A method for generating composite equivalent taper of wheelsets, computer equipment, and readable storage medium

Abstract

This application discloses a method, computer equipment, and readable storage medium for generating composite equivalent conicity of wheelsets, belonging to the field of rail transit operation and maintenance. The method includes the following steps: acquiring relevant data on the equivalent conicity of the wheelsets of a target railway vehicle; determining calculation parameters for the equivalent conicity of the wheelsets based on the relevant data; and generating the composite equivalent conicity of the wheelsets by calling a target calculation model based on the calculation parameters. The target calculation model is constructed based on a target calculation formula, which is as follows: where represents the composite equivalent conicity of the wheelsets, represents the nominal equivalent conicity of the wheelsets, and represents the nonlinear factor. The nonlinear factor characterizes the degree of distortion of the target curve, and the nominal equivalent conicity of the wheelsets is the equivalent conicity of the wheelsets at a lateral displacement of 3 mm on the target curve. The target curve characterizes the relationship between the equivalent conicity of the wheelsets and the lateral displacement of the wheelsets. This method can reduce safety hazards in rail transit and improve the service life of wheels and rails.

Description

Technical Field

[0001] This application belongs to the field of rail transit operation and maintenance, and specifically relates to a method for generating composite equivalent taper of wheelsets, computer equipment, and readable storage medium. Background Technology

[0002] Rail transit is one of the important modes of transportation to ensure the normal operation of urban functions. It improves people's travel efficiency, shortens the distance between cities, and effectively alleviates traffic congestion. The wheel-rail system is one of the key components of rail transport vehicles (railway vehicles). Rail vehicles must achieve traction, operation, and braking through the rolling friction contact between the wheels and rails.

[0003] As is well known, friction is always accompanied by wear, and wear directly changes the macroscopic dimensions and performance of railway vehicle wheels and rails. This causes the wheel-rail geometric contact relationship to change as the service time of railway vehicles increases, and further affects the rolling contact dynamics of railway vehicle wheels and rails.

[0004] The wheelset equivalent taper is an important parameter in the wheel-rail geometric contact relationship. It determines the degree of fit between the wheel and rail, and wear increases the wheelset equivalent taper. A reasonable wheelset equivalent taper can reduce wheel wear on the rail, improve the stability of railway vehicles when passing through curves, reduce vibration and noise, and extend the service life of the wheels and rails.

[0005] Currently, based on industry experience, when the equivalent taper of the wheelset exceeds a certain value (e.g., 0.4), the stability of railway vehicles traversing curves is considered poor, and serpentine instability is likely to occur. Therefore, to avoid serpentine instability, wheel-rail maintenance is generally performed when the equivalent taper of the wheelset reaches or approaches this certain value. However, in practical applications, serpentine instability often occurs when railway vehicles traverse curves even when the equivalent taper of the wheelset is still less than this certain value. This makes timely wheel-rail maintenance impossible, leading to unpredictable safety hazards in rail transit and a reduction in wheel-rail lifespan. Summary of the Invention

[0006] The purpose of this application is to provide a method, computer device, and readable storage medium for generating wheelset composite equivalent taper, which can solve the problem of how to improve the accuracy of wheelset equivalent taper, maintain wheel and rail in a timely manner, thereby reducing safety hazards in rail transit and increasing wheel and rail service life.

[0007] To solve the above-mentioned technical problems, this application is implemented as follows:

[0008] In a first aspect, embodiments of this application provide a method for generating a wheelset composite equivalent taper, the method comprising:

[0009] Obtain relevant data on the equivalent conicity of the wheelsets of the target railway vehicle;

[0010] The calculation parameters for the equivalent taper of the wheelset are determined based on the relevant data of the equivalent taper of the wheelset;

[0011] Based on the calculation parameters, the target calculation model is invoked to generate the wheelset composite equivalent taper;

[0012] The target calculation model is constructed based on the target calculation formula; the target calculation formula is as follows:

[0013]

[0014] in, Indicates the combined equivalent taper of the wheelset. Indicates the nominal equivalent taper of the wheelset. Indicates the nonlinear factor;

[0015] The nonlinear factor characterizes the degree of distortion of the target curve, and the nominal equivalent taper of the wheelset is the equivalent taper of the wheelset at the point on the target curve where the lateral displacement of the wheelset is 3 mm; the target curve characterizes the relationship between the equivalent taper of the wheelset and the lateral displacement of the wheelset.

