A modeling method for continuous four-dimensional evolution of wheel-rail profile throughout its life cycle

Abstract

The present invention discloses a method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its entire life cycle, comprising: constructing a standard wheel-rail discrete voxelized model based on wheel-rail design data; collecting wheel-rail profile data in different service cycles based on a non-contact two-dimensional laser displacement sensor; predicting wheel-rail profile wear data in different service cycles based on a multi-body dynamics simulation model; mapping measured wheel-rail discrete profile data / wheel-rail profile simulation data to the standard wheel-rail discrete voxelized model to obtain the wheel-rail voxelized model evolution velocity field; the method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its entire life cycle proposed by the present invention can realize the continuous evolution of the physical geometric model of the serving wheel-rail throughout its entire life cycle, provides a continuous modeling method for the wheel-rail service process morphology, can effectively explain the profile evolution mechanism during the wheel-rail service process, provides theoretical support for wheel-rail profile design and optimization, and improves the physical performance and service cycle of the product during the entire service life cycle.

Description

Technical Field

[0001] The present invention relates to the technical field of rail vehicle wheel and rail profile modeling, and in particular to a full life cycle continuous four-dimensional evolution modeling method for a wheel and rail profile. Background Art

[0002] As rail transit vehicles develop towards high speed and heavy load, wheel-rail profiles will inevitably wear out during service, and their contact relationship will gradually deteriorate as material wear intensifies. Wheel-rail profile wear not only affects vehicle operation stability, safety and passenger comfort, but also significantly increases vehicle structural vibration, increases wheel-rail noise, and shortens the service life of key vehicle components.

[0003] Wheel-rail surface wear is a continuous process from a standard profile to a severely worn profile, and the geometry of the wheel-rail surface changes continuously throughout its life cycle. However, the current geometric modeling of wheel-rail surface products is still at the static, ideal two-dimensional or three-dimensional model stage, and is unable to reveal the dynamic evolution of the geometry and physical properties of wheel-rail surface products.

[0004] However, the existing geometric modeling methods of wheel-rail profile products are unable to achieve continuous evolution modeling of wheel-rail profiles, and it is difficult to effectively explain the evolution mechanism of the physical properties of wheels and rails during service. Summary of the invention

[0005] The main purpose of the present invention is to provide a method for continuous four-dimensional evolution modeling of wheel-rail profile throughout its life cycle, aiming to solve the problem that the existing wheel-rail profile product geometric modeling method cannot achieve continuous evolution modeling of wheel-rail profile.

[0006] The technical solution proposed by the present invention is:

[0007] A modeling method for continuous four-dimensional evolution of wheel-rail profile throughout its life cycle, comprising:

[0008] Construct a standard wheel-rail discrete voxel model based on wheel-rail design data;

[0009] A non-contact two-dimensional laser displacement sensor is used to collect the measured discrete wheel-rail profile data on the wheel-rail surface during different service cycles.

[0010] Based on the vehicle system dynamics model, wheel-rail contact mechanics model and wheel-rail material contact wear model, obtain the simulation data of wheel-rail profile evolution during the service life;

[0011] Based on the measured wheel-rail discrete profile data, the velocity field function of the wheel-rail measured model evolution is constructed;

[0012] Based on the simulation data of wheel-rail profile evolution during the service life, the evolution velocity field function of the wheel-rail simulation model is constructed;

[0013] Based on the standard wheel-rail discrete voxelized model, the evolution velocity field function of the wheel-rail measured model and the evolution velocity field function of the wheel-rail simulation model, the level set editing method is used to perform continuous four-dimensional evolution of the wheel-rail profile on the service wheel-rail discrete voxelized model.

[0014] Preferably, the method further comprises: based on the standard wheel-rail discrete voxelized model, the wheel-rail measured model evolution velocity field function and the wheel-rail simulation model evolution velocity field function, a level set editing method is used to realize the continuous four-dimensional evolution of the service wheel-rail discrete voxelized model, and then further comprises:

[0015] The marching cubes algorithm is used to realize the geometric modeling and visualization expression of the discrete voxelized model of the in-service wheel-rail after continuous four-dimensional evolution.

[0016] Preferably, the construction of a standard wheel-rail discrete voxel model based on wheel-rail design data comprises:

[0017] Based on the wheel-rail design data, the wheel-rail physical design model is constructed by using the physical modeling design method;

[0018] Extract wheel-rail geometric surface model from wheel-rail entity design model;

[0019] The standard wheel-rail discrete voxel model is constructed based on the wheel-rail geometric surface model using the geometric surface discretization method.

[0020] Preferably, the step of extracting the wheel-rail geometric surface model from the wheel-rail entity design model comprises:

[0021] Selecting the surface geometric features of the wheel-rail entity design model, and converting the surface features of the wheel-rail entity design model into a triangular mesh surface model through triangular mesh discretization;

[0022] The method of using a geometric surface discretization method to construct a standard wheel-rail discrete voxel model based on a wheel-rail geometric surface model comprises:

[0023] The voxelization method is used to discretize the triangular mesh surface model into a standard wheel-rail discrete voxel model, and the position relationship between each voxel vertex in the standard wheel-rail discrete voxel model and the triangular mesh surface model is calculated. The calculation formula is:

[0024]

[0025] Where, the voxel vertices in the standard wheel-rail discrete voxel model are p(x, y, z), the interior of the standard wheel-rail discrete voxel model is Ω, and the exterior of the standard wheel-rail discrete voxel model is Ω c The surface of the standard wheel-rail discrete voxel model is And satisfy disp(p)=min(|pp I |), that is, min(|pp I|) is the shortest distance between each voxel vertex in the standard wheel-rail discrete voxel model and the triangular mesh surface model;

[0026] For the data of the standard wheel-rail discrete voxel model, the voxel vertices within the preset range of the zero level set surface model are assigned as active points in a narrowband manner, and the rest are inactive vertices.

