WO2022183392A1

WO2022183392A1 - Wheel tread roundness monitoring method, device, and system

Info

Publication number: WO2022183392A1
Application number: PCT/CN2021/078819
Authority: WO
Inventors: 郭磊; 胡华亮; 邱志
Original assignee: 舍弗勒技术股份两合公司; 郭磊
Priority date: 2021-03-03
Filing date: 2021-03-03
Publication date: 2022-09-09

Abstract

A wheel tread roundness monitoring method. Wheels are used for wheel sets of a train, and each wheel corresponds to an axle box. The monitoring method comprises: establishing a transmission relationship (H) between axle box acceleration and a wheel-rail excitation force (S10); establishing an alarm threshold of an evaluation amount, the evaluation amount being a wheel-rail excitation force or a wheel radial dimension deviation (S20); during running of the train, obtaining real-time axle box acceleration at a measurement moment, and calculating according to the transmission relationship (H) to obtain a real-time evaluation amount at the measurement moment (S30); and comparing the real-time evaluation amount with the alarm threshold, and outputting a determination result (S40). Further disclosed is a wheel tread roundness monitoring device and system.

Description

Monitoring method, monitoring device and monitoring system for wheel tread roundness

technical field

The present invention relates to the field of vehicle monitoring, and in particular, to a monitoring method, a monitoring device and a monitoring system for the out-of-roundness of wheel treads of a train.

Background technique

For trains (or rail vehicles), the roundness of the wheel tread (also referred to as wheel roundness) is very important to the running of the train.

The monitoring of wheel roundness can be divided into offline and online.

The off-line method uses the wheel diameter jumper to directly measure the wheel diameter jump data, and judge the roundness of the wheel accordingly. This measurement method usually needs to be used when the train is in storage for maintenance, and cannot reflect the abnormality of wheel roundness in real time.

Online monitoring methods mostly use data processing methods based on vibration signals or noise signals to obtain a specific eigenvalue to describe the roundness of the wheel. However, these methods do not consider the influence of the mechanical structure on the signal transmission path, so the monitoring results are not ideal.

SUMMARY OF THE INVENTION

The purpose of the present invention is to overcome or at least alleviate the above-mentioned deficiencies of the prior art, and to provide a monitoring method, monitoring device and monitoring system for the roundness of wheel tread.

According to a first aspect of the present invention, a method for monitoring the roundness of a wheel tread is provided, the wheel is used for a wheelset of a train, and each wheel corresponds to an axle box, and the monitoring method includes:

Establish the transmission relationship between the acceleration of the axle box and the excitation force of the wheel and rail;

Establishing an alarm threshold of an evaluation quantity, the evaluation quantity being the wheel-rail excitation force or the radial dimension deviation of the wheel;

During the running process of the train, obtain the real-time axle box acceleration at the measurement moment, and calculate the real-time evaluation amount at the measurement moment according to the transfer relationship;

The real-time evaluation value is compared with the alarm threshold, and a judgment result is output.

In at least one embodiment, establishing the transmission relationship between the axle box acceleration and the wheel-rail excitation force includes:

When the train is in a non-running state, the wheel is loaded with the wheel-rail excitation force, and the axle box acceleration is measured at the same time;

Calculate the algebraic relationship between the wheel-rail excitation force and the axle box acceleration to obtain the transfer relationship.

In at least one embodiment, loading the wheel with the wheel-rail excitation force includes:

The wheel is hit with an external force equal to the wheel-rail excitation force.

In at least one embodiment, the establishing an alarm threshold for the evaluation metric includes:

establishing the excitation force order curve of the wheel in the initial driving process;

Measuring the radial dimension deviation of the wheel at each position in the circumferential direction, obtaining a radial dimension deviation matrix, and converting the radial dimension deviation matrix into a roundness order curve;

Comparing the excitation force order curve and the roundness order curve, a curve transfer relationship between the excitation force order curve and the roundness order curve is obtained.

In at least one embodiment, the establishing the excitation force order curve of the wheel during the initial driving process includes:

In the initial running process of the train, measure the axle box acceleration of the axle box during the rotation of the wheel at least one circle to obtain an acceleration matrix;

According to the transfer relationship, the excitation force matrix corresponding to the acceleration matrix is obtained;

Convert the excitation force matrix into an excitation force order curve.