[0016] In a second aspect, embodiments of this application provide a computer device including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0017] Thirdly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0018] Fourthly, embodiments of this application provide a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the method described in the first aspect.

[0019] In this embodiment, when calculating the wheelset equivalent taper, a nonlinear factor is added to the nominal equivalent taper to obtain the wheelset composite equivalent taper. This ensures that the wheelset equivalent taper takes into account the degree of curve distortion, meaning it considers the potential significant numerical differences in the wheelset equivalent taper at different lateral displacements. According to the target calculation formula, there is a positive correlation between the degree of distortion (nonlinear factor) and the wheelset composite equivalent taper; that is, the greater the degree of distortion, the greater the wheelset composite equivalent taper. Therefore, the wheelset composite equivalent taper can reach a certain value mentioned in the background art before the nominal equivalent taper, thus enabling timely wheel-rail maintenance, reducing safety hazards in rail transit, and increasing wheel-rail service life. Attached Figure Description

[0020] Figure 1 These are equivalent taper curves for two degrees of distortion involved in some embodiments of this application;

[0021] Figure 2 This is a flowchart illustrating a method for generating composite equivalent taper of wheelsets according to some embodiments of this application;

[0022] Figure 3a These are some embodiments of this application providing simulation diagrams showing the change in wheel dent wear with operating mileage under various operating conditions;

[0023] Figure 3b These are the wheelset equivalent taper curves under different wheel wear characteristics corresponding to different operating mileages provided in some embodiments of this application;

[0024] Figure 4 This is a schematic diagram of fitting a wheel tread using a 3rd order NURBS curve, provided in some embodiments of this application;

[0025] Figure 5 This is a comparison diagram of the nominal equivalent taper, nonlinear factor, and composite equivalent taper of wheelsets under different operating mileages provided by some embodiments of this application;

[0026] Figure 6a This is a statistical diagram of the concave wear characteristics of seven vehicles with two wheels, based on some embodiments of this application;

[0027] Figure 6b This is a statistical diagram of the concave wear characteristics of eight-car, one-position wheels involved in some embodiments of this application;

[0028] Figure 7 This is a schematic diagram of the process for obtaining a set of optimal control points according to some embodiments of this application;

[0029] Figure 8 This is a flowchart illustrating the process of using a genetic algorithm to solve for the optimal weight factor, provided in some embodiments of this application. Detailed Implementation

[0030] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0031] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0032] The wheelset composite equivalent taper generation method, computer equipment, and readable storage medium provided in this application will be described in detail below with reference to the accompanying drawings and through specific embodiments and application scenarios.

[0033] As mentioned in the background technology: In practical applications, when the equivalent taper of the wheelset is still less than a certain value, the railway vehicle has already experienced serpentine instability when passing through the curve. This makes it impossible to maintain the wheel and rail in a timely manner, resulting in certain unpredictable safety hazards in rail transit and a reduction in the service life of the wheel and rail.

[0034] The reason for this is the low accuracy of the calculation of the wheelset equivalent taper, leading to inaccurate results. The applicant's research found that the widely adopted UIC519 standard uses the wheelset equivalent taper at a 3mm lateral displacement as the nominal equivalent taper to measure the wheel-rail matching performance. However, during railway vehicle operation, wheels cannot always maintain only minor swaying of less than 3mm; they will experience greater lateral movement due to track irregularities, curves, and other factors. The equivalent taper at different wheelset lateral displacements may differ, and potentially significantly.

[0035] like Figure 1 As shown, Figure 1 The equivalent taper of the wheelset corresponding to 2mm is significantly different from that corresponding to 3mm. Therefore, even if its value at 3mm (e.g.) =0.4) seems "okay", meaning it does not exceed the relevant threshold. However, railway vehicles will also experience severe serpentine instability when moving from 2mm to this larger wheelset lateral displacement.

[0036] In other words, when calculating the equivalent taper of a wheelset, it is not accurate enough to only consider the equivalent taper of the wheelset at a lateral displacement of 3mm. This is because the cause of slithering instability is not only that the equivalent taper of the wheelset is too large, but also that the numerical difference of the equivalent taper of the wheelset under different lateral displacements is too large.

[0037] Based on this, in an exemplary embodiment, such as Figure 2 As shown, this application proposes a method for generating a composite equivalent taper of a wheelset, which includes steps 202-206. Wherein:

[0038] Step 202: Obtain relevant data on the equivalent cone of the wheelset of the target railway vehicle.