[0027] The sparse dynamic OpenVDB data structure is used to store the vertex coordinates and distance attribute data of the standard wheel-rail discrete voxel model to achieve voxel discretization expression of the standard wheel-rail profile.

[0028] Preferably, the method of collecting the measured wheel-rail discrete profile data of the wheel-rail surface in different service cycles by using a non-contact two-dimensional laser displacement sensor comprises:

[0029] A non-contact two-dimensional laser displacement sensor is used to continuously collect displacement data of the service wheel-rail surface to obtain discrete wheel-rail profile data of the wheel-rail surface in different service cycles, wherein the discrete wheel-rail profile data includes three-dimensional point cloud data of the wheel-rail profile;

[0030] The nearest neighbor point pair matching method is used to adjust the measured wheel-rail profile pose to achieve registration of discrete wheel-rail profile data in different coordinate systems.

[0031] The registered wheel-rail discrete profile data mapping is loaded into the standard wheel-rail discrete voxelized model to obtain the service wheel-rail discrete voxelized model.

[0032] Preferably, the method of adjusting the measured wheel-rail profile pose by using the nearest neighbor point pair matching method to achieve registration of discrete wheel-rail profile data in different coordinate systems includes:

[0033] Based on the rotation and translation relationship between the laser displacement sensor coordinate system and the world coordinate system, the wheel-rail discrete profile data is transformed to achieve the registration of the wheel-rail discrete profile data in different coordinate systems. The expression formula of the rotation and translation relationship between the laser displacement sensor coordinate system and the world coordinate system is:

[0034]

[0035] Wherein, M(mx,my,mz) is the designed wheel-rail profile data, N(nx,ny,nz) is the measured wheel-rail profile three-dimensional point cloud data, and the expression formula is composed of the rotation transformation matrix R(α,β,γ) and the translation vector t(t x ,t y ,t z ) is specifically expressed as:

[0036]

[0037] Preferably, the step of acquiring the profile evolution simulation data of the wheel-rail service cycle based on the vehicle system dynamics model, the wheel-rail contact mechanics model and the wheel-rail material contact wear model further comprises:

[0038] Construct a dynamic model of the vehicle system with wheel-rail coupling;

[0039] Calculate the wheel-rail contact geometry parameters during rail vehicle operation based on the vehicle system dynamics model;

[0040] A wheel-rail contact mechanics model is constructed to calculate the local contact parameters during the wheel-rail service process;

[0041] According to the physical properties of wheel-rail contact materials and wheel-rail contact force, a wheel-rail material contact wear model is constructed.

[0042] Preferably, the construction of a wheel-rail coupled vehicle system dynamics model comprises:

[0043] Obtain the numbers of the two-axle bogies and the car body of the rail vehicle to build a vehicle system dynamics model;

[0044] The wheel-rail contact mechanical model is constructed to calculate the local contact parameters of the wheel-rail during service, including:

[0045] The wheel-rail normal contact force during the service life of the wheel-rail is calculated using the following formula:

[0046]

[0047] Where x, y are the longitudinal and transverse coordinates of the wheel-rail contact spot, a, b are the semi-axis values ​​of the elliptical contact spot;

[0048] The Kalker simplified theory Fastsim algorithm is used to calculate the creep force of the wheel-rail during service. The calculation formula is:

[0049]

[0050] In the formula, Δp x (x,y) and Δp y (x, y) are the tangential stress increments along the longitudinal and transverse directions, respectively, expressed as:

[0051]

[0052] In the formula, ξ x is the longitudinal creep rate, ξ y is the lateral creep rate, ξ z is the spin creep rate, which is determined by the vehicle system dynamics model, and L is the tangential compliance coefficient, which is determined by the table lookup method;

[0053] The wheel-rail material contact wear model is constructed according to the physical properties of the wheel-rail contact material and the wheel-rail contact force, including:

[0054] The wheel-rail material contact wear model is constructed based on the Archard wear model, wherein the material loss of the wheel-rail material contact wear model is proportional to the creep distance and the normal force, and inversely proportional to the material hardness. The wheel-rail material contact wear model is expressed as:

[0055]

[0056] Where k(x,y) is the inner wheel-rail friction coefficient of the contact spot unit (x,y), p z (x,y) is the normal contact stress of the contact spot unit (x,y), and s(x,y) is the sliding distance of the contact spot unit (x,y).

[0057] Preferably, the step of constructing the measured wheel-rail model evolution velocity field function based on the measured wheel-rail discrete profile data comprises:

[0058] Obtain the wheel-rail coordinate data and rail coordinate data mapped to the distance field of the standard wheel-rail discrete voxelized model, and use the nearest neighbor search algorithm to search the nearest neighbor points of the voxel vertices to the in-service wheel point cloud and rail point cloud to construct the nearest neighbor point pairs;

[0059] The velocity vector of voxel vertex evolution is constructed according to the relative position of the point pair in space, so as to construct the velocity field function of wheel-rail measured model evolution;

[0060] The method of constructing the wheel-rail simulation model evolution velocity field function based on the wheel-rail profile evolution simulation data during the wheel-rail service period includes:

[0061] Acquiring simulated data of wheel-rail profile evolution during a service life to predict a wheel-rail profile, and further mapping the wheel-rail profile into a discrete voxel distance field;

[0062] The wear amount at each position of the wheel-rail profile is obtained, and the zero isosurface velocity field value is constructed based on the wear amount, and the evolution velocity is further expanded according to the voxel vertex gradient value to construct the wheel-rail simulation model evolution velocity field function.

