A method and system for testing a train crossing a bridge

Abstract

The application discloses a kind of train test method and system through bridge, applied to rail transit technical field, comprising: receiving bridge response data exported by bridge model;Configuration is used to indicate the vertical displacement of different bridge position point Monadic linear regression model;When receiving a group of bridge response data at the same time, fitting is carried out based on the group of bridge response data Monadic linear regression model, obtain the Monadic linear regression model after fitting;Based on the group of bridge response data and corresponding Monadic linear regression model after fitting, corresponding attitude control instruction is sent to the vibration table to adjust the current attitude of the vibration table;In the process that train passes through vibration table, the force of specified position is detected by sensor and used as the input parameter of bridge model, and the record of detection data is carried out.The scheme of the application can effectively carry out the test that train passes through bridge, and accurate and effective test results are obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of rail transit, in particular to a test method and system for train passing through a bridge. BACKGROUND

[0002] In recent years, with the rapid development of highways and railways in China, the trend of heavy load and high speed of vehicles is obvious, and the vehicle-bridge dynamic coupling phenomenon is more and more prominent. The safety and comfort of vehicles on the bridge have become a research hotspot. When the vehicle passes through the bridge, the gravity of the vehicle itself and the impact of the running vehicle will cause the vibration of the bridge, and vice versa. The vibration of the bridge will affect the running vehicle. The vibration phenomenon of mutual coupling between the vehicle and the bridge is called vehicle-bridge coupling vibration. Excessive vibration may cause bridge fatigue and damage, and affect the safety and comfort of vehicle operation.

[0003] In the current scheme, when studying vehicle-bridge coupling vibration, there are wheel-rail relationship normal point calculation algorithm, simulation method based on Client-Server technology, etc. Usually, the test needs to be realized based on a simulation platform. Although it is relatively convenient, the accuracy of the test needs to be improved.

[0004] In summary, how to effectively test the train passing through the bridge and obtain accurate and effective test results is a technical problem that needs to be solved by the technical personnel in the field at present. SUMMARY

[0005] The purpose of the present application is to provide a test method and system for train passing through a bridge, so as to effectively test the train passing through the bridge and obtain accurate and effective test results.

[0006] To solve the above technical problems, the present application provides the following technical scheme:

[0007] A test method for train passing through a bridge, comprising:

[0008] receiving bridge response data output by a bridge model;

[0009] configuring a linear regression model for representing the vertical displacement of different bridge position points;

[0010] Whenever a group of bridge response data at the same time is received, fitting the linear regression model based on the group of bridge response data is performed, and a fitted linear regression model is obtained;

[0011] Based on the group of bridge response data and the corresponding fitted linear regression model, a corresponding attitude control instruction is sent to the vibration table to adjust the current attitude of the vibration table;

[0012] During the train passing through the shaking table, the force at the designated position is detected by the sensor and taken as the input parameter of the bridge model, and the detection data is recorded.

[0013] Preferably, the shaking table is two-dimensional or three-dimensional.

[0014] Preferably, the shaking table is two-dimensional, and the bridge response data includes vertical displacement of the bridge.

[0015] According to the first period, the bridge model outputs the bridge vertical displacement based on the linear interpolation algorithm.

[0016] The linear interpolation algorithm is expressed as: m is an integer and sequentially takes values from 0 to B / A; k represents different sampling time points; z ba (k) represents the bridge vertical displacement output by the bridge model at the kth sampling time point; represents the bridge vertical displacement at the kth sampling time point obtained by the linear interpolation algorithm; B represents the time step of the bridge model without interpolation; and A represents the period of the bridge response data output by the bridge model after interpolation.

[0017] Preferably, the shaking table is two-dimensional, and the bridge response data at the same time includes the bridge vertical displacement at n different positions at the time; n is a positive integer not less than 2.

[0018] Correspondingly, the fitting of the one-dimensional linear regression model is performed based on the bridge response data at the same time, and a fitted one-dimensional linear regression model is obtained.

[0019] The fitting of the one-dimensional linear regression model is performed based on the n bridge vertical displacements at the same time, and a fitted one-dimensional linear regression model is obtained. wherein, x t represents the center position of the shaking table, z t represents the vertical displacement of the center position of the shaking table at the time, and are model parameters of the one-dimensional linear regression model at the time determined by fitting.