In at least one embodiment, the evaluation quantity is a wheel-rail excitation force, and the establishing an alarm threshold of the evaluation quantity further comprises: determining an alarm threshold of the radial dimension deviation of the wheel, and according to the curve transfer relationship, the The radial dimension deviation alarm threshold is converted into a wheel-rail excitation force alarm threshold, and the wheel-rail excitation force alarm threshold is used as the alarm threshold; or

The evaluation amount is the radial dimension deviation of the wheel. During the running process of the train, the real-time axle box acceleration at the measurement time is obtained, and according to the transfer relationship, the real-time evaluation amount at the measurement time is calculated and obtained, including:

According to the transfer relationship, calculate the real-time excitation force corresponding to the real-time axle box acceleration;

According to the curve transfer relationship, the real-time excitation force is converted into real-time wheel radial dimension deviation.

In at least one embodiment, the comparing the real-time evaluation with the alarm threshold includes:

The real-time evaluation amount belonging to the abnormal value is eliminated, and the real-time evaluation amount belonging to the non-outlier value is compared with the alarm threshold.

In at least one embodiment, the "removing the real-time evaluation quantities belonging to outliers" comprises:

Based on continuous measurement data, high-pass filtering is performed on the real-time evaluation quantity at the measurement moment.

In at least one embodiment, the comparing the real-time evaluation value with the alarm threshold and outputting a judgment result includes:

When the real-time evaluation value is greater than or equal to the alarm threshold, an alarm signal is output.

In at least one embodiment, the measurement location of the axlebox acceleration is located at the top of the wheel corresponding to the axlebox.

According to a second aspect of the present invention, a device for monitoring the roundness of a wheel tread is provided. The wheel is used for a wheelset of a train, and each wheel corresponds to an axle box, including:

The first preprocessing module is used to establish the transmission relationship between the acceleration of the axle box and the excitation force of the wheel and rail;

The second preprocessing module is used to establish the alarm threshold of the evaluation quantity, and the evaluation quantity is the wheel-rail excitation force or the radial dimension deviation of the wheel;

The signal acquisition module is used to obtain the real-time axle box acceleration at the measurement time during the running process of the train; the processing module is used to calculate and obtain the real-time axle box acceleration at the measurement time based on the real-time axle box acceleration and according to the transmission relationship. amount of evaluation;

A judging module for comparing the real-time evaluation value with the alarm threshold to determine whether an alarm is required; and

The alarm module is used to obtain the judgment result of the judgment module and output an alarm signal.

In at least one embodiment, the first preprocessing module is configured to obtain the wheel-rail excitation force loaded on the wheel when the train is in a non-running state, and simultaneously measure the axle box acceleration, The algebraic relationship between the wheel-rail excitation force and the axle box acceleration is calculated to establish the transfer relationship.

In at least one embodiment, the second preprocessing module is configured to, during the initial driving of the wheel,

Establish the order curve of the incentive force;

Obtaining the radial dimension deviation of each part of the wheel in the circumferential direction, obtaining a radial dimension deviation matrix, and converting the radial dimension deviation matrix into a roundness order curve;

In at least one embodiment, the second preprocessing module is further configured to acquire the axlebox acceleration of the axlebox in the process of at least one rotation of the wheel during the initial running of the train to obtain an acceleration matrix ;

And convert the excitation force matrix into an excitation force order curve.

In at least one embodiment, the evaluation quantity is wheel-rail excitation force, the second preprocessing module is further configured to determine an alarm threshold of radial dimension deviation of the wheel, and according to the curve transfer relationship, the diameter Convert to Dimension Deviation Alarm Threshold to Wheel Rail Excitation Force Alarm Threshold; or

The evaluation quantity is the radial dimension deviation of the wheel, and the processing module is further configured to calculate the real-time excitation force corresponding to the real-time axle box acceleration according to the transmission relationship; and according to the curve transmission relationship, the real-time excitation force is Forces are converted to real-time wheel radial dimension deviations.

In at least one embodiment, the judging module is further configured to perform high-pass filtering processing on the real-time evaluation value at the measurement moment based on the obtained continuous real-time evaluation value, and use the processed real-time evaluation value to process the real-time evaluation value after processing. The evaluation quantity is compared to the alarm threshold.

