and dL is obtained; and fitting to obtain local damage function model parameters and / or reversely deducing macroscopic damage function model parameters of another railway system.

Description

Technical Field

[0001] The present invention relates to the technical field of wheel-rail rolling contact fatigue damage, and more particularly, to a method for determining parameters of a wheel-rail rolling contact fatigue damage function. Background Art

[0002] Rolling contact fatigue is a common form of wheel-rail damage. In modern railways, it mostly results from excessive creep rate / force between the wheel and the rail when passing through curves. It is a direct result of the accumulation of plastic strain in the surface material under the action of rolling contact load cycles until ductility is exhausted. In the early stage, it appears as surface cracks, and in the later stage, it mostly develops into spalling and chipping. In rare cases, it can lead to the fracture of the wheel rim and the rail, threatening the safety of train operation.

[0003] Heavy-haul railways are characterized by large axle loads and high traffic volumes. In small-radius curve sections with intense wheel-rail interaction, rolling contact fatigue of the rail often occurs, which is one of the main reasons for rail replacement. Rolling contact fatigue of the rail in curve sections is often referred to as inclined cracks or fish-scale patterns. Generally, they are continuously and approximately stably distributed longitudinally, the crack interval can be as low as 1 mm, and the crack direction forms a certain angle with the longitudinal direction, belonging to typical continuous rolling contact fatigue.

[0004] The stage from the initial crack initiation to the visible crack size of 2 mm is defined as the crack initiation stage. The research on the mechanism of wheel-rail rolling contact fatigue initiation and development and treatment measures has always been one of the research hotspots in the industry. For continuous rolling contact fatigue, the academic community has established some widely accepted wheel-rail rolling contact fatigue initiation prediction models, such as the BS11 rail damage function model proposed by the Railway Safety and Standards Board (RSSB) in the UK and the U75V quenched rail damage function model obtained according to the hardness calculation recommendations. The damage function model is a piecewise linear relationship curve between the wear number and the damage value.

[0005] However, due to the huge differences in railway systems around the world, the applicability of the damage function model in different line systems is not good. The main reasons for the poor usability of the damage function are as follows: First, the proposed process of the damage function model is relatively single, and the number and types of selected stations are limited, unable to take into account the characteristics of a wider range of railway systems; Second, for the crack initiation size, the calculation of wear and energy in the damage function model is based on the macroscopic rolling contact solution (referred to as the macroscopic damage function model in this patent), without considering the mesoscopic contact solution within the contact patch, and unable to accurately reflect the local load conditions of the rail.

[0006] Despite the above application problems, the relationship between fatigue and wear supported by the damage function model has been widely proven to be correct. Therefore, when many scholars conduct predictive analysis within the target line system using the damage function model, there are generally two improvement strategies: one is to adjust the parameters of the damage function model according to the hardness and shear strength of the wheel-rail material, including adjusting the wear values at key points and the extreme values of the damage function; the other is to carry out on-site measurements and multi-body dynamics simulation calibration work to obtain the damage function parameters applicable to the target line system. The former is relatively simple, but the prediction effect of the new damage function obtained is average; the latter has a better prediction effect for the new damage function obtained, but requires a large amount of time and economic cost.

[0007] Generally speaking, the widespread application of the damage function model still lacks a general method for quickly determining the damage function model parameters applicable to different railway systems. This method needs to improve the disadvantage of the inaccurate original damage function model and also overcome the deficiencies of the above two damage function parameter adjustment strategies. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a method for determining the parameters of the wheel-rail rolling contact fatigue damage function that takes into account the mesoscopic contact solution within the contact patch, can be applicable to different railway systems, and has a good prediction effect.

[0009] To achieve the above object, the present invention provides a method for determining the parameters of the wheel-rail rolling contact fatigue damage function, and the technical solution is as follows:

[0010] The method for determining the parameters of the wheel-rail rolling contact fatigue damage function includes the following steps:

[0011] Step 100: Establish a vehicle system dynamics model of a certain railway system and solve the transient wheel-rail contact state parameters;

[0012] Step 200: According to the transient wheel-rail contact state parameters, establish a three-dimensional wheel-rail transient rolling contact finite element model, and calculate the macroscopic rolling contact solution and the mesoscopic contact solution within the contact patch; the macroscopic rolling contact solution includes the rigid creepage rate and the rigid creep force at the wheel-rail contact interface; the mesoscopic contact solution within the contact patch includes the local creepage rate and the local tangential contact stress within the wheel-rail contact patch.