[0020] ​Preferably, based on the bridge response data described in this group and the corresponding fitted univariate linear regression model, a corresponding attitude control command is sent to the shaking table to adjust the current attitude of the shaking table, including:

[0021] Based on the bridge response data described in this group and the corresponding fitted univariate linear regression model, a signal carrying z-axis data was sent to the shaking table. to and pitch to The attitude control commands are used to adjust the current attitude of the vibration table;

[0022] Among them, z to This represents the vertical displacement of the center position of the shaking table, which is superimposed with track irregularities, and z to =z t +z ir , z ir For track irregularities, pitch to This represents the rotation angle of the vibration table around the y-axis, and

[0023]

[0024] Preferably, the vibration mode of the shaking table is three-dimensional vibration, and the bridge response data includes the vertical displacement of the bridge and the rotation angle of the bridge about the x-axis. Correspondingly, the bridge response data output by the bridge model includes:

[0025] According to the first cycle, the vertical displacement of the bridge and the rotation angle of the bridge around the x-axis are output by the bridge model based on the linear interpolation algorithm.

[0026] The linear interpolation algorithm is expressed as follows: m is an integer and takes values ​​from 0 to B / A; k represents different sampling time points; z ba (k) represents the vertical displacement of the bridge output by the bridge model at the k-th sampling time point; roll ba (k) represents the angle of rotation of the bridge around the x-axis output by the bridge model at the k-th sampling time point; This represents the result obtained through the linear interpolation algorithm at the 1st... The vertical displacement of the bridge at any given moment; This represents the result obtained through the linear interpolation algorithm at the 1st... B represents the angle of rotation of the bridge around the x-axis at time t; B represents the time step of the bridge model without interpolation; A represents the period of the bridge response data output by the bridge model after interpolation.

[0027] Preferably, the vibration mode of the vibration table is three-dimensional vibration, the vibration table is connected by a vibration units in x-axis direction, the bridge response data at the same time is specifically: vertical displacement of the bridge at q different positions at the time, and rotation angle of the bridge around x-axis at q different positions; q is a positive integer not less than 2; the one-dimensional linear regression model comprises a one-dimensional linear regression sub-models;

[0028] Correspondingly, each time a group of bridge response data at the same time is received, fitting of the one-dimensional linear regression model is performed based on the group of bridge response data, to obtain a fitted one-dimensional linear regression model, comprising:

[0029] Each time a group of bridge response data at the same time is received, the q vertical displacements of the bridge in the group are divided into corresponding one-dimensional linear regression sub-models according to positions, and fitting of each one-dimensional linear regression sub-model is performed, to obtain a fitted a one-dimensional linear regression sub-models, and the jth fitted one-dimensional linear regression sub-model is expressed as: Wherein, x tj represents the central position of the jth vibration unit, z tj represents the vertical displacement of the central position of the jth vibration unit at the time, and are model parameters of the jth one-dimensional linear regression sub-model at the time determined by fitting, j is a positive integer and 1≤j≤a.

[0030] Preferably, based on the group of bridge response data and the corresponding fitted one-dimensional linear regression model, a corresponding attitude control instruction is sent to the vibration table to adjust the current attitude of the vibration table, comprising:

[0031] Based on the group of bridge response data and the corresponding fitted one-dimensional linear regression model, an attitude control instruction carrying z to1 to z toa , pitch to1 to pitch toa , roll t1 to roll ta is sent to the vibration table to adjust the current attitude of the vibration table.

[0032] Wherein, z to1 represents the vertical displacement of the central position of the first vibration unit superimposed with track irregularity, z toa represents the vertical displacement of the central position of the a vibration unit superimposed with track irregularity, and z toj =z tj +z ir , z irj is track irregularity.

[0033] pitch to1 represents the rotation angle of the first vibration unit around the y-axis, pitch toa represents the rotation angle of the a-th vibration unit around the y-axis, and

[0034] roll t1 represents the rotation angle of the first vibration unit around the x-axis, roll ta represents the rotation angle of the a-th vibration unit around the y-axis.