According to a third aspect of the present invention, a monitoring system for the roundness of a wheel tread is provided, comprising:

an acceleration sensor arranged on the axle box corresponding to the wheel; and

According to the monitoring device for wheel tread roundness of the present invention, the acceleration sensor is used for communicating with the signal acquisition module of the monitoring device.

The monitoring method according to the present invention can accurately and timely monitor the roundness of the wheel tread.

Description of drawings

FIG. 1 is a schematic diagram of a wheel provided on a bogie according to an embodiment of the present invention.

2 to 4 are flowcharts of a method for monitoring the roundness of a wheel tread according to an embodiment of the present invention.

5 is a schematic diagram of an excitation force order curve and a circularity order curve according to an embodiment of the present invention.

FIG. 6 is a flowchart of a method for monitoring the roundness of a wheel tread according to an embodiment of the present invention.

7 is a schematic diagram of a monitoring device for the roundness of a wheel tread according to an embodiment of the present invention.

8 is a schematic diagram of a monitoring system for wheel tread roundness according to an embodiment of the present invention.

Detailed ways

Exemplary embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that these specific descriptions are only used to teach those skilled in the art how to implement the present invention, and are not used to exhaust all possible ways of the present invention, nor to limit the scope of the present invention.

Taking a wheel set on a wheelset of a bogie of a train as an example, the method for monitoring the roundness of a wheel tread according to the present invention will be described with reference to FIGS. 1 to 6 .

The roundness of the wheel referred to in the present invention refers to the difference in the radial dimension of the wheel at various points in the circumferential direction, and the smaller the difference is, the better the roundness of the wheel is considered to be.

The monitoring method measures the real-time axle box acceleration at the axle box online, calculates the real-time wheel-rail excitation force at the measurement time, compares the real-time wheel-rail excitation force with the preset wheel-rail excitation force alarm threshold, and judges the wheel circle at the measurement time. degree is acceptable.

Optionally, the measurement position of the axle box acceleration is located at the top of the axle box, close to the position where the wheel is located, for example, an acceleration sensor may be set at this position to measure the axle box acceleration. Preferably, the axle box acceleration refers specifically to the acceleration in the vertical direction of the above-mentioned measurement position of the axle box.

Figure 1 shows a scheme with four wheels on the bogie. In order to grasp the roundness of each wheel, the real-time axle box acceleration is measured at the top of the axle box corresponding to each wheel, and the real-time wheel-rail excitation force is calculated accordingly. For example, at the moment of measurement, the real-time axlebox accelerations corresponding to the four wheels are A1, A2, A3 and A4, respectively, and the corresponding calculated real-time wheel-rail excitation forces of the four wheels are F1, F2, F3, and F4, respectively.

Specifically, referring to Fig. 2, the monitoring method includes:

S10, establishing the transmission relationship H between the axle box acceleration and the wheel-rail excitation force.

S20, establishing the alarm threshold of the wheel-rail excitation force (also called the evaluation quantity);

S30, during the running process of the train, obtain the real-time axle box acceleration at the measurement time, and calculate the real-time wheel-rail excitation force at the measurement time according to the transfer relationship H;

S40, compare the real-time wheel-rail excitation force with the wheel-rail excitation force alarm threshold, and output a judgment result.

It should be understood that the execution order of the above steps may be adjusted according to actual conditions, for example, step S20 may be earlier or later than step S30.

Referring to FIG. 3, for the above step S10, optionally, the establishment of the transfer relationship H is realized by offline measurement. It includes:

S11, when the train is in a non-running state, load wheel-rail excitation force on the wheels, and measure the acceleration of the axle box at the same time;

S12, calculate the algebraic relationship between the wheel-rail excitation force and the axle box acceleration, and obtain the transfer relationship H.

Optionally, for step S11, the wheel is loaded with an external force equal to the wheel-rail excitation force, for example, a heavy hammer can be used to load the external force. Hit the lowest point of the wheel with a heavy hammer in the vertical direction. The magnitude of the force exerted by the heavy hammer is equal to the excitation force of the wheel and rail, and the direction is vertical upward.

Optionally, for step S12, the transfer relationship H may be used to reflect the algebraic relationship between the rail excitation force corresponding to the multiple wheels and the axle box acceleration.

For example, for the four wheels shown in Figure 1, the axlebox acceleration A of the four axleboxes and the wheel-rail excitation force F of the four wheels can be recorded in the form of a matrix.

Then, the transitive relation H can be written as a 4×4 matrix,

A=H×F, so H=A×F ⁻¹ .