[0013] Step 300: Adopt a macroscopic damage function model based on the macroscopic rolling contact solution, and solve the transient wear number Tγ, the total transient damage value D within the contact patch, t and the peak value h of the transient damage value m at any moment according to the rigid creepage rate;

[0014] Step 400: Solve the local wear number τγ within the contact patch according to the local creepage rate and the tangential contact stress; L ;

[0015] Step 500: Using the peak value h of the transient damage value m as the maximum local wear number τγ within the contact patch L to obtain the local damage value d L , and obtain the corresponding relationship between the local wear number τγ L and the peak value h of the transient damage value m ;

[0016] Step 600: By changing the working conditions, obtain the relationship between the transient wear number Tγ and the local wear number τγ that cover the entire domain of the macroscopic damage function model under different traction coefficients L ;

[0017] Step 700: For the local wear number τγ at the critical point Lc L , reduce it according to the wear reduction amount until the total local damage value d t within the corresponding contact patch is equal to the total transient damage value D t within the contact patch, and obtain the relationship between the reduced local wear number τγ L and the local damage value d L ;

[0018] Step 800: Fit to obtain the parameters of the local damage function model based on the mesoscopic contact solution within the patch for a certain railway system and / or reverse-infer the parameters of the macroscopic damage function model for another railway system.

[0019] It can be seen that the method for determining the parameters of the wheel-rail rolling contact fatigue damage function of the present invention has the following advantages:

[0020] First of all, for rolling contact fatigue, the present invention adopts the mesoscopic contact solution within the patch, uses the macroscopic damage function model based on the macroscopic rolling contact solution, local wear number reduction, and working condition changes to fit the parameters of the local damage function model based on the mesoscopic contact solution within the patch. Furthermore, a rail rolling contact fatigue initiation model based on the mesoscopic rolling contact solution within the contact patch can be established, which can quickly and simply predict the fatigue initiation situation at local positions at a scale closer to the essence of fatigue initiation, with better and more comprehensive prediction effects, and has important guiding significance for preventing and controlling this type of fatigue failure disease, especially suitable for predicting the initiation of heavy-haul rail rolling contact fatigue.

[0021] Secondly, when there are no significant changes in the material properties (wear resistance and fatigue resistance) of the wheel-rail, the relationship between the number of local wear in the wheel-rail contact patch and the initiation life of rolling contact fatigue is the same for different railway systems. However, the parameters of the macroscopic damage function model corresponding to different railway systems are different, which is also the main reason for the poor applicability of the existing damage function models in different railway line systems. The present invention can be used as a general method for quickly determining the parameters of the damage function model applicable to different railway systems, that is, through the method for determining the parameters of the damage function model of the present invention, using the parameters of the local damage function model of a certain railway system as the calibration basis to quickly deduce and determine the parameters of the macroscopic damage function model of another railway system with no significant changes in the wheel-rail material properties.

[0022] The following further describes the present invention in conjunction with the accompanying drawings and specific embodiments. The additional aspects and advantages of the present invention will be partially given in the following description, partially become apparent from the following description, or be understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The drawings constituting a part of the present invention are used to assist in understanding the present invention. The content provided in the drawings and the related descriptions in the present invention can be used to explain the present invention, but do not constitute an improper limitation to the present invention. In the drawings:

[0024] Figure 1 It is a photo of the rolling contact fatigue crack on the rail surface of a 800m radius circular curve section of a heavy-haul railway.

[0025] Figure 2 It is a schematic diagram of a three-dimensional wheel-rail transient rolling contact finite element model.

[0026] Figure 3 It is a curve showing the variation of the creep forces and their angles of the inner rail and outer rail in the vehicle system dynamics model and the three-dimensional wheel-rail transient rolling contact finite element model with the rolling distance.

[0027] Figure 4 It is a projection diagram of the local tangential contact stress of the outer rail in the three-dimensional wheel-rail transient rolling contact finite element model on the xOy plane with respect to the rolling distance and the lateral position.

[0028] Figure 5 It is a projection diagram of the local tangential contact stress of the inner rail in the three-dimensional wheel-rail transient rolling contact finite element model on the xOy plane with respect to the rolling distance and the lateral position.

[0029] Figure 6 It is a schematic diagram of the damage function model of U75V quenched rail.

[0030] Figure 7Curves showing the variation of the transient wear number Tγ with the rolling distance in the vehicle system dynamics model and the three-dimensional wheel-rail transient rolling contact finite element model.

[0031] Figure 8 For the local wear numbers τγ of the inner and outer rails in the contact patch of the three-dimensional wheel-rail transient rolling contact finite element model L Three-dimensional distribution diagram with respect to the lateral position and the rolling distance.

[0032] Figure 9 For the local wear number τγ of the inner rail in the contact patch of the three-dimensional wheel-rail transient rolling contact finite element model L Three-dimensional distribution diagram with respect to the lateral position and the rolling distance.

[0033] Figure 10 For the local wear number τγ in the three-dimensional wheel-rail transient rolling contact finite element model L And the peak value h of the transient damage value m Corresponding relationship diagram.

[0034] Figure 11 For the transient wear number Tγ and the local wear number τγ on the outer rail side in the wheel-rail transient rolling contact finite element model under typical working conditions L Curves showing the variation of the peak values with the rolling distance.