[0035] Preferably, q = 2a + 1, and the q different positions specifically include: two end positions of each of the a vibration units, and a center position of each of the a vibration units;

[0036] Correspondingly, roll 2j-1 represents the rotation angle of the bridge around the x-axis at the 2j-1 position; roll 2j+1 represents the rotation angle of the bridge around the x-axis at the 2j+1 position, roll tj represents the rotation angle of the j-th vibration unit around the x-axis.

[0037] A test system for a train passing through a bridge, comprising:

[0038] A bridge response data receiving module, configured to receive bridge response data output by a bridge model;

[0039] A linear regression model configuration module, configured to configure a linear regression model for representing vertical displacement of different bridge position points;

[0040] A fitting module, configured to, when a group of the bridge response data at the same time is received, perform fitting of the linear regression model based on the group of the bridge response data, to obtain a fitted linear regression model;

[0041] A posture control module, configured to, based on the group of the bridge response data and the corresponding fitted linear regression model, send a corresponding posture control instruction to a vibration table to adjust a current posture of the vibration table;

[0042] A feedback recording module, configured to, in a process of the train passing through the vibration table, detect force at a specified position by a sensor and take the force as an input parameter of the bridge model, and record detection data.

[0043] By using the technical solution provided by the embodiment of the present application, the test of the train passing through the bridge is realized based on the boundary coordination algorithm, specifically, after receiving the bridge response data output by the bridge model, the attitude control instruction is obtained through the boundary coordination algorithm to adjust the current attitude of the vibration table, and in the process of the train passing through the vibration table, the force at the specified position can be detected by the sensor and then used as the input parameter of the bridge model, and the detection data can also be recorded. Since the test of the train passing through the bridge is realized based on the boundary coordination algorithm in the present application, the bridge is simulated by the vibration table, which is conducive to obtaining accurate and effective test results under the premise of ensuring the cost. In the process of the train passing through the vibration table, since the vibration table cannot directly simulate the deformation of the bridge in the form of displacement curve, the present application configures a linear regression model for representing the vertical displacement of different bridge position points, and after receiving a group of bridge response data at the same time, the linear regression model can be fitted based on the group of bridge response data to obtain the fitted linear regression model, so that the attitude of the vibration table can be controlled based on the fitted linear regression model. In summary, the scheme of the present application can effectively perform the test of the train passing through the bridge and obtain accurate and effective test results. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0045] Figure 1 An implementation flowchart of a train passing through a bridge test method in the present application;

[0046] Figure 2a A principle diagram of a bridge and a vibration table 9-point fitting in a specific embodiment;

[0047] Figure 2b A flowchart of a train passing through a vibration table hybrid experiment in a specific embodiment;

[0048] Figure 3 A principle diagram of a bridge and a vibration table 2-point fitting in a specific embodiment;

[0049] Figure 4 A structure diagram of a train passing through a bridge test system in the present application. DETAILED DESCRIPTION

[0050] The core of the present application is to provide a train passing through a bridge test method, which can effectively perform the test of the train passing through the bridge and obtain accurate and effective test results.

[0051] In order to make the person skilled in the art better understand the present application, the present application is further described in detail below in combination with the drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0052] Please refer to Figure 1 , Figure 1 For the implementation flowchart of the train passing through the bridge test method in the present application, the train passing through the bridge test method can include the following steps:

[0053] Step S101: receiving the bridge response data output by the bridge model.

[0054] Specifically, in the scheme of the present application, the deformation of the bridge is simulated by the vibration table, so that the real bridge does not need to be established in order to save the cost, but the bridge model needs to be established, so as to provide the basis for the attitude change of the vibration table in the process of the train passing through the vibration table.

[0055] The specific content of the bridge response data output by the bridge model can be set according to the actual needs, but generally at least the vertical displacement of the bridge at different positions is needed, that is, the displacement perpendicular to the z-axis direction of the bridge surface.

[0056] In a specific embodiment of the present application, the vibration mode of the vibration table can be set as two-dimensional vibration, at this time the bridge response data includes the vertical displacement of the bridge, such an embodiment is relatively simple. Another embodiment is to set the vibration mode of the vibration table as three-dimensional vibration, such an embodiment is beneficial to obtain more accurate test results, the bridge response data needs to include the vertical displacement z of the bridge and the roll angle of the bridge around the x-axis, in the embodiments hereinafter of the present application, the vibration mode of the vibration table is taken as two-dimensional vibration or three-dimensional vibration as an example, which is described respectively.