[0039] Step 204: Determine the calculation parameters of the equivalent taper of the wheelset based on the relevant data of the equivalent taper of the wheelset.

[0040] Step 206: Based on the calculation parameters, call the target calculation model to generate the wheelset composite equivalent taper.

[0041] The target calculation model is constructed based on the target calculation formula; the target calculation formula is as follows: Formula (1):

[0042] (1)

[0043] in, Indicates the combined equivalent taper of the wheelset. Indicates the nominal equivalent taper of the wheelset. Indicates the nonlinear factor;

[0044] The nonlinear factor characterizes the degree of distortion of the target curve, and the nominal equivalent taper of the wheelset is the equivalent taper of the wheelset at the point on the target curve where the lateral displacement of the wheelset is 3 mm; the target curve characterizes the relationship between the equivalent taper of the wheelset and the lateral displacement of the wheelset.

[0045] Railway vehicles include passenger cars, freight cars, and urban rail transit vehicles. Passenger cars include conventional trains and high-speed trains. The following describes the implementation of each method step using high-speed trains as an example.

[0046] Among them, the relevant data on the equivalent taper of the wheelset can be determined by referring to the "Method for Determining the Equivalent Taper of Wheel-Rail Wheelset in Railway Applications" (TB / T3332-2013). This data includes at least the data used to parametrically describe the wheel-rail profile, such as the data on the wheel tread.

[0047] The calculation parameters include, in addition to the parameters related to the calculation of the equivalent taper of the wheelset specified in TB / T3332-2013, parameters used to calculate the nonlinear factor.

[0048] The degree of distortion (curve shape) refers to the degree of nonlinearity. In this embodiment, the "curve shape" is quantified by a nonlinear factor. That is, the ineffable "degree of distortion" is transformed into a calculable value through a relatively simple but effective method (the standard deviation of N points on the curve). The standard deviation is used to measure the fluctuation range of the curve, meaning that the value of the nonlinear factor is positively correlated with the fluctuation range.

[0049] Specifically, multiple equivalent wheel pair tapers can be obtained by sampling on the target curve, and the standard deviation of all sampled equivalent wheel pair tapers can be calculated, thus using the standard deviation as a nonlinear factor. For example, if N is 8, the equivalent wheel pair tapers at eight points with wheel pair lateral displacements of 1mm, 2mm, 3mm, 4mm, 5mm, 6mm, 7mm, and 8mm can be sampled, and the standard deviation of these eight equivalent wheel pair tapers can be calculated. The formula for calculating the nonlinear factor is as follows: Formula (2):

[0050] (2)

[0051] in, This indicates that the lateral displacement of the wheelset on the target curve is... Wheelset equivalent taper at mm; This is the average value of the equivalent taper of all wheelsets sampled from the target curve.

[0052] The target computational model can be implemented using MATLAB.

[0053] This embodiment calculates the wheelset equivalent taper by adding a nonlinear factor to the nominal equivalent taper, resulting in a composite equivalent taper. This ensures the equivalent taper takes into account the curve's distortion, recognizing the potential for significant numerical differences in equivalent taper values ​​at different lateral displacements. According to the target calculation formula, there is a positive correlation between the distortion degree (nonlinear factor) and the composite equivalent taper; the greater the distortion, the larger the composite equivalent taper. Therefore, the composite equivalent taper can reach the specified value mentioned in the background art before the nominal equivalent taper, enabling timely wheel-rail maintenance, reducing safety hazards in rail transit, and increasing wheel-rail service life.

[0054] In one embodiment, such as Figure 3a As shown, Figure 3a This is a graph showing the change in wheel dent wear as a function of operating mileage under simulated operating conditions. The graph reflects the gradual increase in wheel wear with increasing operating mileage. Figure 3b The above, Figure 3b These are the wheelset equivalent taper curves under different wheel wear characteristics corresponding to different operating mileages. Under the simulation conditions, the maximum wear of the wheel is close to 1.9 mm, and the maximum nominal equivalent taper of the wheelset at the lateral displacement of 3 mm also reaches more than 0.6. According to experience, if the wheelset equivalent taper exceeds 0.4, the high-speed train will experience lateral instability. However, here, the wheelset equivalent taper is around 0.35, and the high-speed train has already experienced serpentine instability. It can be seen that the wheelset equivalent taper is closely related to the wear of the train.