[0035] Figure 12 For the transient wear number Tγ and the local wear number τγ on the inner rail side in the wheel-rail transient rolling contact finite element model under typical working conditions L Curves showing the variation of the peak values with the rolling distance.

[0036] Figure 13 For the transient wear number Tγ and the local wear number τγ on the outer rail side in the wheel-rail transient rolling contact finite element model under different traction coefficients μ L Curves showing the variation of the peak values with the rolling distance.

[0037] Figure 14 For the transient wear number Tγ and the local wear number τγ on the inner and outer rail sides under different traction coefficients μ L Correlation diagram between the peak values.

[0038] Figure 15 For the transient wear number Tγ and the local wear number τγ at all times under different traction conditions L Relationship diagram of the peak values.

[0039] Figure 16 For the local wear number τγ before and after reduction L And the local damage value d L Variation curve.

[0040] Figure 17Variation diagram of the local damage value within the contact patch on the outer rail side predicted by the local damage function model with respect to the transient wear number Tγ.

[0041] Figure 18 Prediction results of the rolling contact fatigue damage values on the inner and outer rail surfaces under typical working conditions with respect to the rolling distance (longitudinal) and lateral position.

[0042] Figure 19 Distribution diagram of the rolling contact fatigue damage values of the inner and outer rails predicted by the damage function model and the local damage function model within a certain cross-section under typical working conditions for U75V quenched rails. Detailed implementation manner

[0043] The present invention will be clearly and completely described below with reference to the accompanying drawings. Those of ordinary skill in the art will be able to implement the present invention based on these descriptions. Before describing the present invention with reference to the accompanying drawings, it should be particularly noted that:

[0044] The technical solutions and technical features provided in each part including the following description of the present invention can be combined with each other without conflict.

[0045] In addition, the embodiments of the present invention involved in the following description are usually only a part of the embodiments of the present invention, rather than all the embodiments. Therefore, all other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts should fall within the protection scope of the present invention.

[0046] Regarding the terms and units in the present invention. The terms "comprising", "having" and any variations thereof in the description, claims and relevant parts of the present invention are intended to cover non-exclusive inclusion.

[0047] The detailed implementation manner of the method for determining the parameters of the wheel-rail rolling contact fatigue damage function of the present invention includes the following steps:

[0048] Step 100, establish a vehicle system dynamics model of a certain railway system, and solve the transient wheel-rail contact state parameters; preferably, establish the vehicle system dynamics model in SIMPACK.

[0049] Step 200, according to the transient wheel-rail contact state parameters, establish a three-dimensional transient wheel-rail rolling contact finite element model, and calculate the macroscopic rolling contact solution and the mesoscopic contact solution within the contact patch; the macroscopic rolling contact solution includes the rigid creep rate and the rigid creep force at the wheel-rail contact interface; the mesoscopic contact solution within the contact patch includes the local creep rate and the local tangential contact stress within the wheel-rail contact patch; preferably, establish the three-dimensional transient wheel-rail rolling contact finite element model in ANSYS / Ls-dyna.

[0050] Step 300: Using a macroscopic damage function model based on the macroscopic rolling contact solution, solve for the transient wear number \(T_{\gamma}\), the total transient damage value \(D\) within the contact patch, and the peak value \(h\) of the transient damage value at any given time according to the rigid creep rate. t And the peak value \(h\) of the transient damage value m ;

[0051] Step 400: Solve for the local wear number \(\tau_{\gamma}\) within the contact patch according to the local creep rate and the tangential contact stress. L ;

[0052] Step 500: Take the peak value \(h\) of the transient damage value m as the local damage value \(d\) of the maximum local wear number \(\tau_{\gamma}\) within the contact patch L , and obtain the corresponding relationship between the local wear number \(\tau_{\gamma}\) L and the peak value \(h\) of the transient damage value L ; m ;

[0053] Step 600: Through changes in the operating conditions, obtain the relationship between the transient wear number \(T_{\gamma}\) and the local wear number \(\tau_{\gamma}\) that covers the entire domain of the macroscopic damage function model under different traction coefficients. L ;

[0054] Step 700: Reduce the local wear number \(\tau_{\gamma}\) at the critical point according to the wear reduction amount until the corresponding total local damage value \(d\) within the contact patch L is equal to the total transient damage value \(D\) within the contact patch t , and obtain the relationship between the reduced local wear number \(\tau_{\gamma}\) t and the local damage value \(d\) L ; L ;

[0055] Step 800: Fit to obtain the parameters of the local damage function model based on the mesoscopic contact solution within the patch for a certain railway system and / or reverse-infer the parameters of the macroscopic damage function model for another railway system.