[0057] In a specific embodiment of the present application, the vibration mode of the vibration table is two-dimensional vibration, and the bridge response data includes the vertical displacement of the bridge. Correspondingly, step S101 can specifically include:

[0058] According to the first period, the bridge vertical displacement output by the bridge model based on the linear interpolation algorithm is received;

[0059] Wherein, the linear interpolation algorithm is represented as: m is an integer and sequentially takes the value of 0 to B / A; k represents different sampling time points; z ba(k) represents the vertical displacement of the bridge output by the bridge model at the k-th sampling time point; This represents the result obtained through a linear interpolation algorithm at the 1st... B represents the vertical displacement of the bridge at time t; B represents the time step of the bridge model without interpolation; A represents the period of the bridge response data output by the bridge model after interpolation.

[0060] As described above, when the vibration mode of the shaking table is two-dimensional vibration, the bridge response data includes the vertical displacement of the bridge, that is, the bridge response data at a certain moment, which is composed of the vertical displacement of the bridge at multiple different bridge positions at that moment. The different bridge positions described here refer to different positions in the x-axis direction. The x-axis direction described in this application is also the extension direction of the bridge.

[0061] In this implementation, the vertical displacement of the bridge generated by the bridge model is interpolated. This is because in practical applications, the bridge response data output by the bridge model is usually received periodically. However, in some cases, without interpolation, the period of the bridge response data output by the bridge model is long. In order to obtain more accurate test results, subsequent steps should be executed at a higher frequency. In other words, if the period of the bridge response data output by the bridge model is long, it will not meet the requirements. For example, in a specific case, the time step of the bridge model is 1 / 128 seconds, while the time step of other modules is 1 / 2048 seconds.

[0062] Therefore, in this implementation, the vertical displacement of the bridge generated by the bridge model is interpolated, which effectively increases the frequency of the bridge model outputting bridge response data.

[0063] In this implementation, the linear interpolation algorithm is expressed as:

[0064] B represents the time step of the bridge model without interpolation, that is, the period at which the bridge model outputs the bridge response data without interpolation. For example, in the example above, B = 1 / 128 seconds. A represents the period at which the bridge model outputs the bridge response data after interpolation, for example, A = 1 / 2048 seconds in the example above. In this example, the linear interpolation algorithm is specifically as follows: m takes values ​​from 0 to 16. In this example, the first period is A, which is 1 / 2048 seconds.

[0065] Step S102: Configure a univariate linear regression model to represent the vertical displacement of different bridge locations.

[0066] Step S103: After receiving a set of bridge response data at the same time, a fitting of the one-dimensional linear regression model is performed based on the set of bridge response data, and a fitted one-dimensional linear regression model is obtained.

[0067] Still taking the two-dimensional vibration as an example, the steps S102 to S105 are described. As described above, the bridge response data includes the vertical displacement of the bridge.

[0068] It can be understood that, in the subsequent steps of the present application, a fitting of the one-dimensional linear regression model based on the bridge response data is required. Therefore, each time the bridge response data output by the bridge model is received, a set of data at the same time is received, which represents the vertical displacement of the bridge at different positions at the same time.

[0069] After obtaining a set of data at a certain time, a fitting of the one-dimensional linear regression model can be performed based on the set of data.

[0070] In one specific embodiment of the present application, the vibration mode of the vibration table is two-dimensional vibration. The set of bridge response data at the same time specifically includes the vertical displacement of the bridge at n different positions at the same time.

[0071] Correspondingly, the step S103 can specifically include:

[0072] After receiving a set of bridge response data at the same time, a fitting of the one-dimensional linear regression model is performed based on the n vertical displacements of the bridge at the same time, and a fitted one-dimensional linear regression model is obtained. The fitted one-dimensional linear regression model is represented as: wherein x t represents the center position of the vibration table, z t represents the vertical displacement of the center position of the vibration table at the same time, and are model parameters of the one-dimensional linear regression model at the same time determined through the fitting.