[0063] Preferably, the wheel-rail profile of the service wheel-rail discrete voxelized model is continuously evolved in four dimensions using a level set editing method based on the standard wheel-rail discrete voxelized model, the wheel-rail measured model evolution velocity field function and the wheel-rail simulation model evolution velocity field function, including:

[0064] The distance field of the standard wheel-rail discrete voxelized model is superimposed with the evolution velocity field of the wheel-rail measured model and the evolution velocity field of the wheel-rail simulation model;

[0065] The wheel / rail profile of the evolved service wheel / rail discrete voxel model is updated by using the level set method. After a preset number of evolution steps, the distance field value of the standard wheel / rail discrete voxel model is recalculated and a narrow band is established.

[0066] The method of updating the wheel-rail profile of the discrete voxelized model of the wheel-rail in service by using the level set method comprises:

[0067] The level set method is used to perform continuous four-dimensional evolution of the wheel-rail profile on the standard wheel-rail discrete voxel model. The evolution formula is:

[0068] φ(p(x,y,z),t)=0,

[0069] In the formula, the evolution formula can be obtained by using the chain differential rule:

[0070]

[0071] The distance field of the standard wheel-rail discrete voxel model is initialized by solving the steady-state solution of the partial differential equation, where the initialization operation formula is:

[0072]

[0073] By determining the voxel distance field value near the zero level set, the continuous four-dimensional evolution of the wheel and rail profile of the service wheel and rail discrete voxel model is realized;

[0074] The method of recalculating the distance field value of the standard wheel-rail discrete voxelized model after a preset number of evolution steps and establishing a narrow band includes:

[0075] Taking the zero level set surface as the center, determine the narrow band local area with a width of 2δ around the zero level set surface;

[0076] The voxel vertices in the marked narrow band area are activated points, and the other voxel vertices are inactivated points. During the continuous four-dimensional evolution of the wheel-rail profile of the service wheel-rail discrete voxelization model, only the level set function values ​​of the activated points are iteratively updated.

[0077] Through the above technical solution, the following beneficial effects can be achieved:

[0078] The method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle proposed in the present invention can realize the continuous evolution of the physical geometric model of the serving wheel-rail throughout its life cycle, and provides a continuous modeling method for the morphology of the wheel-rail service process. It can effectively explain the evolution mechanism of the physical properties of the wheel-rail service process, provide theoretical support for wheel-rail profile design and optimization, shorten the wheel-rail profile design optimization cycle, and improve the physical performance and service cycle of the product during its entire life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0080] Figure 1 A flowchart of a first embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention;

[0081] Figure 2 A result diagram of the distance field of the rail profile evolution process in the first embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention;

[0082] Figure 3 A result diagram of the velocity field of the rail profile evolution process in the first embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention;

[0083] Figure 4 A voxel model diagram of the rail profile evolution process in the first embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention;

[0084] Figure 5 The error information between the geometric model of the rail evolution process and the expected result in the first embodiment of the method for modeling the continuous four-dimensional evolution of the wheel-rail profile throughout its life cycle proposed by the present invention;

[0085] Figure 6 A result diagram of the distance field of the wheel-rail profile evolution process in the second embodiment of the method for modeling the continuous four-dimensional evolution of the wheel-rail profile throughout its life cycle proposed by the present invention;

[0086] Figure 7 This is a result diagram of the velocity field during the wheel-rail profile evolution process in the second embodiment of a method for modeling a continuous four-dimensional evolution of wheel-rail profile throughout its life cycle proposed by the present invention;.

[0087] Figure 8 A voxel model diagram of the wheel-rail profile evolution process in the second embodiment of the wheel-rail profile full life cycle continuous four-dimensional evolution modeling method proposed by the present invention;

[0088] Fig. 9 It is the error information between the geometric model of the wheel evolution process and the expected result in the second embodiment of the method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle proposed by the present invention. DETAILED DESCRIPTION

[0089] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0090] The present invention proposes a modeling method for continuous four-dimensional evolution of wheel-rail profile throughout its entire life cycle.

[0091] As attached Figure 1 As shown, in a first embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, this embodiment includes the following steps:

[0092] Step S110: constructing a standard wheel-rail discrete voxel model based on the wheel-rail design data.

[0093] Step S120: using a non-contact two-dimensional laser displacement sensor to collect the measured wheel-rail discrete profile data of the wheel-rail surface in different service cycles.

[0094] Step S130: Based on the vehicle system dynamics model, the wheel-rail contact mechanics model and the wheel-rail material contact wear model, the profile evolution simulation data within the wheel-rail service cycle is obtained.

[0095] Step S140: constructing a velocity field function of the measured wheel-rail model evolution based on the measured wheel-rail discrete profile data.

[0096] Step S150: constructing a wheel-rail simulation model evolution velocity field function based on the wheel-rail profile evolution simulation data during the wheel-rail service period.

[0097] Step S160: Based on the standard wheel-rail discrete voxelized model, the wheel-rail measured model evolution velocity field function and the wheel-rail simulation model evolution velocity field function, the level set editing method is used to perform continuous four-dimensional evolution of the wheel-rail profile on the in-service wheel-rail discrete voxelized model.

[0098] The method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle proposed in the present invention can realize the continuous evolution of the physical geometric model of the serving wheel-rail throughout its life cycle, and provides a continuous modeling method for the morphology of the wheel-rail service process. It can effectively explain the evolution mechanism of the physical properties of the wheel-rail service process, provide theoretical support for wheel-rail profile design and optimization, shorten the wheel-rail profile design optimization cycle, and improve the physical performance and service cycle of the product during its entire life cycle.