[0056] In the above steps, the calculation expressions for the rigid creep rate, local creep rate, transient wear number \(T_{\gamma}\), and peak value \(h\) of the transient damage value are as follows: m ;

[0057]

[0058]

[0059]

[0060]

[0061] \(T_{\gamma}=T\) x \(\gamma\) x +T y \(\gamma\)y (Formula 5)

[0062]

[0063] τγ I = τ x γ Lx + τ y γ Ly (Formula 7)

[0064] Wherein, γ x is the longitudinal rigid creepage rate; γ y is the lateral rigid creepage rate; γ Lx is the longitudinal local creepage rate; γ Ly is the lateral local creepage rate; ω is the wheel rotational speed; r is the actual rolling circle radius of the selected tread output point set; v 0 is the longitudinal speed of the wheel axis; v y is the lateral translation speed of the wheel relative to the track; Δv x is the longitudinal relative sliding speed between the wheel and the rail at any point within the contact patch; Δv y is the lateral relative sliding speed between the wheel and the rail at any point within the contact patch; T x is the longitudinal rigid creepage force; T y is the lateral rigid creepage force; b is the lateral semi-axis length of the contact patch, π is the pi; τ x is the longitudinal component of the local tangential contact stress; τ y is the lateral component of the local tangential contact stress.

[0065] The beneficial effects of the present invention are illustrated below through specific application examples.

[0066] In a domestic 25t axle load heavy-haul coal transportation railway, serious rolling contact fatigue occurred on the circular curve section with a radius of less than 1000m. Figure 1 Figure 52 is a photo of the rail surface fatigue crack of the circular curve section with a radius of 800m of a certain heavy-haul railway. As Figure 1 shown, continuous inclined cracks exist at both the outer (higher) rail gauge corner and the inner (lower) rail top surface. There are also dispersed small pieces of spalling on the inner rail; the fatigue area at the outer rail gauge corner is about 11.50mm wide, about 18.50mm from the center to the rail top center line, the crack spacing is about 2mm, and the included angle with the driving direction is about -50°; the fatigue area of the inner rail is approximately at the center of the rail top (i.e., the rail top approximately coincides with the center line of the fatigue area), about 28.50mm wide, the crack spacing is about 1.5mm, and the included angle with the driving direction is about 75°. Combining with relevant literature, it can be determined that this belongs to typical surface-initiated rolling contact fatigue, dominated by the tangential contact load between the wheel and the rail, and a damage function model can be used for prediction.

[0067] Step 100

[0068] For Figure 1 the working conditions of the 800 m radius circular curve section in the middle, a vehicle system dynamics model considering a single C80 freight car was established in SIMPACK, and its key parameters are shown in Table 1. The vehicle system dynamics model adopted new wheel-rail profiles, namely LM and CN60N. The car body, side frame, wheelset, etc. were simplified into rigid bodies with six degrees of freedom, and the total degrees of freedom of the model were 68. Suspension elements were all simulated by springs and damping force elements. The primary (axle box) suspension connected the wheelset and the side frame of the bogie, and the secondary suspension (bearing saddle and bolster) connected the side frame and the car body. The track sub-model included straight, transition curve and circular curve sections, with lengths of 150 m, 380 m and 300 m respectively, and the total length was 830 m. The wheel-rail normal and tangential contacts were calculated by the widely accepted Hertz contact spring and FASTSIM respectively, and the friction between them was characterized by the Coulomb friction model, and the friction coefficient f was taken as the typical value of 0.5 under dry conditions. Further considering the actual operation, it was assumed that the vehicle passed through the curve at a constant speed, and the calculation time step was taken as 0.02 s. The rigid rail hypothesis was adopted, and factors such as gauge, superelevation, rail profile, and rail bottom slope were considered, but the sub-rail structure was ignored.

[0069] Table 1

[0070]

[0071] Thus, when the vehicle passed through the circular curve section at a constant speed of 80 km / h in a steady state, the transient wheel-rail contact state parameters of the leading wheelset of the front bogie were: the lateral displacement was 8.2 mm, the roll angle was 0.0784°, and the yaw angle was 0.0882°.

[0072] Step 200

[0073] Using the wheel-rail transient rolling contact modeling method, taking the passage of the leading wheelset of the C80 freight car through the circular curve section as the object, a Figure 2 three-dimensional wheel-rail transient rolling contact finite element model as shown was established in ANSYS / Ls-dyna to simulate the wheel-rail transient rolling contact behavior when the wheelset passed through the 800 m radius circular curve section.

[0074] The finite element model considered the real geometry of the wheelset and the rail, and the discrete support structure of the ballast track. The track length was taken as 14.4 m, including 25 sets of fasteners (span 0.6 m). The vehicle only considered the vertical suspension characteristics and ignored the structure above the primary suspension that had little effect on the transient rolling contact behavior (only its mass was considered). The wheelset structure, wheel-rail profile, suspension parameters, speed, superelevation, rail bottom slope, etc. were all the same as those of the vehicle system dynamics model (see Table 1 and Table 2) to ensure comparability.