[0073] In this embodiment, each time the bridge response data output by the bridge model is received, n data at the same time are received, which represent the vertical displacement of the bridge at n different positions in the x-axis direction. n is a positive integer not less than 2, representing the number of points to be fitted, and the specific value can be set and adjusted as required, for example, n is set to 9 in one specific embodiment, i.e., 9 sampling points are used for fitting.

[0074] The fitted one-dimensional linear regression model can be represented as: wherein the and The value of the fitting needs to be determined by fitting. The specific implementation of the fitting can be various, for example, one commonly used way is to calculate the residual sum of squares to implement the fitting.

[0075] Specifically, the residual sum of squares Q is Here, z i is the vertical displacement of the bridge at the i-th position among the n different positions at this moment. i is the i-th position among the n different positions at this moment.

[0076] The purpose of the linear fitting is to minimize the value of the residual sum of squares Q, which becomes an extreme value problem, which can be obtained by derivation, that is, the Q value calculation formula and are derived, respectively, so that the value of Q is minimized and Therefore, the one-dimensional linear regression model after fitting based on the bridge response data of the group is obtained

[0077] For reference Figure 2a is the schematic diagram of the bridge and the shaking table 9 fitting, that is, n is set to 9. In Figure 2a , 21 needs to represent the upper surface of the shaking table; the serial number 22 represents the bridge response point, that is, it reflects the vertical displacement of the bridge at a certain position among the n different positions, and the serial number 23 represents the pitch angle, that is, the rotation angle of the shaking table around the y axis.

[0078] Step S104: based on the bridge response data of the group and the corresponding one-dimensional linear regression model after fitting, a corresponding attitude control instruction is sent to the shaking table to adjust the current attitude of the shaking table.

[0079] After obtaining the one-dimensional linear regression model after fitting, based on the bridge response data of the group and the one-dimensional linear regression model after fitting, a corresponding attitude control instruction is sent to the shaking table, so that the current attitude of the shaking table can be adjusted.

[0080] When the vibration mode of the shaking table is two-dimensional vibration, the attitude control instruction sent to the shaking table only needs to include two contents, that is, the vertical displacement z of the shaking table and the rotation angle pitch around the y axis. For the vertical displacement z of the shaking table, the center position of the shaking table can be substituted into the one-dimensional linear regression model after fitting, and the vertical displacement of the center position of the shaking table obtained is the required vertical displacement of the shaking table, and the rotation angle pitch of the shaking table around the y axis can be determined by the one-dimensional linear regression model.

[0081] In one specific embodiment of the present application, step S104 can specifically include:

[0082] Based on this set of bridge response data and the corresponding fitted univariate linear regression model, a signal carrying z-axis data was sent to the shaking table. to and pitch to The attitude control commands are used to adjust the current attitude of the vibration table;

[0083] Among them, z to This represents the vertical displacement of the center position of the shaking table, which is superimposed with track irregularities, and z to =z t +z ir , z ir For track irregularities, pitch to This represents the rotation angle of the vibration table around the y-axis, and

[0084]

[0085] This implementation takes into account that uneven wear on the rail surface due to train operation causes track irregularities. Therefore, when sending vertical displacement to the vibration table, the track irregularities can be superimposed to obtain more accurate test results. Thus, in this implementation, through... The vertical displacement z at the center of the shaking table is obtained. t Then, the orbital irregularity z will be superimposed. ir , will z to This is the vertical displacement sent to the vibration table.

[0086] Track irregularity z ir There are various ways to calculate it. For example, in one specific implementation, the orbital irregularity z can be generated based on the power spectral density formula. ir Let f(S) be the expression. Ω represents the spatial frequency, and A v Ω is the track roughness constant. r and Ω c These are the relevant parameters.

[0087] When the vibration mode of the vibration table is two-dimensional vibration, the vibration table only needs to be set with one vibration element, that is, the vibration table has only one vibration table surface. The vibration table receives the signal carrying z-axis. to and pitch to After receiving the attitude control command, the vertical displacement of the center position of the vibration table will be set to z. to Set the angle between the vibration table surface and the x-axis as pitch. to That is, the rotation angle of the vibration table surface around the y-axis is set as pitch. to .