[0099] In a second embodiment of a method for modeling a continuous four-dimensional evolution of a wheel / rail profile throughout its life cycle proposed by the present invention, based on the first embodiment, step S160 further includes the following steps:

[0100] Step S210: using the marching cubes algorithm to achieve geometric modeling and visual expression of the discrete voxelized model of the in-service wheel-rail after continuous four-dimensional evolution.

[0101] In a third embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the first embodiment, step S110 includes the following steps:

[0102] Step S310: Based on the wheel-rail design data, a wheel-rail physical design model is constructed using a physical modeling design method.

[0103] Step S320: extracting the wheel-rail geometric surface model from the wheel-rail entity design model.

[0104] Step S330: constructing a standard wheel-rail discrete voxel model based on the wheel-rail geometric surface model using a geometric surface discretization method.

[0105] This embodiment improves the specific steps of step S110 in the first embodiment, and its purpose is to extract the geometric surface model through the wheel-rail entity design model.

[0106] In a fourth embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the third embodiment, step S320 includes the following steps:

[0107] Step S410: Selecting surface geometric features of the wheel-rail entity design model, and converting the surface features of the wheel-rail entity design model into a triangular mesh surface model by triangular mesh discretization.

[0108] Step S330 includes the following steps:

[0109] Step S420: discretize the triangular mesh surface model into a standard wheel-rail discrete voxel model by using a voxelization method, and calculate the positional relationship between each voxel vertex in the standard wheel-rail discrete voxel model and the triangular mesh surface model. The calculation formula is:

[0110]

[0111] Where, the voxel vertices in the standard wheel-rail discrete voxel model are p(x, y, z), the interior of the standard wheel-rail discrete voxel model is Ω, and the exterior of the standard wheel-rail discrete voxel model is Ω c The surface of the standard wheel-rail discrete voxel model is And satisfy disp(p)=min(|pp I |), that is, min(|pp I |) is the shortest distance between each voxel vertex in the standard wheel-rail discrete voxel model and the triangular mesh surface model;

[0112] Step S420: For the data of the standard wheel-rail discrete voxelized model, a narrowband method is used to assign voxel vertices within a preset range of the zero level set surface model (for example, the preset range is three voxel widths, that is, the zero level set and the inner and outer narrowband widths are all three voxels in size) as activation points, and the rest are inactive vertices.

[0113] Step S430: using a sparse dynamic OpenVDB data structure to store vertex coordinates and distance attribute data of the standard wheel-rail discrete voxel model, so as to achieve voxel discretization expression of the standard wheel-rail profile.

[0114] In a fifth embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the first embodiment, step S120 includes the following steps:

[0115] Step S510: using a non-contact two-dimensional laser displacement sensor to continuously collect displacement data of the service wheel-rail surface to obtain discrete wheel-rail profile data of the wheel-rail surface in different service cycles, wherein the discrete wheel-rail profile data includes three-dimensional point cloud data of the wheel-rail profile;

[0116] Step S520: using the nearest neighbor point pair matching method to adjust the measured wheel-rail profile pose to achieve registration of the wheel-rail discrete profile data in different coordinate systems;

[0117] Step S530: Loading the registered wheel-rail discrete profile data mapping into the standard wheel-rail discrete voxelized model to calculate the measured wheel-rail profile evolution velocity field.

[0118] Specifically, this embodiment improves step S120. The purpose of step S120 is to realize coordinate system 1 through point cloud registration based on the in-service wheel-rail profile data measured by the two-dimensional laser displacement sensor, and map the measured wheel-rail profile data after registration to the discrete voxel model of the standard wheel-rail.

[0119] In a sixth embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the fifth embodiment, step S520 includes the following steps:

[0120] Step S610: Based on the rotation and translation relationship between the laser displacement sensor coordinate system and the world coordinate system, coordinate transformation is performed on the wheel-rail discrete profile data to achieve registration of the wheel-rail discrete profile data in different coordinate systems, wherein the expression formula for the rotation and translation relationship between the laser displacement sensor coordinate system and the world coordinate system is:

[0121]

[0122] Wherein, M(mx,my,mz) is the designed wheel-rail profile data, N(nx,ny,nz) is the measured wheel-rail profile three-dimensional point cloud data, and the expression formula is composed of the rotation transformation matrix R(α,β,γ) and the translation vector t(t x ,t y ,t z ) is specifically expressed as:

[0123]

[0124] In the seventh embodiment of the method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle proposed by the present invention, based on the first embodiment, step S130 also includes the following steps:

[0125] Step S710: constructing a wheel-rail coupled vehicle system dynamics model based on vehicle system dynamics software.

[0126] Step S720: Calculating wheel-rail contact geometric parameters during the operation of the rail vehicle based on the vehicle system dynamics model.

[0127] Specifically, the wheel-rail contact geometric parameters here include contact point, creep rate, etc.

[0128] Step S730: constructing a wheel-rail contact mechanics model to calculate local contact parameters during the wheel-rail service process.

[0129] Specifically, the local contact parameters here include wheel-rail normal contact force and creep force.

[0130] Step S740: constructing a wheel-rail material contact wear model according to the physical properties of the wheel-rail contact material and the wheel-rail contact force.

[0131] Specifically, the physical properties here include material hardness, friction coefficient, etc.

[0132] In the eighth embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the seventh embodiment, step S710 comprises the following steps:

[0133] Step S810: Obtain the number of two-axle bogies and car bodies of the rail vehicle to construct a vehicle system dynamics model.