[0075] Table 2

[0076]

[0077] The wheel-rail is discretized by hexahedron solid elements, taking its three-dimensional real geometry into account. A non-uniform grid is adopted, with the finest mesh on the contact surface (element size 1.0 mm) to ensure the accuracy of contact calculation. The primary suspension is characterized by spring-damping elements, connecting the unsprung mass simulated by mass elements and the axis nodes, and coupling the lateral and longitudinal degrees of freedom of the two ends of the primary suspension nodes to ensure that the unsprung mass moves translationally with the wheel set. For the "fastener-sleeper-ballast" support structure discretized under the rail, vertical spring-damping elements are used to characterize the fasteners and ballast, ignoring the elasticity and mass in the other two directions. The sleeper is discretized by sparse hexahedron solid elements, taking into account its approximate geometry, mass and other characteristics. The total number of elements and nodes in the model is approximately 1.57 million and 1.73 million respectively.

[0078] The "surface-to-surface" contact algorithm based on the penalty function method and Coulomb friction model is used to calculate the normal / tangential contact between the wheel and rail, and the implicit-explicit combined strategy is adopted for time integration. During specific calculations, the wheel set starts to roll forward from the initial position A in Figure 2 the middle. The initial attitude of the wheel set is set according to the above transient wheel-rail contact state parameters. First, the implicit integration method is used to obtain the equilibrium deformation field of the system under the action of gravity, and the subsequent explicit integration calculation is initialized. The length of the dynamic relaxation zone AB is 0.52 m, which attenuates the excitation caused by the initialization of imperfect stress. In the solution zone BC, the wheel set approximately reaches the quasi-steady state rolling. The explicit time integration has the characteristics of conditional stability, and the time step is taken as 5.67×10 -8 s, which is convenient for capturing the transient changes of the wheel-rail contact behavior. An overall coordinate system Oxyz with the origin at the center of the corresponding track of A is established, and x, y, and z are along the longitudinal, lateral, and vertical directions respectively.

[0079] Equation 1-2 is used to calculate the corresponding rigid creepage rate and rigid creep force (hereinafter simply referred to as creepage rate and creep force). Figure 3 Figure 1-2 shows the curves of the creep forces and their angles of the inner rail and outer rail in the vehicle system dynamics model and the three-dimensional wheel-rail transient rolling contact finite element model changing with the rolling distance. As Figure 3 shown, on the premise of considering the dynamic fluctuations, the prediction results of the two models are still roughly consistent. Among them, the creep force predicted by the finite element model on the outer rail side fluctuates around the steady-state value of 29.30 kN predicted by the dynamics model, and the angle with the running direction is around 36.97°. The amplitude of the creep force on the inner rail side is smaller, fluctuating around 24.54 kN, but slightly larger than the prediction result of the dynamics model, and the angle with the running direction is about 187.33°.

[0080] Only using the three-dimensional wheel-rail transient rolling contact finite element model, at the moment of t = 0.0256 s (the wheel set is at the position of x = 647.25 mm in the refined solution area), Equation 3-4 is used to calculate the corresponding local creepage rate and local tangential contact stress. Figures 4 - 5They are the projection diagrams of the local tangential contact stresses of the outer rail and the inner rail in the three-dimensional wheel-rail transient rolling contact finite element model on the xOy plane with respect to the rolling distance and the lateral position. As Figures 4 - 5 shown, the contact patch on the outer rail side is close to the gauge corner side (the center of the rail top y = 741 mm), while the contact patch on the inner rail side is approximately at the center of the rail top (y = -760 mm); the maximum local tangential contact stress on the outer rail side is located at the trailing edge of the contact patch and is biased towards the gauge corner side, about 379 MPa, and the influence of spin is clearly visible; the maximum local tangential contact stress on the inner rail side is also at the trailing edge of the contact patch, but the direction is basically along the longitudinal direction, that is, the spin can be ignored.

[0081] Step 300

[0082] The macroscopic damage function model based on the macroscopic rolling contact solution used in this step is the U75V quenched rail damage function model. Figure 6 is a schematic diagram of the U75V quenched rail damage function model, which is a three-segment linear curve of the transient wear number Tγ and the transient damage value, with three critical points La, Lb, and Lc. Among them, the critical point Lc is the demarcation point between the fatigue initiation and the severe wear state. Fatigue occurs before the critical point Lc, and severe wear will replace fatigue after the critical point Lc. Table 3 shows the transient wear number Tγ and the transient damage value at the three critical points La, Lb, and Lc.