[0055] Therefore, in this embodiment, the nonlinear factor involves not only the degree of distortion of the target curve, but also the wear information carried by the target curve, that is, the combined effect of the wear and distortion degree reflected by the target curve on the composite equivalent taper of the wheelset is considered.

[0056] The nonlinear factor includes three factorization formulas: sign factorization, wear factorization, and standard deviation factorization.

[0057] The standard deviation factorization formula is used to calculate the index value of the degree of torsion, specifically referring to the formula for calculating the standard deviation of the equivalent taper of all wheelsets obtained from the above sampling. That is, the index value is the aforementioned standard deviation.

[0058] Among them, the symbolic factorization is used to determine the curve trend of the curve segment of the target curve in the area near the lateral displacement of the wheelset of 3 mm.

[0059] Among them, the wear factorization formula is used to calculate the characteristic value of the wheel-rail wear characteristics reflected by the target curve.

[0060] For example, the expression for the nonlinear factor is as follows: Formula (3):

[0061] (3)

[0062] in, For symbolic factorization, This is the decomposition formula for wear factors. This is the standard deviation factorization expression; This represents the equivalent taper of the wheelset at a point on the target curve where the wheelset lateral displacement is 2 mm. This represents the equivalent taper of the wheelset at a point on the target curve where the wheelset lateral displacement is 4 mm. This indicates that the lateral displacement of the wheelset on the target curve is... Wheelset equivalent taper at mm; This is the average value of the equivalent taper of all wheelsets sampled from the target curve; It represents the degree of wear of the current wheel relative to a new wheel.

[0063] for Its value is ±1, thereby achieving the purpose of adjusting the nonlinear factor, where "+1" indicates that the curve trend is downward and "-1" indicates that the curve trend is upward.

[0064] It should be noted that a downward trend is generally less conducive to the stability of high-speed train operation. Therefore, a downward trend corresponds to "+1", which, in the target calculation formula, means increasing the value based on the nominal equivalent cone of the wheelset.

[0065] It should be noted that the target curve is obtained through computer simulation. During the simulation, wheel and rail wear data from the railway department can be collected to statistically determine the wheel-rail wear characteristics. These wheel-rail wear characteristics are then used to fit a curve that approximates the realistic wheel tread profile during the simulation.

[0066] Specifically, in simulation, curves such as Bézier curves, B-spline curves, and NURBS curves derived from both can be used.

[0067] The following example illustrates an implementation of the wear factorization formula, using NURBS curves in simulation:

[0068] The formula for NURBS curves will be introduced below:

[0069] NURBS curves are developed based on the Bézier method and the B-spline method. They retain the powerful functions of the Bézier method and the B-spline method, and can accurately represent quadratic curves and surfaces. The NURBS method is widely used in the fields of CAD / CAM and computer graphics due to its powerful flexibility and good geometric properties. The rational fraction of a k-th degree NURBS curve is expressed as follows (4):

[0070] (4)

[0071] in, ; Using control points as control points, connect the control vertices sequentially with polylines to form a control polygon; For the weight factors corresponding to the control points, It is a k-th degree B-spline basis function. It is a parameter, and its value range is within the node vector. Within the defined interval, there are m+1 items, where m satisfies m=n+k+1.

[0072] Wherein, basis functions It can be obtained from the Debussy-Cockell recursive formula, as shown in formula (5):

[0073] (5)

[0074] in, These are called nodes.

[0075] In summary, we can draw the following two conclusions:

[0076] (1) It is a step function, and its value is in All values ​​outside the range are 0.

[0077] (2) It is a linear combination of two k-1 degree basis functions.

[0078] It should be noted that for NURBS curves used in general applications, the nodes at both ends of the node vector are usually taken as k+1 times multiple nodes, that is, the nodes at both ends are... The purpose of this selection is to ensure that the two endpoints of the NURBS curve are tangent to the start and end points of the control polygon while passing through them. In practical applications, the endpoint values ​​of the node vector are usually 0 and 1, and the node vector is calculated using the following formula (6):

[0079] (6)

[0080] Therefore, the domain of the curve .

[0081] Specifically, when simulating using NURBS curves, the expression for the wear factorization formula is as follows (7):

[0082] (7)

[0083] in, For adjustment coefficients, The mean of the weighting factors for the NURBS curve used to describe the tread profile of the new wheel. This is the mean of the weighting factors used to describe the current wheel tread profile in the NURBS curve.