[0088] Figure 2b This is a schematic diagram of the process of a train undergoing a mixed test on a vibration table. Figure 2bIn an embodiment, the track irregularity amount is sent to the boundary coordination algorithm, the boundary coordination algorithm performs fitting of the one-dimensional linear regression model based on the received bridge response data output by the bridge model, and then sends the attitude control instruction carrying z to and pitch to to the vibration table.

[0089] Step S105: During the process of the train passing through the vibration table, the force at the specified position is detected by the sensor as an input parameter of the bridge model, and the detection data is recorded.

[0090] During the process of the train passing through the vibration table, the bridge response data output by the bridge model is received each time to determine the corresponding attitude control instruction, which is then sent to the vibration table to adjust the current attitude of the vibration table.

[0091] During the process of the train passing through the vibration table, the attitude of the vibration table changes continuously, and the train also moves forward continuously. The scheme of the present application can detect the force at the specified position by the sensor as an input parameter of the bridge model. The specific sensor detection position can be set and adjusted according to actual needs. After the bridge model receives the detection data of the corresponding sensor, the bridge model changes and outputs corresponding bridge response data.

[0092] During the process of the train passing through the vibration table, the data detected by the sensor can be recorded for subsequent data analysis. In addition, during the process of the train passing through the vibration table, the attitude change of the vibration table and the related parameters of the train can also be recorded to realize the test of the train passing through the bridge.

[0093] In the foregoing, the vibration mode of the vibration table is taken as two-dimensional vibration as an example for description, and in the following, the vibration mode of the vibration table is taken as three-dimensional vibration as an example for description.

[0094] In a specific embodiment of the present application, the vibration mode of the vibration table is three-dimensional vibration, the bridge response data includes the bridge vertical displacement and the bridge rotation angle around the x-axis, and correspondingly, step S101 can specifically include:

[0095] According to the first period, the bridge vertical displacement and the bridge rotation angle around the x-axis output by the bridge model based on the linear interpolation algorithm are received;

[0096] The linear interpolation algorithm is expressed as: m is an integer and sequentially takes values from 0 to B / A; k represents different sampling time points; z ba (k) represents the bridge vertical displacement output by the bridge model at the k-th sampling time point; roll ba(k) represents the angle of rotation of the bridge around the x-axis output by the bridge model at the k-th sampling time point; This represents the result obtained through a linear interpolation algorithm at the 1st... The vertical displacement of the bridge at any given moment; This represents the result obtained through a linear interpolation algorithm at the 1st... B represents the bridge's rotation angle around the x-axis at time t; B represents the time step of the bridge model without interpolation; A represents the period of the bridge model's output bridge response data after interpolation.

[0097] In one embodiment described above, since the vibration mode of the shaking table is two-dimensional vibration, the bridge response data includes the vertical displacement of the bridge, and then the vertical displacement of the bridge generated by the bridge model is interpolated.

[0098] In this embodiment, since the shaking table vibrates in three dimensions, the bridge response data includes not only the bridge's vertical displacement but also its roll angle around the x-axis. When interpolating, it is necessary to interpolate both the bridge's vertical displacement and the roll angle around the x-axis. Since the principle is the same as in the embodiment described above, it will not be elaborated further here.

[0099] In one specific embodiment of the present invention, the vibration mode of the vibration table is three-dimensional vibration. The vibration table is composed of a vibration units connected in the x-axis direction. A set of bridge response data at the same moment specifically includes: the vertical displacement of the bridge at q different positions at that moment, and the rotation angle of the bridge around the x-axis at q different positions; q is a positive integer not less than 2; the univariate linear regression model includes a univariate linear regression sub-models.

[0100] Accordingly, step S103 may specifically include:

[0101] Upon receiving a set of bridge response data at the same time, the vertical displacements of the q bridges in this set are divided into corresponding univariate linear regression sub-models according to their location. Each univariate linear regression sub-model is then fitted to obtain a fitted univariate linear regression sub-models. The fitted j-th univariate linear regression sub-model is expressed as: Where, x tj This represents the center position of the j-th vibrating element, z. tj This represents the vertical displacement of the center position of the j-th vibrating element at that moment. and All of these are model parameters of the j-th univariate linear regression sub-model at that moment, determined by fitting, where j is a positive integer and 1≤j≤a.