[0083] The transient wear number Tγ and the total transient damage value D within the contact patch are solved at any moment according to the rigid creep rate using Equation 5 and the U75V quenched rail damage function model. t . Figure 7 is the curve of the transient wear number Tγ with respect to the rolling distance in the vehicle system dynamics model and the three-dimensional wheel-rail transient rolling contact finite element model. As Figure 7 shown, the values predicted by the finite element model show significant fluctuations with the contact patch position (proportional to time), but their average values are roughly consistent with the prediction results of the multi-body dynamics model. The difference between the two models is that the prediction results of the finite element model include the fluctuations caused by the vibrations of the system structure, continuum, etc. under non-perfect initial conditions, while in the prediction of the dynamics model, due to the neglect of the interface excitation, the obtained results correspond to the passing of the quasi-steady state curve, that is, they do not include the influence of the above vibrations. Therefore, in order to restore the real wheel-rail contact behavior, the data used in the subsequent steps are all obtained from the three-dimensional wheel-rail transient rolling contact finite element model.

[0084] Then, the total transient damage value D within the contact patch is t discretized transversely along the contact patch according to the semi-elliptical distribution method, that is, the peak value h of the transient damage value is obtained. m .

[0085] Step 400

[0086] Only adopt the three-dimensional wheel-rail transient rolling contact finite element model, and solve the local wear number ργ within the contact patch according to the local creep rate by using Equation 7 L . Figures 8 - 9 They are the local wear numbers ργ of the outer rail and inner rail within the contact patch in the three-dimensional wheel-rail transient rolling contact finite element model respectively L Three-dimensional distribution diagrams with respect to the lateral position and rolling distance. Combining Figures 4 - 5 with Figs. 8-9, it can be seen that at the same moment, the distribution of the local wear number ργ L is consistent with that of the local tangential contact stress. The peak value of the local wear number ργ L within the contact patch on the outer rail side is at the trailing edge of the contact patch and is biased towards the gauge corner side. Among them, the local wear numbers ργ L corresponding to the positions of the three critical points La, Lb, and Lc are 0.69 MPa, 2.21 MPa, and 3.80 MPa respectively

[0087] Step 500

[0088] Take the peak value h m of the transient damage value as the local damage value d L of the maximum local wear number ργ L within the contact patch, and obtain the corresponding relationship diagram between the local wear number ργ L and the peak value h m (abbreviated as peak h in the figure m ) as shown in Figure 10 . Each point in the figure represents a moment. For easy understanding, the parameters of the local damage function model finally proposed by the present invention (see details below) are also shown in Figure 9 . It can be seen from Figure 10 that although the results at each moment have a certain degree of discreteness, the correlation between the peak value h m of the transient damage value and the peak value of the local wear number ργ L is high, which verifies the feasibility of the parameter determination method of the present invention

[0089] Step 600

[0090] The working conditions set in the above steps are typical working conditions (traction coefficient μ = 0), Figures 11 - 12 which are the curves of the peak values of the transient wear number Tγ and the local wear number ργ of the outer rail side and inner rail side in the wheel-rail transient rolling contact finite element model under typical working conditions respectively L changing with the rolling distance. As shown in Figures 11 - 12 , whether it is the outer rail side or the inner rail side, the peak values of the transient wear number Tγ and the local wear number ργ L are basically the same, and the change trends are roughly the same. Compared with the outer rail side, the peak values of the transient wear number Tγ and the local wear number ργ L of the inner rail side are lower

[0091] Specifically, the present invention applies corresponding traction torques to the wheelsets to simulate different adhesion utilization conditions, ensuring that the obtained transient wear number Tγ covers the entire domain of the damage function model of U75V quenched steel rails, and further obtaining the local damage function model parameters based on the mesoscopic contact solution within the contact patch.

[0092] Figure 13 For the transient wear number Tγ and the local wear number τγ on the outer rail side in the finite element model of transient rolling contact between wheel and rail under different traction coefficients μ L The curve of the peak value varying with the rolling distance. Figure 14 For the transient wear number Tγ and the local wear number τγ on the inner rail side and the outer rail side under different traction coefficients μ L The correlation diagram between the peak values, and the correlation coefficient R is calculated by the Pearson linear correlation test method. Combining Figures 13 - 14 It can be seen that the peak values of the transient wear number Tγ and the local wear number τγ on the inner rail side and the outer rail side L Are still correlated under different traction coefficients.

[0093] Figure 15 For the relationship diagram of the peak values of the transient wear number Tγ and the local wear number τγ at all times under different traction conditions, the demarcation points of sections I, II, and III in the figure correspond to the two critical wear numbers of the damage function model of U75V quenched steel rails, namely 23 N and 100 N, and the black solid line is the curve obtained by non-linear fitting with a quartic polynomial. As L Shown, the peak values of the transient wear number Tγ and the local wear number τγ Figure 15 Have a strong correlation. The longitudinal creep force generated on the outer rail side when the wheelset with traction and the wheelset without traction pass through the circular curve section is in the same direction, while it is opposite on the inner rail side. Therefore, the transient wear number Tγ on the outer rail side increases with the increase of the traction coefficient μ, while the transient wear number Tγ on the inner rail side first decreases and then increases, that is, the average coverage range of the transient Tγ on the outer rail side is wider. L

[0094]