[0134] Specifically, in this embodiment, the vehicle is composed of two two-axle bogies and one car body, so the constructed vehicle system dynamics model includes one car body, two frames, four wheel pairs, and eight axle boxes, a total of 15 rigid bodies, and the constructed vehicle system dynamics model includes primary suspension and secondary suspension.

[0135] Step S720 includes the following steps:

[0136] Step S820: Calculate the wheel-rail normal contact force during the wheel-rail service process, and the calculation formula is:

[0137]

[0138] Where x, y are the longitudinal and transverse coordinates of the wheel-rail contact spot, a, b are the semi-axis values ​​of the elliptical contact spot;

[0139] Step S830: The Kalker simplified theory Fastsim algorithm is used to calculate the creep force of the wheel-rail during service. The calculation formula is:

[0140]

[0141] In the formula, Δp x (x,y) and Δp y (x, y) are the tangential stress increments along the longitudinal and transverse directions, respectively, expressed as:

[0142]

[0143] In the formula, ξ x is the longitudinal creep rate, ξ y is the lateral creep rate, ξ z is the spin creep rate, which is determined by the vehicle system dynamics model, and L is the tangential compliance coefficient, which is determined by the table lookup method;

[0144] Step S740 includes the following steps:

[0145] Step S840: constructing the wheel-rail material contact wear model based on the Archard wear model, wherein the material loss of the wheel-rail material contact wear model is proportional to the creep distance and the normal force, and inversely proportional to the material hardness. The wheel-rail material contact wear model is expressed as:

[0146]

[0147] Where k(x,y) is the inner wheel-rail friction coefficient of the contact spot unit (x,y), p z (x,y) is the normal contact stress of the contact spot unit (x,y), and s(x,y) is the sliding distance of the contact spot unit (x,y).

[0148] Specifically, the wheel-rail profile update and wear prediction are realized by constructing a vehicle-rail coupling dynamics physical simulation model. First, a vehicle dynamics model is established, and then a wheel-rail profile contact mechanics model is constructed to realize the calculation of the wheel-rail normal contact force and creep force during vehicle movement. Then, a wheel-rail contact material wear model is established to determine the amount of wheel-rail profile material wear at a certain operating mileage of the vehicle, and the wheel-rail profile evolution velocity field is determined according to the amount of profile wear. Then, the continuous four-dimensional evolution modeling of the service wheel-rail is realized through the overall evolution idea shown in the first embodiment.

[0149] In a ninth embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the first embodiment, step S140 comprises the following steps:

[0150] Step S910: Obtain wheel coordinate data and rail coordinate data mapped to the standard wheel-rail discrete voxelized model distance field from the measured wheel-rail discrete profile data, and use the nearest neighbor search algorithm to search for the nearest neighbor points of the voxel vertices to the in-service wheel point cloud and rail point cloud to construct neighbor point pairs.

[0151] Step S920: construct the velocity vector of the voxel vertex evolution according to the relative position of the point pair in space, so as to construct the velocity field function of the wheel-rail measured model evolution.

[0152] Step S150 includes the following steps:

[0153] Step S930: Acquire the profile evolution simulation data within the wheel-rail service period to predict the wheel-rail profile, and further map the wheel-rail profile into a discrete voxel distance field.

[0154] Step S940: Obtain the wear amount at each position of the wheel-rail profile, and construct a zero isosurface velocity field value based on the wear amount, and further expand the evolution velocity according to the voxel vertex gradient value to construct an evolution velocity field function of the wheel-rail simulation model.

[0155] Specifically, the surface profile data points of the in-service wheel and rail are obtained through the vehicle profile detection system and the track profile measurement system, and then the iterative closest point method is used to achieve wheel / rail profile matching in different coordinate systems. Then, the wheel / rail profile in the unified coordinate system is mapped to a discrete voxelized model to construct a wheel-rail evolution velocity field, and then the overall evolution idea shown in the first embodiment is used to achieve continuous four-dimensional evolution modeling of the in-service wheel and rail.

[0156] In the tenth embodiment of a method for modeling a continuous four-dimensional evolution of a wheel-rail profile throughout its life cycle proposed by the present invention, based on the first embodiment, step S160 comprises the following steps:

[0157] Step S1010: superimposing the distance field of the standard wheel-rail discrete voxelized model with the evolution velocity field of the wheel-rail measured model and the evolution velocity field of the wheel-rail simulation model.

[0158] Step S1020: The wheel / rail profile of the evolved service wheel / rail discrete voxelized model is updated using the level set method. After a preset number (e.g., 5) of evolution steps, the value of the distance field of the standard wheel / rail discrete voxelized model is recalculated and a narrow band is established to ensure the distance attribute of the voxel vertices during the wheel / rail evolution process and improve the level set update efficiency.

[0159] The step S1020 of updating the wheel-rail profile of the evolving service wheel-rail discrete voxel model using the level set method comprises the following steps:

[0160] Step S1030: Using the level set method to perform continuous four-dimensional evolution of the standard wheel-rail discrete voxel model, the evolution formula is:

[0161] φ(p(x,y,z),t)=0,

[0162] In the formula, in order to clearly express the voxel model evolution process, the evolution formula is obtained by using the chain differential rule:

[0163]

[0164] It can be seen that by defining the distance field and velocity field required for wheel-rail evolution, a continuous four-dimensional wheel-rail evolution model can be constructed.

[0165] Step S1040: In order to maintain the distance field properties of the voxel vertices of the level set evolution, the distance field of the standard wheel-rail discrete voxel model is initialized by solving the steady-state solution of the partial differential equation, wherein the initialization operation formula is:

[0166]

[0167] Step S1050: The wheel-rail evolution process is actually the evolution of the zero level set in the level set equation. Therefore, the continuous four-dimensional evolution of the wheel-rail profile of the service wheel-rail discrete voxelization model can be achieved by determining the voxel distance field value near the zero level set.