[0102] The embodiment considers that, for the vibration table of three-dimensional vibration, the vibration table can be connected by a vibration units in the x-axis direction, that is, the bridge is simulated by a connected vibration table surface, that is, the bridge curve is replaced by a segmented line, and the specific posture of the a connected vibration table surfaces can be set respectively.

[0103] Since the a vibration units are arranged, the one-dimensional linear regression model can include a one-dimensional linear regression sub-model, that is, a straight line needs to be fitted to control the posture of the a vibration units respectively.

[0104] Taking a = 8 and p = 17 as an example, at a certain moment, the bridge response data output by the bridge model can include: z b1 , z b2 ,..., z b17 , roll1, roll2,..., roll 17 Here, z b1 is the vertical displacement of the bridge at the first position of the 17 different positions, and correspondingly, z b2 is the vertical displacement of the bridge at the second position of the 17 different positions, and z b17 is the vertical displacement of the bridge at the 17th position of the 17 different positions. roll1 is the rotation angle of the bridge around the x-axis at the first position of the 17 different positions, and correspondingly, roll2 is the rotation angle of the bridge around the x-axis at the second position of the 17 different positions, and roll 17 is the rotation angle of the bridge around the x-axis at the 17th position of the 17 different positions.

[0105] In the embodiment, according to the different positions, the q bridge vertical displacements in the group need to be divided into the corresponding one-dimensional linear regression sub-models. For example, in a specific case, the q different positions specifically include: the two end positions of each vibration unit in the a vibration units, and the center position of each vibration unit in the a vibration units. Taking the first vibration unit as an example, that is, in the example, the first position of the 17th different position corresponds to the head end of the first vibration unit, the third position of the 17th different position corresponds to the tail end of the first vibration unit, and the second position of the 17th different position corresponds to the center position of the first vibration unit.

[0106] Therefore, in the example, the first, second and third bridge vertical displacements of the q bridge vertical displacements in the group can be divided into the first one-dimensional linear regression sub-model, the third, fourth and fifth bridge vertical displacements can be divided into the second one-dimensional linear regression sub-model, and so on.

[0107] After dividing the vertical displacements of the q bridges in the group to the corresponding linear regression sub-models according to the positions, the fitting of each linear regression sub-model can be performed, and in the foregoing example, eight linear regression sub-models can be obtained after fitting, wherein the jth linear regression sub-model is represented as In the fitting, as described above, the value of the residual sum of squares Q is minimized, and the parameters in the linear regression sub-model can be obtained by derivation and

[0108] In one specific embodiment of the present application, the step S104 can specifically include:

[0109] Based on the bridge response data in the group and the corresponding fitted linear regression sub-models, the vibration table is sent with the attitude control instructions carrying z to1 to z toa , pitch to1 to pitch toa , and roll t1 to roll ta to adjust the current attitude of the vibration table.

[0110] wherein z to1 represents the vertical displacement of the center position of the first vibration unit superimposed with the track irregularity, z toa represents the vertical displacement of the center position of the ath vibration unit superimposed with the track irregularity, and z toj =z tj +z ir , and z irj is the track irregularity.

[0111] pitch to1 represents the rotation angle of the first vibration unit around the y-axis, pitch toa represents the rotation angle of the ath vibration unit around the y-axis, and

[0112] roll t1 represents the rotation angle of the first vibration unit around the x-axis, and roll ta represents the rotation angle of the ath vibration unit around the y-axis.

[0113] In this embodiment, for each vibration unit, the track irregularity z irj is superimposed to improve the accuracy of the test results, and the calculation principle of the track irregularity z irj is the same as above, which will not be repeated here.

[0114] In one specific embodiment of the present application, q=2a+1, and the q different positions specifically include: two end positions of each of the a vibration units, and a center position of each of the a vibration units.

[0115] Correspondingly, roll 2j-1 represents the rotation angle of the bridge around the x axis at the 2j-1 position; roll 2j+1 represents the rotation angle of the bridge around the x axis at the 2j+1 position; roll tj represents the rotation angle of the jth vibration unit around the x axis.

[0116] Taking j=1 as an example, the average of roll1 and roll3 can be obtained, that is, roll t1 , which is the angle between the table surface of the first vibration unit and the y axis, that is, the rotation angle of the table surface of the first vibration unit around the x axis. Such an embodiment calculates roll tj is relatively simple and convenient.