[0094] Steps 700 - 800

[0095] Through the above calculation, the total local damage d within the contact patch corresponding to Lc at the critical point t Is not 0. Therefore, in order to fully consider the influence of wear in the LbLc section, the local damage function model based on the mesoscopic contact solution within the contact patch proposed by the present invention can suppress the occurrence of rolling contact fatigue on the premise of considering the lateral displacement of the wheelset, and introduces a wear reduction amount, that is, the local wear number τγ corresponding to Lc at the critical point L Is reduced according to the wear reduction amount until the total local damage d within the corresponding contact patch t Is equal to the total transient damage D within the contact patch t After the reduction calculation, the local wear number τγ corresponding to Lc at the critical pointL It decreased from 3.80 MPa to 3.04 MPa, and finally obtained the local wear numbers τγ L / N and the local damage values as shown in Table 3.

[0096] Table 3

[0097]

[0098] By analogy with the construction of the damage function model of traditional U75V quenched steel rails, the local damage function model proposed in the present invention is also linearly constructed in three segments, that is, the coordinates corresponding to the three critical points (local wear number τγ L / N, local damage value) are connected (the function of the critical points is not constructed in the figure). Figure 16 For the local wear numbers τγ L before and after reduction and the local damage value d L The change curve. As Figure 16 shown, the change curve after reduction (the function after wear reduction is not shown in the figure) is the local damage function model proposed in the present invention based on the mesoscopic contact solution within the contact patch.

[0099] Effect verification: Prediction of fatigue initiation within the contact patch

[0100] Typical transient wear numbers Tγ at different stages in the damage function model of U75V quenched steel rails were selected, and the local damage distribution within the contact patch on the outer rail side was predicted using the above-obtained local damage function model. Figure 17 It is a diagram of the change of the local damage value within the contact patch on the outer rail side predicted by the local damage function model with the transient wear number Tγ. Among them, the local damage values greater than, equal to, and less than 0 correspond to rolling contact fatigue (the corresponding local damage value is called the rolling contact fatigue damage value), slight wear, and severe wear regions respectively, and their area ratios within the contact patch are shown in Table 4. These detailed distribution results within the contact patch cannot be obtained by traditional damage function models (such as the BS11 steel rail damage function model and the U75V quenched steel rail damage function model).

[0101] Table 4

[0102] Transient wear number Tγ 64N 100N 189N 253N 302N 371N Slight wear area 89.22％ 80.64％ 63.51％ 53.64％ 45.45％ 37.45％ Rolling contact fatigue area 10.78％ 19.36％ 35.14％ 38.25％ 40.00％ 38.00％ Severe wear area 0％ 0％ 1.35％ 8.11％ 14.55％ 24.55％

[0103] As Figure 17 and Table 4 show, the area of the rolling contact fatigue region increases with the increase of the transient wear number Tγ. When the transient wear number Tγ is close to or greater than Lc = 253 N, a severe wear region begins to appear inside the rolling contact fatigue region, which can inhibit the occurrence of rolling contact fatigue. It is worth noting that due to the continuity and non-uniformity of the damage value distribution within the contact patch, when the transient wear number Tγ is greater than 253 N, there is still a rolling contact fatigue region within the contact patch. However, the position change of the severe wear region caused by different lateral displacements can inhibit the initiation of rolling contact fatigue.

[0104] Figure 18 The predicted results of the rolling contact fatigue damage values (abbreviated as damage values in the figure) on the surfaces of the inner rail and the outer rail under typical working conditions with respect to the rolling distance (longitudinal) and the lateral position. As Figure 18 shown, the rolling contact fatigue damage values are non-uniformly distributed longitudinally, reflecting the dynamic changes in the wheel-rail contact state, that is, the uneven dissipation of energy on the rail surface, which is consistent with the phenomenon that the distribution of rolling contact fatigue cracks on site is not completely consistent ( Figure 1 ), demonstrating the advantages of the local damage function model of the present invention.

[0105] Figure 19 The distribution diagrams of the rolling contact fatigue damage values (abbreviated as damage values in the figure) of the inner rail and the outer rail predicted by the damage function model and the local damage function model in a certain cross-section under typical working conditions. As Figure 19 shown, the maximum damage values predicted by the two models in the cross-section are basically the same, which to a certain extent proves the rationality of the local damage function model of the present invention. However, the damage function model of U75V quenched rail predicts the phenomenon that there is also damage at the edge of the contact patch where the contact stress is very low, which is obviously inconsistent with the physical law, and this point is corrected in the local damage function model. It can be seen from the figure that the rolling contact fatigue area predicted by the local damage function model is narrower.

[0106] Application: Determination of the parameters of the macroscopic damage function model for any railway system

[0107] In summary, the local damage function model obtained by the present invention can accurately calculate the mesoscopic tangential contact load within the contact patch and can predict the fatigue initiation situation at local positions at a scale closer to the essence of fatigue initiation. The local wear number τγ L reflects the essence of the prediction results of the damage function and is more relevant to the mechanism and life leading to fatigue initiation.