[0084] As can be seen from the formula for the NURBS curve above, obtaining the NURBS curve through simulation requires determining the control points, node vectors, and weight factors. Among these, the node vectors can be determined through the control points.

[0085] This embodiment only discusses the influence of control points and weighting factors on the NURBS curve fitting process.

[0086] Control points are used to determine the approximate shape of the NURBS curve, while weight factors are used to fine-tune the coordinates of the control points, thereby determining the final shape of the NURBS curve.

[0087] It is understandable that, given a fixed coordinate system of control points, different weighting factors can produce different NURBS curves.

[0088] Based on this, such as Figure 4 As shown, the following explanation uses a cubic NURBS curve as an example. The shape of this cubic NURBS curve is determined by five control points and three weighting factors. The five control points include P... 6-0 ,P 6-1 ,P 6-2 ,P 6-3 and P 6-4P 6-0 and P 6-4 The beginning and end points of the NURBS curve; the three weighting factors include , and , respectively with P 6-1 P 6-2 and P 6-3 Correspondingly; that is, during the fitting process, P 6-0 and P 6-4 It is fixed and unmoving, and based on , and P 6-1 P 6-2 and P 6-3 The coordinates are adjustable.

[0089] Here, "thin rim tread" refers to the curve of the wheel tread profile. Generally, since the relative displacement of the wheelset occurs within a certain range, the wheel tread (wheel-rail contact part) is part of, but not all of, the overall wheel profile; that is, wheel wear is concentrated in a certain part of the overall wheel profile. Considering the characteristics of this wheel wear and in order to improve the fitting accuracy of the NURBS curve, this embodiment uses the NURBS curve to simulate the part of the wheel wear.

[0090] In summary, the target curves for wheel tread profiles under different operating mileages can be obtained through NURBS curve fitting. Based on these target curves and the measured wear amounts, the following table (Table 1) can be obtained:

[0091] Table 1

[0092]

[0093] Using the data in Table 1 above as an example, the characteristic values ​​of wheel-rail wear characteristics under different operating mileages can be calculated. The following table 2 is obtained:

[0094] Table 2

[0095]

[0096] in, The mean of the weighting factors.

[0097] Analysis of Tables 1 and 2 reveals that the amount of wear is related to... There is a positive correlation. However, as wear and tear increases, the risk of high-speed trains experiencing serpentine instability also increases. Therefore, through... It can improve the accuracy of the wheelset equivalent taper. That is, by setting the wear factor decomposition formula in the target calculation formula corresponding to the wheelset composite equivalent taper, the accuracy of the wheelset composite equivalent taper can be improved, thereby accurately detecting the possible serpentine instability of high-speed trains and avoiding it in advance.

[0098] For example, such as Figure 5 As shown, Figure 5 In this study, when a high-speed train reaches 200,000 kilometers of operating mileage, the corresponding nominal equivalent cone of the wheelset is 0.33, and the nonlinear factor is 0.08. Based on empirical values, a high-speed train may experience serpentine instability when the equivalent cone of the wheelset reaches 0.4. At this point, the nominal equivalent cone of the wheelset is 0.33, not yet reaching 0.4. However, after adding the nonlinear factor of 0.08, the final composite equivalent cone of the wheelset is 0.41, reaching the instability limit of 0.4. Actual measurement data shows that this high-speed train experienced serpentine instability after reaching 200,000 kilometers of operating mileage. It can be understood that the setting of the nonlinear factor is equivalent to correcting the deviation in the equivalent cone of the wheelset.

[0099] In one embodiment, determining the weight factor of the cubic NURBS curve of the current wheel tread profile includes: within a refining cycle, solving for a set of candidate control points for the cubic NURBS curves corresponding to different preset mileages reached by the high-speed train; the cubic NURBS curves are subject to constraints; determining an optimal control point from multiple sets of candidate control points within the refining cycle; and determining the weight factor of the cubic NURBS curve of the current wheel tread profile based on the optimal control point.

[0100] The wheel turning cycle refers to the operating mileage interval for high-speed train wheelset turning. For example, if a wheel is turned every 200,000 kilometers, then candidate control points can be solved every N0,000 kilometers. For example, if N is 2, then the preset mileage is 20,000 kilometers, 40,000 kilometers, ..., 200,000 kilometers.