[0168] The step S1020 of recalculating the distance field value of the standard wheel-rail discrete voxelized model after a preset number of evolution steps and establishing a narrowband comprises the following steps:

[0169] Specifically, in order to improve the efficiency of wheel-rail updating by level set, the narrowband level set method is used to realize the evolution of wheel-rail profile. The specific steps of narrowband construction are as follows:

[0170] Step S1060: Taking the zero level set surface as the center, determine a narrowband local region with a width of 2δ around the zero level set surface.

[0171] Specifically, the widths of the inner and outer sides are δ respectively.

[0172] Step S1070: Mark the voxel vertices in the narrowband area as active points, and the remaining voxel vertices as inactive points. In the process of continuous four-dimensional evolution of the wheel-rail profile of the in-service wheel-rail discrete voxelized model, only the level set function values ​​of the active points are iteratively updated.

[0173] In the eleventh embodiment of the method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle proposed by the present invention, based on the second embodiment, step S210 comprises the following steps:

[0174] Step S1110: The discrete voxelized model of the in-service wheel-rail in the continuous four-dimensional evolution process is reconstructed using the marching cubes algorithm, and the distance field and the velocity field in the evolution process are visualized using cloud maps and vector arrows, respectively, to verify whether the continuous four-dimensional evolution process of the discrete voxelized model of the in-service wheel-rail evolves in the expected manner.

[0175] Step S1120: Based on the distance field properties of the in-service wheel-rail discrete voxelized model during the continuous four-dimensional evolution process, the process of continuous four-dimensional evolution of the wheel-rail profile of the in-service wheel-rail discrete voxelized model is corrected.

[0176] Step S1130: using a cloud diagram to check whether the final result of the continuous four-dimensional evolution of the wheel-rail profile of the discrete voxelized model of the in-service wheel-rail meets the expected evolution accuracy requirement.

[0177] The overall flow chart of the present invention is shown in the first embodiment and the second embodiment. The main idea is: first, the standard wheel-rail geometric surface model is discretized into voxels to construct a voxelized model with distance field attributes, then the measured wheel-rail service profile data / multi-body dynamics physical simulation data are mapped to the distance field, and the velocity field operator required for level set tracking of wheel-rail evolution is constructed, and then the level set is updated by superimposing the distance field and the velocity field. After a certain number of evolutions, the voxel model distance field is recalculated and a narrow band is constructed, and then the wheel-rail profile is continuously evolved in four dimensions. Finally, the geometric surface of the voxel model is reconstructed by the marching cube algorithm to achieve visual expression of the service wheel-rail model.

[0178] Embodiment 1:

[0179] This paper takes 50 rails currently in service as an example to further illustrate the modeling effect of this method on rail evolution. The schematic results are shown in Figure 2-5 shown.

[0180] in Figure 2(a) is the initial distance field, where (a) is the initial distance field, (b)-(f) are the distance fields of the evolution model after 100, 200, 300, 400 and 500 iterations, respectively; Figure 2 (b)-(f) are the distance fields of the evolution model with 100 to 500 iteration steps, respectively.

[0181] Figure 3 (a) is the initial velocity field, Figure 3 (b)-(f) are the velocity fields of the evolution model with 100-500 iteration steps, respectively. It can be seen that as the number of evolution steps increases, the model evolution velocity field tends to 0. That is, (a) is the initial velocity field, and (b)-(f) are the velocity fields of the evolution model with 100, 200, 300, 400 and 500 iteration steps, respectively.

[0182] Figure 4 (a) is the initial voxel model, Figure 4 (b)-(f) are voxelized models evolved at 100-500 iterations, respectively; that is, (a) is the initial voxel model, and (b)-(f) are voxelized models evolved at 100, 200, 300, 400 and 500 iterations, respectively.

[0183] Figure 5 The error information between the reconstructed model and the expected evolution result during the evolution process is compared, that is, (a) is the error information between the initial geometric model and the expected result, and (b)-(d) are the error information of different evolution processes; the geometric model evolution error is within 0-2mm. It can be seen that as the number of evolution steps increases, the reconstructed geometric model is closer to the expected evolution result, and the cloud map of the final evolution result shows that the evolved geometric model is consistent with the expected evolution result. It can be seen that the method of the present invention can realize the continuous evolution modeling of rail profile.

[0184] Embodiment 2:

[0185] This paper takes the JM3 type wheel currently in service as an example to further illustrate the effect of this method in the wheel evolution modeling process. The results are as follows: Figure 6-9 shown.

[0186] in Figure 6 (a) is the initial distance field, Figure 6 (b)-(f) are the distance fields of the evolution model with iterations of 100-500 steps respectively; that is, (a) is the initial distance field, and (b)-(f) are the distance fields of the evolution model with iterations of 100 steps, 200 steps, 300 steps, 400 steps and 500 steps respectively.

[0187] Figure 7 (a) is the initial velocity field, Figure 7(b)-(f) are the velocity fields of the evolution model with iterations of 100 to 500 steps, respectively; that is, (a) is the initial velocity field, and (b)-(f) are the velocity fields of the evolution model with iterations of 100, 200, 300, 400 and 500 steps, respectively.

[0188] Figure 8 (a) is the initial voxel model, Figure 8 (b)-(f) are voxelized models evolved at 100-500 iterations, respectively; that is, (a) is the initial voxel model, and (b)-(f) are voxelized models evolved at 100, 200, 300, 400 and 500 iterations, respectively.