It should be understood that the transfer relationship H can also be calculated separately for a single wheel.

Referring to FIG. 4, for the above step S20, optionally, the establishment of the wheel-rail excitation force alarm threshold includes:

S21, establishing the excitation force order curve Cf of the wheel during the initial running process.

The above-mentioned initial running process refers to the running process of the train at the initial stage of each working cycle. During this running process, the wheels are assumed to have no wear or only a small amount of wear. The obtained excitation force order curve is the same as the circle introduced below. The direction of the degree-order curve is theoretically the same. For example, the initial driving process is a driving process in which the wheel is put into use for the first time after the radial dimension deviation is collected offline as described below.

The specific method for establishing the excitation force order curve Cf will be further described below with reference to FIG. 5 .

S22, measure the radial dimension deviation of the wheel at each position in the circumferential direction, obtain a radial dimension deviation curve, and convert the radial dimension deviation curve into a roundness order curve Cr.

Optionally, the radial dimension deviation of the wheel at each position in the circumferential direction can be collected, for example, by using a wheel diameter jumper in an offline state (when the train is not running). The radial dimension deviation data of each part in the circumferential direction may be continuous data, or may be discrete data with small intervals in the circumferential direction. These data form radial dimensional deviation curves.

The Fourier expansion of the radial dimension deviation curve can obtain the roundness order curve Cr.

S23 , comparing the excitation force order curve Cf and the roundness order curve Cr to obtain a curve transfer relationship Hc between the excitation force order curve Cf and the roundness order curve Cr.

Theoretically, the direction of the excitation force order curve and the circularity order curve are the same, and only show differences in the amplitude. The difference in magnitude can be represented by the curve transfer relationship Hc.

For example, referring to FIG. 5 , at each order, the value Cf0 on the excitation force order curve Cf and the value Cr0 on the roundness order curve Cr have the following relationship: Cf0=Hc×Cr0.

S24, determine the radial size deviation alarm threshold of the wheel, and convert the radial size deviation alarm threshold into the wheel-rail excitation force alarm threshold according to the curve transfer relationship Hc.

Referring to FIG. 6, for the above step S21, optionally, the establishment of the excitation force order curve includes:

S211, during the initial running of the train, measure the acceleration of the axle box corresponding to the axle box in the process of at least one rotation of the wheel to obtain an acceleration curve;

S212, according to the transfer relationship H, obtain the excitation force curve corresponding to the acceleration curve;

S213, convert the excitation force curve into the excitation force order curve Cf. Specifically, this transformation is achieved by Fourier transform.

Preferably, when step S40 is performed, it is necessary to perform an abnormal value judgment on the calculated real-time wheel-rail excitation force. That is, it is determined whether the real-time wheel-rail excitation force obtained at the measurement moment belongs to an abnormal value, and the real-time wheel-rail excitation force belonging to the abnormal value will be eliminated.

This is because the wheels may be subjected to abnormal wheel-rail excitation force due to some interference factors during the operation of the train. For example, due to the accumulation of foreign objects such as small stones on the track, or due to the wear of the track, the wheel will be subjected to abnormal wheel-rail excitation force during the process of passing the above-mentioned track. The above-mentioned abnormal wheel-rail excitation force cannot directly indicate that the wheel roundness is abnormal. Therefore, the above-mentioned abnormal wheel-rail excitation force needs to be eliminated as an abnormal value, and the real-time wheel-rail excitation force and wheel-rail excitation force belonging to non-abnormal values Alarm thresholds are compared.

Preferably, the method for removing outliers includes: based on continuous measurement data, performing high-pass filtering processing on the real-time wheel-rail excitation force at the measurement moment.

For step S40, optionally, when the real-time wheel-rail excitation force is greater than or equal to the wheel-rail excitation force alarm threshold, it is determined that the wheel roundness is abnormal, and an alarm signal is output.

Referring to FIG. 7 , the present invention also provides a device for monitoring the roundness of the tread of a wheel. The wheels are the wheels of a train wheelset, and an axle box is correspondingly provided beside each wheel. The monitoring device includes a first preprocessing module, a second preprocessing module, a signal acquisition module, a processing module, a judgment module and an alarm module.

The first preprocessing module is used to establish the transmission relationship H between the axle box acceleration and the wheel-rail excitation force.