[0101] This involves identifying a set of candidate control points for the cubic NURBS curves corresponding to different preset mileages reached during the high-speed train's operation within the overhaul cycle. The least squares method is a commonly used approach for solving optimization problems. The measured wheel treads represent a series of data points. The core idea of ​​the least squares method is to minimize the sum of squares of the errors between the cubic NURBS curve and the data points. For each data point, the difference between it and the corresponding point on the cubic NURBS curve is calculated, and all differences are squared and summed.

[0102] By adjusting the control points of the cubic NURBS curve, the sum of squares is minimized, thereby making the cubic NURBS curve as close as possible to the data points, and thus obtaining the optimal fitting curve in the sense of minimizing the sum of squared errors.

[0103] Specifically, assume that the data point consists of p+1 ordered sequence points. The structure is first determined by using the cumulative chord length parameter method to determine the parameter values ​​of the data points. The calculation formula for the cumulative chord length parameter method is formula (8):

[0104] (8)

[0105] in, Forward difference vector, That is, the chord vector.

[0106] The goal of fitting is to make the cubic NURBS curve as close as possible to these data points. The objective function is constructed using the least squares method, as shown in the following formula (9):

[0107]

[0108] in, For the first One candidate control point; For the first The measured data points refer to a set of data points measured from the wheel tread when the high-speed train reaches any preset mileage.

[0109] Regarding constraints, it should be noted that while searching for candidate control points, it is also necessary to set constraints on the curve to ensure the smoothness of the curve fitting.

[0110] Specifically, the primary fitting area is defined as the 20mm interval to the left and right of the nominal rolling circle region, which is the main area where wheel tread wear occurs. The reasoning is as follows:

[0111] When high-speed trains are brought into the depot for nighttime maintenance, tracked tests were conducted on the wear changes of the wheels of 8-car trains at operating mileages of 20,000 km, 57,000 km, 80,000 km, 100,000 km, 116,000 km, 137,000 km, 163,000 km, 185,000 km, and 200,000 km. Taking the second wheel of a seven-car train and the first wheel of an eight-car train as examples, ... Figure 6a and Figure 6b As shown, wheel wear mainly occurs in three places: near the nominal rolling circle of the wheel tread, at the junction of the tread and the rim, and at the rim. Among these, the concave wear of the tread is most severe at 20mm on both sides of the wheel rolling circle radius, within the range of [-20, 20].

[0112] Therefore, the 20mm interval to the left and right of the nominal rolling circle region is used as the main fitting area. To ensure the smooth continuity of the fitted tread surface at the connection point, the second control point P of the cubic NURBS curve is given. 6-1 and the fourth control vertex P6-3 Add a constraint: At the curve connection point, the left and right adjacent control points are collinear.

[0113] Reference Figure 4 In the rim wear section DE, control point P 6-1 It needs to pass through the previous arc to point P 6-0 Tangent at point P, control point P 6-3 It needs to go through the control endpoint P 6-4 The tangent.

[0114] In one embodiment, after obtaining candidate control points corresponding to different preset mileages within the overhaul cycle, a set of optimal control points is selected from the candidate control points. These optimal control points are used for fitting the 3rd NURBS curves corresponding to all preset mileages.

[0115] Specifically, if the weighting factor is to be used as a characteristic parameter of the wheel tread under different wear conditions, the control points under different wear conditions must be the same. In other words, within a refinishing cycle, the wheel tread under various wear stages needs to be fitted with NURBS curves three times at the same set of control points, and the only variable in the fitting process is the weighting factor.

[0116] It is understandable that the weighting factors of the curves at each wear stage (corresponding to the preset mileage) can be used as the characteristic parameters of the wear profile. Therefore, the set of optimal control points to be found not only needs to be able to adapt to multiple wheels (a high-speed train has 64 wheels), but also needs to consider the fitting situation of a turning cycle, that is, to ensure the fitting accuracy of the wheel tread profile under the set of optimal control points within a turning cycle.

[0117] Specifically, firstly, starting with new wheels, the wheel tread profile data of a train carriage with 64 wheels is measured every 20,000 kilometers until 200,000 kilometers before overhaul, for a total of 11 samplings. The average value of each measurement for the 64 wheelsets is taken as the representative profile, i.e., 11 sets of representative wheel tread profiles. Secondly, the optimal control points that minimize the fitting error of the 11 sets of representative profiles are determined.

[0118] The optimal control point is obtained by solving the following formula (10):

[0119] (10)

[0120] in, For the first One candidate control point; For the first One optimal control point; This represents the number of nodes.