[0189] Fig. 9 The error information between the reconstructed model and the expected evolution result during the evolution process is compared, that is, (a) is the error information between the initial geometric model and the expected result, and (b)-(d) are the error information of different evolution processes; the geometric model evolution error is within 0-4mm. It can be seen that as the number of evolution steps increases, the reconstructed service wheel geometric model is closer to the expected evolution result, and the cloud map shows that the geometric model of the final evolution result of the wheel is consistent with the expected evolution result. It can be seen that the method of the present invention can realize the continuous evolution modeling of the vehicle profile.

[0190] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0191] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0192] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are within the protection of the present invention.

Claims

1. A modeling method for continuous four-dimensional evolution of wheel-rail profile throughout its life cycle, characterized in that: include: Construct a standard wheel-rail discrete voxel model based on wheel-rail design data; A non-contact two-dimensional laser displacement sensor is used to collect the measured discrete wheel-rail profile data on the wheel-rail surface during different service cycles. Based on the vehicle system dynamics model, wheel-rail contact mechanics model and wheel-rail material contact wear model, obtain the simulation data of wheel-rail profile evolution during the service life; Based on the measured wheel-rail discrete profile data, the velocity field function of the wheel-rail measured model evolution is constructed; Based on the simulation data of wheel-rail profile evolution during the service life, the evolution velocity field function of the wheel-rail simulation model is constructed; Based on the standard wheel-rail discrete voxelized model, the evolution velocity field function of the wheel-rail measured model and the evolution velocity field function of the wheel-rail simulation model, the level set editing method is used to perform continuous four-dimensional evolution of the wheel-rail profile on the service wheel-rail discrete voxelized model.

2. A wheel-rail profile full life cycle continuous four-dimensional evolution modeling method according to claim 1, characterized in that: The method further comprises: based on the standard wheel-rail discrete voxelized model, the wheel-rail measured model evolution velocity field function and the wheel-rail simulation model evolution velocity field function, a level set editing method is used to realize the continuous four-dimensional evolution of the service wheel-rail discrete voxelized model, and then: The marching cubes algorithm is used to realize the geometric modeling and visual expression of the discrete voxelized model of the in-service wheel-rail after continuous four-dimensional evolution.

3. A wheel-rail profile full life cycle continuous four-dimensional evolution modeling method according to claim 1, characterized in that: The method of constructing a standard wheel-rail discrete voxel model based on wheel-rail design data includes: Based on the wheel-rail design data, the wheel-rail physical design model is constructed by using the physical modeling design method; Extract wheel-rail geometric surface model from wheel-rail entity design model; The standard wheel-rail discrete voxel model is constructed based on the wheel-rail geometric surface model using the geometric surface discretization method.

4. A wheel-rail profile full life cycle continuous four-dimensional evolution modeling method according to claim 3, characterized in that: The step of extracting the wheel-rail geometric surface model from the wheel-rail entity design model comprises: Selecting the surface geometric features of the wheel-rail entity design model, and converting the surface features of the wheel-rail entity design model into a triangular mesh surface model through triangular mesh discretization; The method of using a geometric surface discretization method to construct a standard wheel-rail discrete voxel model based on a wheel-rail geometric surface model comprises: The voxelization method is used to discretize the triangular mesh surface model into a standard wheel-rail discrete voxel model, and the position relationship between each voxel vertex in the standard wheel-rail discrete voxel model and the triangular mesh surface model is calculated. The calculation formula is: Where, the voxel vertices in the standard wheel-rail discrete voxel model are p(x, y, z), the interior of the standard wheel-rail discrete voxel model is Ω, and the exterior of the standard wheel-rail discrete voxel model is Ω c The surface of the standard wheel-rail discrete voxel model is And satisfy disp(p)=min(|pp I |), that is, min(|pp I |) is the shortest distance between each voxel vertex in the standard wheel-rail discrete voxel model and the triangular mesh surface model; For the data of the standard wheel-rail discrete voxel model, the voxel vertices within the preset range of the zero level set surface model are assigned as active points in a narrowband manner, and the rest are inactive vertices. The sparse dynamic OpenVDB data structure is used to store the vertex coordinates and distance attribute data of the standard wheel-rail discrete voxel model to achieve voxel discretization expression of the standard wheel-rail profile.

5. The method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle according to claim 1 is characterized in that: The non-contact two-dimensional laser displacement sensor is used to collect the measured wheel-rail discrete profile data on the wheel-rail surface in different service cycles, including: A non-contact two-dimensional laser displacement sensor is used to continuously collect displacement data of the service wheel-rail surface to obtain discrete wheel-rail profile data of the wheel-rail surface in different service cycles, wherein the discrete wheel-rail profile data includes three-dimensional point cloud data of the wheel-rail profile; The nearest neighbor point pair matching method is used to adjust the measured wheel-rail profile pose to achieve registration of discrete wheel-rail profile data in different coordinate systems. The registered wheel-rail discrete profile data map is loaded into the standard wheel-rail discrete voxelized model to calculate the measured wheel-rail profile evolution velocity field.