The second preprocessing module is used to establish an alarm threshold for an evaluation variable (eg wheel-rail excitation force).

The signal acquisition module is used to obtain the real-time axle box acceleration at the measurement moment during the running process of the train.

The processing module is used to calculate and obtain the real-time wheel-rail excitation force at the measurement moment based on the real-time axle box acceleration and the transfer relation H.

The judging module is used to compare the real-time wheel-rail excitation force with the wheel-rail excitation force alarm threshold to determine whether an alarm is required.

The alarm module is used to obtain the judgment result of the judgment module, and output an alarm signal when an alarm is required.

Optionally, the first preprocessing module is used to obtain the wheel-rail excitation force loaded on the wheels when the train is in a non-running state, measure the axle box acceleration, and calculate the algebra between the wheel-rail excitation force and the axle box acceleration. relationship, thereby establishing a transitive relationship H.

Optionally, the transfer relationship H may be pre-stored in the first preprocessing module, or may be input or imported into the first preprocessing module by an operator, for example, in stages for updating.

Optionally, the second preprocessing module is configured to establish the excitation force order curve Cf during the initial running of the wheel;

Obtain the radial dimension deviation of each part of the wheel in the circumferential direction, obtain the radial dimension deviation matrix, and convert the radial dimension deviation matrix into the roundness order curve Cr;

Comparing the excitation force order curve Cf and the roundness order curve Cr, the curve transfer relationship Hc between the excitation force order curve Cf and the roundness order curve Cr is obtained;

Determine the radial size deviation alarm threshold of the wheel, and convert the radial size deviation alarm threshold into the wheel-rail excitation force alarm threshold according to the curve transfer relationship Hc.

Optionally, the second preprocessing module is also used for:

In the initial running process of the train, obtain the axle box acceleration of the axle box in the process that the wheel rotates at least one circle, and obtain the acceleration matrix;

According to the transfer relation H, the excitation force matrix corresponding to the acceleration matrix is obtained;

And convert the excitation force matrix into the excitation force order curve Cf.

Optionally, the wheel-rail excitation force alarm threshold may be pre-stored in the second preprocessing module, or may be input or imported into the second preprocessing module by the operator, for example, in stages for updating.

Optionally, the judgment module is also used to perform high-pass filter processing on the real-time wheel-rail excitation force at the measurement time based on the obtained continuous real-time wheel-rail excitation force, and compare the processed real-time wheel-rail excitation force with the alarm threshold. .

Optionally, the alarm signal output by the alarm module may be displayed in the form of, for example, sound, light, text, and image, and may also be, for example, a digital signal or an analog signal transmitted to other control units.

Referring to FIG. 8 , the present invention further provides a monitoring system for the roundness of the wheel tread, which includes an acceleration sensor and the above-mentioned monitoring device for the roundness of the wheel tread. The acceleration sensor is arranged on the axle box corresponding to the wheel (closest to the wheel), preferably on the top of the axle box. The acceleration sensor is used to communicate with the signal acquisition module of the monitoring device.

Some beneficial effects of the above-mentioned embodiments of the present invention are briefly described below.

(i) The present invention establishes the relationship between the three physical quantities of axle box vibration (axle box acceleration), wheelset excitation force and wheel roundness, so that wheel roundness can be evaluated by measuring axle box vibration data in real time, and the wheel The monitoring of tread roundness is accurate and timely.

(ii) When evaluating the wheel tread roundness based on the real-time axlebox acceleration (or the calculated real-time wheelset excitation force), first determine the abnormal value caused by the interference factor, and eliminate the abnormal value through, for example, high-pass filtering to ensure that the accuracy of the monitoring method.

It should be understood that the above-mentioned embodiments are only exemplary, and are not intended to limit the present invention. Those skilled in the art can make various modifications and changes to the above-described embodiments under the teachings of the present invention without departing from the scope of the present invention. E.g:

(i) In the case of considering the wheel mass, the wheel-rail excitation force obtained by the above calculation can also be converted into the wheel tread acceleration, so that the numerical value of the wheel tread acceleration can be used for the calculation of the intermediate process.