[0121] like Figure 7 As shown, Figure 7 This is a flowchart illustrating the process of finding a set of optimal control points. The overall algorithm at this point focuses on three internal control points (…). Figure 4 P in 6-1 P 6-2 and P 6-3 Within the feasible region, adjust each candidate control point group. ( The weighting factor for each curve corresponding to different control point groups (numbered as follows) is used to calculate the root mean square error of the 11 curves within one refining cycle. The minimum error is then output, along with the error of the curve with the largest mean square error among the 11 tread curves fitted using the candidate control point group for one turning cycle. Finally, output all. The candidate control point group corresponding to the minimum value in This is the optimal control vertex we are looking for.

[0122] Furthermore, such as Figure 8 As shown, Figure 8 This is a flowchart illustrating the process of using a genetic algorithm to solve for the optimal weight factors. The genetic algorithm calculates the optimal combination of weight factors by parameterizing the cumulative chord length of measured wheel tread profile data at different preset mileages, setting the NURBS curve iterations and the number of control points, and using the 200,000 km wear limit of the wheel tread as a benchmark. The coordinates of the optimal control points are sequentially input into the objective function, with only the weight factors as variables. The objective value is the minimum area Smin between the measured value and the NURBS curve fitting value. Through repeated iterations using the genetic algorithm, the optimal combination of weight factors is finally output.

[0123] In an alternative embodiment, in conventional technology, wheel and rail are often repaired separately with reference to the equivalent taper of the wheelset within a single repair cycle. However, this “planned repair” approach is often delayed, meaning that the high-speed train may have already experienced serpentine instability before the repair is completed. This can also lead to certain unpredictable safety hazards in rail transit and a reduction in the service life of the wheels and rails.

[0124] Based on this, this embodiment integrates machine learning ideas to quantify the nonlinear relationship between the wheelset composite equivalent taper and the dynamic characteristics of the bogie, and constructs a closed-loop model from data preprocessing to engineering application. It can invert the wheelset composite equivalent taper through dynamic characteristics and issue an early warning when it exceeds the empirical value (threshold), promote the transformation of wheel-rail maintenance from "planned maintenance" to "condition-based maintenance", reduce maintenance costs, improve the accuracy of matching performance monitoring, reduce safety hazards, and increase wheel-rail service life.

[0125] Specifically, the influence of wheel wear on the composite equivalent taper of the wheelset and the vibration signal at the axle box is analyzed. The characteristics of the vibration signal at the axle box are analyzed, and a dataset of vibration signal at the axle box (feature data of dynamic characteristics) and wheel wear and corresponding equivalent taper of the wheelset is constructed. A nonlinear support vector machine (SVM) model (i.e. prediction model) is established to verify the mapping relationship between the lateral vibration signal at the axle box and wheel wear and the composite equivalent taper of the wheelset, thereby realizing the monitoring of wheel-rail matching status.

[0126] For example, this embodiment uses spectrum recognition to monitor wheel-rail matching status. Specifically, the input data originates from the lateral and axial vibrations of the train axle boxes. These time-series data contain the dynamic characteristics of the axle boxes during operation, i.e., their kinetic features. First, a spectrum transformation is performed on these time-series data to convert them into frequency domain information. This step is to reveal the inherent characteristics and structure of the data in the frequency dimension.

[0127] This embodiment uses the short-time Fourier transform (STFT) to calculate the spectral characteristics of a signal over time. Specifically, the STFT divides the signal into overlapping blocks using a sliding window and calculates the Fourier transform for each block. For a continuous time series signal... The formula for calculating STFT is as follows: (11)

[0128] (11)

[0129] in, It is a complex-valued window function, and its complex conjugate is... This complex-valued window function can be interpreted as determining... With window scalar product, the window Through time Translate, then frequency shift.

[0130] After obtaining the frequency domain information, a Visual Transformer (ViT) architecture is further employed for in-depth learning and analysis of the data. Specifically, the spectral image is first segmented into multiple small patches, which spatially deconstruct the information of the lateral and axial vibration signals. Then, these vibration signal patches are mixed and fed into a fully connected layer. In this layer, the network extracts features from the image data of each patch, aiming to capture temporal features in different frequency bands.

[0131] The extracted time-series features and predicted class results are then fed into the encoder of the Transformer model. The encoder encodes the features through a series of self-attention mechanisms and fully connected layers, improving the richness and discriminative power of the feature representation.

[0132] Finally, the encoded features are output through another fully connected layer to estimate the equivalent taper and wear of the wheelset corresponding to the timing signal.