[0108] The results of a large number of wear tests prove that the solid contact friction wear amount / rate is determined by the wear resistance of the material itself in addition to being related to the energy input to the contact interface. The present invention believes that the parameters of the local damage function model are only related to the wheel-rail material properties and are the mesoscopic expression of the macroscopic damage function model. Therefore, based on the characteristics and calculation logic of the local damage function model of the present invention, it can be reasonably inferred that it is reasonable and practical to determine the damage function of another railway system (system B) with no obvious change in the wheel-rail material characteristics according to the damage function parameters of a known railway system (system A). The specific implementation method is as follows:

[0109] (1) Obtain the key parameters of the local damage function model calibrated by system A according to the macroscopic damage function model according to steps 100 to 700.

[0110] (2) Conduct the same simulation work within System B, and reverse-infer the macroscopic damage function model parameters applicable to System B based on the local damage function model obtained in step (1).

[0111] The above describes the relevant content of the present invention. Those of ordinary skill in the art will be able to implement the present invention based on these descriptions. Based on the above content of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

Claims

1. A method for determining parameters of wheel-rail rolling contact fatigue damage function, characterized in that: The following steps are involved: Step 100, establishing a vehicle system dynamics model of a railway system and solving transient wheel-rail contact state parameters; Step 200, based on the transient wheel-rail contact state parameters, a three-dimensional wheel-rail transient rolling contact finite element model is established to calculate a macroscopic rolling contact solution and an intra-spot microscopic contact solution; the macroscopic rolling contact solution includes a rigid creep rate and a rigid creep force at the wheel-rail contact interface; the intra-spot microscopic contact solution includes a local creep rate and a local tangential contact stress within the wheel-rail contact spot; Step 300, using a macro damage function model based on a macro rolling contact solution, solve the transient wear number Tγ and the total transient damage value D in the contact spot at any time according to the rigid creep rate. t and the peak value of transient damage h m ; Step 400, solving the local wear number τγ in the contact spot according to the local creep rate and the tangential contact stress L ; Step 500, taking the transient damage value peak h m As the maximum local wear number τγ in the contact patch L The local damage value d L , and the local wear number τγ is obtained L The peak value of transient damage h m The corresponding relationship; Step 600, by changing the working conditions, obtain the transient wear number Tγ and the local wear number τγ covering the entire domain of the macro damage function model under different traction coefficients L Relationship between Step 700: calculate the local wear number τγ at the critical point Lc L According to the wear reduction amount, the reduction is reduced until the corresponding total value of local damage in the contact spot is d t The total value of transient damage in the contact spot D t Equal to obtain the reduced local wear number τγ L and local damage value d L relationship; Step 800, fitting to obtain local damage function model parameters of a railway system based on intra-spot microscopic contact solution and / or inferring macroscopic damage function model parameters of another railway system.

2. The method for determining the parameters of the wheel-rail rolling contact fatigue damage function according to claim 1, characterized in that: In step 100, a vehicle system dynamics model is established in the multi-body dynamics software SIMPACK.

3. The method for determining the parameters of the wheel-rail rolling contact fatigue damage function according to claim 1, characterized in that: In step 200, a three-dimensional wheel-rail transient rolling contact finite element model is established in the finite element solver ANSYS / Ls-dyna.

4. The method for determining the parameters of the wheel-rail rolling contact fatigue damage function according to claim 3, characterized in that: In step 200, the calculation expression of the rigid creep rate is: The calculation expression of local creep rate is: In the formula, γ x is the longitudinal rigid creep rate; γ y is the lateral rigid creep rate; γ Lx is the local creep rate in the longitudinal direction; γ Ly is the lateral local creep rate; ω is the wheel speed; r is the actual rolling circle radius of the selected tread output point set; v0 is the longitudinal speed of the wheel axis; v y is the lateral moving speed of the wheel relative to the rail; Δv x is the longitudinal relative sliding velocity between the wheel and rail at any point in the contact patch; Δv y is the lateral relative sliding velocity between the wheel and rail at any point in the contact patch.

5. The method for determining the parameters of the wheel-rail rolling contact fatigue damage function according to claim 4, characterized in that: In step 300, the calculation expression of the transient wear number Tγ is: Transient damage peak value h m The calculation expression is: Where, T x is the longitudinal rigid creep force; T y is the lateral rigid creep force; b is the lateral semi-axis length of the contact spot, and π is the pi.

6. The method for determining the parameters of the wheel-rail rolling contact fatigue damage function according to claim 4, characterized in that: In step 400, the local wear number τγ L The calculation expression is: Tg L =t x c Lx +t y c Ly ; In the formula, τ x is the longitudinal component of the local tangential contact stress; τ y is the transverse component of the local tangential contact stress.

Images