(ii) Since there is a curve transfer relationship Hc between the excitation force order curve Cf and the roundness order curve Cr, the physical quantity involved in the judgment can also be replaced by the excitation force during the monitoring process. The radial dimension deviation, that is, The radial dimension deviation of the wheel is used as the evaluation quantity to substitute the wheel-rail excitation force. For example, monitoring methods include:

S100, establishing the transmission relationship H between the acceleration of the axle box and the excitation force of the wheel and rail;

S200, establishing an alarm threshold for the radial dimension deviation of the wheel (ie, the evaluation quantity);

S300, during the running of the train, obtain the real-time axle box acceleration at the measurement time, and calculate the real-time wheel radial dimension deviation at the measurement time according to the transfer relationship H;

S400, compare the real-time wheel radial size deviation with the wheel radial size deviation alarm threshold, and output a judgment result.

Wherein, step S200 may be implemented with reference to the aforementioned steps S21, S22 and S23.

Wherein, obtaining the real-time axle box acceleration at the measurement moment in step S21, and according to the transfer relationship H, calculating the real-time evaluation quantity at the measurement moment includes:

According to the transfer relation H, calculate the real-time excitation force corresponding to the real-time axle box acceleration;

According to the curve transfer relationship Hc, the real-time excitation force is converted into real-time wheel radial dimension deviation.

(iii) Using the method of matrix operation, the tread roundness of multiple wheels can be monitored at the same time.

Claims

A method for monitoring the roundness of a wheel tread, wherein the wheel is used for a wheelset of a train, and each wheel corresponds to an axle box, wherein the monitoring method comprises:

Establish the transmission relationship (H) between the acceleration of the axle box and the excitation force of the wheel and rail;

Establishing an alarm threshold of an evaluation quantity, the evaluation quantity being the wheel-rail excitation force or the radial dimension deviation of the wheel;

During the running process of the train, the real-time axle box acceleration at the measurement time is obtained, and the real-time evaluation value at the measurement time is obtained by calculation according to the transfer relationship (H);

The real-time evaluation value is compared with the alarm threshold, and a judgment result is output.
The method for monitoring wheel tread roundness according to claim 1, wherein the establishing a transmission relationship (H) between the acceleration of the axle box and the wheel-rail excitation force comprises:

When the train is in a non-running state, the wheel is loaded with the wheel-rail excitation force, and the axle box acceleration is measured at the same time;

Calculate the algebraic relationship between the wheel-rail excitation force and the axle box acceleration to obtain the transfer relationship (H).
The method for monitoring wheel tread roundness according to claim 2, wherein the loading the wheel-rail excitation force to the wheel comprises:

The wheel is hit with an external force equal to the wheel-rail excitation force.
The method for monitoring wheel tread roundness according to claim 1, wherein the establishing an alarm threshold of the evaluation quantity comprises:

establishing the excitation force order curve (Cf) of the wheel in the initial driving process;

Measure the radial dimension deviation of the wheel at each position in the circumferential direction, obtain a radial dimension deviation matrix, and convert the radial dimension deviation matrix into a roundness order curve (Cr);

Comparing the excitation force order curve (Cf) and the roundness order curve (Cr), the curve transfer between the excitation force order curve (Cf) and the roundness order curve (Cr) is obtained Relationship (Hc).
The method for monitoring wheel tread roundness according to claim 4, wherein the establishing the excitation force order curve (Cf) of the wheel in the initial running process comprises:

In the initial running process of the train, measure the axle box acceleration of the axle box during the rotation of the wheel at least one circle to obtain an acceleration matrix;

According to the transfer relationship (H), the excitation force matrix corresponding to the acceleration matrix is obtained;

Convert the excitation force matrix into an excitation force order curve (Cf).
The method for monitoring wheel tread roundness according to claim 4 or 5, wherein the evaluation quantity is wheel-rail excitation force, and establishing an alarm threshold for the evaluation quantity further comprises: determining the radial dimension of the wheel Deviation alarm threshold, according to the curve transfer relationship (Hc), converting the radial dimension deviation alarm threshold into a wheel-rail excitation force alarm threshold, and the wheel-rail excitation force alarm threshold as the alarm threshold; or

The evaluation quantity is the radial dimension deviation of the wheel. During the running process of the train, the real-time axle box acceleration at the measurement time is obtained, and according to the transfer relationship (H), the real-time evaluation quantity at the measurement time is calculated to include: :

According to the transfer relationship (H), calculate the real-time excitation force corresponding to the real-time axle box acceleration;