[0133] It should be noted that, due to the relatively small sample size of the track vibration dataset used in this embodiment, the Transformer structure, which relies on a large amount of data and lacks inductive bias, limits the network's learning performance. Therefore, to avoid collecting large-scale data, the model parameters of the pre-trained model are used for pre-training the spectrum converter. The pre-trained model can be trained using the ImageNet dataset.

[0134] Optionally, embodiments of this application also provide a computer device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the various processes of the above-described wheelset composite equivalent taper generation method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0135] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described wheelset composite equivalent taper generation method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0136] The processor is the processor in the computer device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0137] This application also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described wheelset composite equivalent taper generation method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0138] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0139] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0140] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0141] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A method for generating a composite equivalent taper of a wheelset, characterized in that, The method for generating the composite equivalent taper of the wheelset includes: Obtain relevant data on the equivalent conicity of the wheelsets of the target railway vehicle; The calculation parameters for the equivalent taper of the wheelset are determined based on the relevant data of the equivalent taper of the wheelset; Based on the calculation parameters, the target calculation model is invoked to generate the wheelset composite equivalent taper; The target calculation model is constructed based on the target calculation formula. The target calculation formula is as follows: in, Indicates the combined equivalent taper of the wheelset. Indicates the nominal equivalent taper of the wheelset. Indicates the nonlinear factor; The nonlinear factor characterizes the degree of distortion of the target curve, and the nominal equivalent taper of the wheelset is the equivalent taper of the wheelset at the point on the target curve where the lateral displacement of the wheelset is 3 mm. The nonlinear factor includes three factorizations: a sign factorization, a wear factorization, and a standard deviation factorization. The standard deviation factorization formula is used to calculate the index value of the degree of distortion. The symbolic factorization is used to determine the curve trend of the curve segment of the target curve in the region near the lateral displacement of the wheelset of 3 mm. The wear factorization formula is used to calculate the characteristic values ​​of the wheel-rail wear characteristics reflected by the target curve. The expression for the nonlinear factor is as follows: in, This represents the equivalent taper of the wheelset at a point on the target curve where the wheelset lateral displacement is 2 mm. This represents the equivalent taper of the wheelset at a point on the target curve where the wheelset lateral displacement is 4 mm. This indicates that the lateral displacement of the wheelset on the target curve is... Wheelset equivalent taper at mm; This is the average value of the equivalent taper of all wheelsets sampled from the target curve; The wear factorization formula represents the degree of wear of the current wheel relative to the new wheel; the expression of the wear factorization formula is as follows: in, For adjustment coefficients, The mean of the weighting factors for the NURBS curve used to describe the tread profile of the new wheel. The mean of the weighting factors used to describe the current wheel tread profile of the NURBS curve; The target curve represents the relationship between the equivalent taper of the wheelset and the lateral displacement of the wheelset.

2. The method for generating a composite equivalent taper of a wheelset according to claim 1, characterized in that, The NURBS curve is a cubic NURBS curve. The weighting factors for determining the cubic NURBS curve of the current wheel tread profile include: During the overhaul cycle, a set of candidate control points for the 3rd NURBS curves corresponding to different preset mileages of high-speed train operation are solved respectively; The cubic NURBS curve is subject to constraints. From multiple candidate control points within the refinishing cycle, determine one set of optimal control points; The weighting factors of the cubic NURBS curve of the current wheel tread profile are determined based on the optimal control point.

3. The method for generating a composite equivalent taper of a wheelset according to claim 2, characterized in that, The candidate control points are obtained by solving the following formula: in, For the first One candidate control point; For the first The measured data points refer to a set of data points measured from the wheel tread when the high-speed train reaches any preset mileage.

4. The method for generating a composite equivalent taper of a wheelset according to claim 3, characterized in that, The optimal control point is obtained by solving the following formula: in, For the first One candidate control point; For the first One optimal control point; This represents the number of nodes.

5. The method for generating a composite equivalent taper of a wheelset according to claim 1, characterized in that, The method further includes: Characteristic data for determining the dynamic characteristics of the bogie of the target railway vehicle; Based on the feature data of the aforementioned dynamic characteristics, the prediction model is invoked to generate a predicted value for the composite equivalent taper of the wheelset.

6. A computer device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the wheelset composite equivalent taper generation method as described in any one of claims 1-5.

7. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the wheelset composite equivalent taper generation method as described in any one of claims 1-5.

Images