6. A method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle according to claim 5, characterized in that: The method of using the nearest neighbor point pair matching method to adjust the measured wheel-rail profile pose to achieve registration of discrete wheel-rail profile data in different coordinate systems includes: Based on the rotation and translation relationship between the laser displacement sensor coordinate system and the world coordinate system, the wheel-rail discrete profile data is transformed to achieve the registration of the wheel-rail discrete profile data in different coordinate systems. The expression formula of the rotation and translation relationship between the laser displacement sensor coordinate system and the world coordinate system is: Wherein, M(mx,my,mz) is the wheel-rail design profile data, N(nx,ny,nz) is the measured wheel-rail profile 3D point cloud data, and the expression formula is composed of the rotation transformation matrix R(α,β,γ) and the translation vector t(t x ,t y ,t z ) is specifically expressed as:

7. The method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle according to claim 1, characterized in that: The method of obtaining the profile evolution simulation data of the wheel-rail service cycle based on the vehicle system dynamics model, the wheel-rail contact mechanics model and the wheel-rail material contact wear model also includes: Construct a dynamic model of the vehicle system with wheel-rail coupling; Calculate the wheel-rail contact geometry parameters during rail vehicle operation based on the vehicle system dynamics model; A wheel-rail contact mechanics model is constructed to calculate the local contact parameters during the wheel-rail service process; According to the physical properties of wheel-rail contact materials and wheel-rail contact force, a wheel-rail material contact wear model is constructed.

8. A method for modeling the continuous four-dimensional evolution of wheel-rail profile throughout its life cycle according to claim 7, characterized in that: The construction of the wheel-rail coupled vehicle system dynamics model comprises: Obtain the numbers of the two-axle bogies and the car body of the rail vehicle to build a vehicle system dynamics model; The wheel-rail contact mechanical model is constructed to calculate the local contact parameters of the wheel-rail during service, including: The wheel-rail normal contact force during the service life of the wheel-rail is calculated using the following formula: Where x, y are the longitudinal and transverse coordinates of the wheel-rail contact spot, a, b are the semi-axis values ​​of the elliptical contact spot; The Kalker simplified theory Fastsim algorithm is used to calculate the creep force of the wheel-rail during service. The calculation formula is: In the formula, Δp x (x,y) and Δp y (x, y) are the tangential stress increments along the longitudinal and transverse directions, respectively, expressed as: In the formula, ξ x is the longitudinal creep rate, ξ y is the lateral creep rate, ξ z is the spin creep rate, which is determined by the vehicle system dynamics model, and L is the tangential compliance coefficient, which is determined by the table lookup method; The wheel-rail material contact wear model is constructed according to the physical properties of the wheel-rail contact material and the wheel-rail contact force, including: The wheel-rail material contact wear model is constructed based on the Archard wear model, wherein the material loss of the wheel-rail material contact wear model is proportional to the creep distance and the normal force, and inversely proportional to the material hardness. The wheel-rail material contact wear model is expressed as: Where k(x,y) is the inner wheel-rail friction coefficient of the contact spot unit (x,y), p z (x,y) is the normal contact stress of the contact spot unit (x,y), and s(x,y) is the sliding distance of the contact spot unit (x,y).

9. A wheel-rail profile full life cycle continuous four-dimensional evolution modeling method according to claim 1, characterized in that: The method of constructing the wheel-rail measured model evolution velocity field function based on the measured wheel-rail discrete profile data includes: Obtain the wheel coordinate data and rail coordinate data mapped from the measured wheel-rail discrete profile data to the distance field of the standard wheel-rail discrete voxelized model, and use the nearest neighbor search algorithm to search for the nearest neighbor points of the voxel vertices to the in-service wheel point cloud and rail point cloud to construct the nearest neighbor point pairs; The velocity vector of voxel vertex evolution is constructed according to the relative position of the point pair in space, so as to construct the velocity field function of wheel-rail measured model evolution; The wheel-rail simulation model evolution velocity field function is constructed based on the wheel-rail profile evolution simulation data during the wheel-rail service period, including: Acquiring simulated data of wheel-rail profile evolution during a service life to predict a wheel-rail profile, and further mapping the wheel-rail profile into a discrete voxel distance field; The wear amount at each position of the wheel-rail profile is obtained, and the zero isosurface velocity field value is constructed based on the wear amount, and the evolution velocity is further expanded according to the voxel vertex gradient value to construct the wheel-rail simulation model evolution velocity field function.

10. A wheel-rail profile full life cycle continuous four-dimensional evolution modeling method according to claim 1, characterized in that: The method uses a level set editing method to continuously evolve the wheel-rail profile of the service wheel-rail discrete voxel model in four dimensions based on the standard wheel-rail discrete voxel model, the wheel-rail measured model evolution velocity field function and the wheel-rail simulation model evolution velocity field function, including: The distance field of the standard wheel-rail discrete voxelized model is superimposed with the evolution velocity field of the wheel-rail measured model and the evolution velocity field of the wheel-rail simulation model; The wheel / rail profile of the evolved service wheel / rail discrete voxel model is updated by using the level set method. After a preset number of evolution steps, the distance field value of the standard wheel / rail discrete voxel model is recalculated and a narrow band is established. The method of updating the wheel-rail profile of the discrete voxelized model of the wheel-rail in service by using the level set method comprises: The level set method is used to perform continuous four-dimensional evolution of the wheel-rail profile on the standard wheel-rail discrete voxel model. The evolution formula is: φ(p(x,y,z),t)=0, In the formula, the evolution formula can be obtained by using the chain differential rule: The distance field of the standard wheel-rail discrete voxel model is initialized by solving the steady-state solution of the partial differential equation, where the initialization operation formula is: By determining the voxel distance field value near the zero level set, the continuous four-dimensional evolution of the wheel and rail profile of the service wheel and rail discrete voxel model is realized; The method of recalculating the distance field value of the standard wheel-rail discrete voxelized model after a preset number of evolution steps and establishing a narrow band includes: Taking the zero level set surface as the center, determine the narrow band local area with a width of 2δ around the zero level set surface; The voxel vertices in the marked narrow band area are activated points, and the other voxel vertices are inactivated points. During the continuous four-dimensional evolution of the wheel-rail profile of the service wheel-rail discrete voxelization model, only the level set function values ​​of the activated points are iteratively updated.

Images