According to the curve transfer relationship (Hc), the real-time excitation force is converted into real-time wheel radial dimension deviation.
The method for monitoring wheel tread roundness according to claim 1, wherein the comparing the real-time evaluation value with the alarm threshold comprises:

The real-time evaluation amount belonging to the abnormal value is eliminated, and the real-time evaluation amount belonging to the non-outlier value is compared with the alarm threshold.
The method for monitoring the roundness of wheel tread according to claim 7, wherein the "removing the real-time evaluation quantity belonging to abnormal values" comprises:

Based on continuous measurement data, high-pass filtering is performed on the real-time evaluation quantity at the measurement moment.
The method for monitoring wheel tread roundness according to claim 1, wherein the comparing the real-time evaluation value with the alarm threshold, and outputting a judgment result, comprises:

When the real-time evaluation value is greater than or equal to the alarm threshold, an alarm signal is output.
The method for monitoring wheel tread roundness according to claim 2 or 5, wherein the measurement position of the axle box acceleration is located at the top of the wheel corresponding to the axle box.
A monitoring device for the roundness of wheel tread, the wheel is used for the wheelset of a train, and each wheel corresponds to an axle box, characterized in that it includes:

The first preprocessing module is used to establish the transmission relationship (H) between the acceleration of the axle box and the excitation force of the wheel and rail;

The second preprocessing module is used to establish an alarm threshold of an evaluation quantity, and the evaluation quantity is wheel-rail excitation force or wheel radial dimension deviation;

a signal acquisition module for acquiring the real-time axlebox acceleration at the measurement moment during the running process of the train; a processing module for obtaining the measurement based on the real-time axlebox acceleration and according to the transfer relationship (H) real-time evaluation of time;

A judging module for comparing the real-time evaluation value with the alarm threshold to determine whether an alarm is required; and

The alarm module is used to obtain the judgment result of the judgment module and output an alarm signal.
The device for monitoring wheel tread roundness according to claim 11, wherein the first preprocessing module is configured to acquire the wheel load on the wheel when the train is in a non-running state The rail excitation force is measured, and the axle box acceleration is measured, and the algebraic relationship between the wheel-rail excitation force and the axle box acceleration is calculated to establish the transfer relationship (H).
The device for monitoring the roundness of wheel tread according to claim 11, wherein the second preprocessing module is configured to:

Establish the excitation force order curve (Cf);

Acquiring the radial dimension deviation of each part of the wheel in the circumferential direction, obtaining a radial dimension deviation matrix, and converting the radial dimension deviation matrix into a roundness order curve (Cr);

Comparing the excitation force order curve (Cf) and the roundness order curve (Cr), the curve transfer between the excitation force order curve (Cf) and the roundness order curve (Cr) is obtained Relationship (Hc).
The device for monitoring the roundness of the wheel tread according to claim 13, wherein the second preprocessing module is further configured to acquire, during the initial running of the train, the process of obtaining at least one rotation of the wheel. The axle box acceleration of the axle box is obtained, and the acceleration matrix is obtained;

According to the transfer relationship (H), the excitation force matrix corresponding to the acceleration matrix is obtained;

And convert the excitation force matrix into an excitation force order curve (Cf).
The monitoring device for wheel tread roundness according to claim 13 or 14, wherein the evaluation quantity is wheel-rail excitation force, and the second preprocessing module is further configured to determine the radial dimension deviation of the wheel an alarm threshold, converting the radial dimension deviation alarm threshold into a wheel-rail excitation force alarm threshold according to the curve transfer relationship (Hc); or

The evaluation quantity is the radial dimension deviation of the wheel, and the processing module is further configured to calculate the real-time excitation force corresponding to the real-time axle box acceleration according to the transfer relationship (H); and according to the curve transfer relationship (Hc) , and convert the real-time excitation force into real-time wheel radial dimension deviation.
The device for monitoring the roundness of the wheel tread according to claim 11, wherein the judgment module is further configured to, based on the acquired continuous real-time evaluations, perform an evaluation on the real-time evaluations at the measurement time. High-pass filtering is performed, and the processed real-time evaluation quantity is compared with the alarm threshold.
A monitoring system for wheel tread roundness, comprising:

an acceleration sensor arranged on the axle box corresponding to the wheel; and

According to the monitoring device for wheel tread roundness according to any one of claims 12 to 16, the acceleration sensor is configured to communicate with the signal acquisition module of the monitoring device.