[0117] In addition, it can be seen that in this embodiment q=2a+1, and the q different positions specifically include: two end positions of each of the a vibration units, and a center position of each of the a vibration units, so when fitting each one-dimensional linear regression sub-model, as described in the above embodiment, only 3 points are needed to realize the fitting of 1 one-dimensional linear regression sub-model, making the fitting very simple and convenient.

[0118] Further, in one embodiment, for the fitting of any one-dimensional linear regression sub-model, only 2 points can be used to realize the fitting, thereby further reducing the calculation amount, for example, in the q different positions, only the two end positions of each of the a vibration units are selected, taking the above q=17 and a=8 as an example, only the two end positions of each of the 8 vibration units are selected, that is, only z b1 , z b3 , z b5 , z b7 , z b9 , z b11 , z b13 , z b15 , z b17 These 9 data can realize the fitting of 8 one-dimensional linear regression sub-models, that is, 2 fitting points are used to realize the fitting of a one-dimensional linear regression sub-model corresponding to a single vibration unit, and the calculation amount is very low. For example, in this example, z b1 and z b3 can be used to fit the first one-dimensional linear regression sub-model, z b3and z b5 The fitting of the second one-dimensional linear regression sub-model can be performed, and so on.

[0119] In Figure 3 In an embodiment of the application, eight vibration units are sequentially connected, wherein the serial number 31 corresponds to the first position of the 17 positions, and the vertical displacement of the bridge at this position is z b1 The serial number 32 corresponds to the fourth position of the 17 positions, and the vertical displacement of the bridge at this position is z b4 The serial number 33 represents the fourth vibration unit of the eight vibration units.

[0120] According to the technical scheme provided by the embodiment of the application, the test of the train passing through the bridge is realized based on the boundary coordination algorithm. Specifically, after receiving the bridge response data output by the bridge model, the attitude control instruction is obtained through the boundary coordination algorithm to adjust the current attitude of the vibration table. During the process of the train passing through the vibration table, the force at the specified position can be detected by the sensor, and then used as the input parameter of the bridge model, and the detection data can also be recorded. Since the test of the train passing through the bridge is realized based on the boundary coordination algorithm in the application, the bridge is simulated by the vibration table, which is beneficial to obtaining accurate and effective test results under the premise of ensuring the cost. During the process of the train passing through the vibration table, since the vibration table cannot directly simulate the deformation of the bridge in the form of displacement curve, the application configures a one-dimensional linear regression model for representing the vertical displacement of different bridge position points. After receiving a group of bridge response data at the same time, the fitting of the one-dimensional linear regression model can be performed based on the group of bridge response data, and the one-dimensional linear regression model after fitting is obtained, so that the attitude of the vibration table can be controlled based on the one-dimensional linear regression model after fitting. In summary, the scheme of the application can effectively perform the test of the train passing through the bridge, and obtain accurate and effective test results.

[0121] Corresponding to the above method embodiment, the embodiment of the application further provides a test system for a train passing through a bridge, which can be mutually corresponding and referred to above.

[0122] Referring to Figure 4 Fig. 1 is a structural schematic diagram of a test system for a train passing through a bridge in the application, which comprises:

[0123] The bridge response data receiving module 401 is configured to receive the bridge response data output by the bridge model.

[0124] The one-dimensional linear regression model configuration module 402 is configured to configure a one-dimensional linear regression model for representing the vertical displacement of different bridge position points.

[0125] The fitting module 403 is configured to perform fitting of the one-dimensional linear regression model based on the current group of bridge response data to obtain the fitted one-dimensional linear regression model each time the current group of bridge response data is received.

[0126] The attitude control module 404 is configured to send a corresponding attitude control instruction to the vibration table to adjust the current attitude of the vibration table based on the current group of bridge response data and the corresponding fitted one-dimensional linear regression model.

[0127] The feedback recording module 405 is configured to detect the force at the specified position through the sensor as the input parameter of the bridge model and record the detection data during the process in which the train passes through the vibration table.

[0128] In an embodiment of the present application, the vibration mode of the vibration table is two-dimensional vibration or three-dimensional vibration.

[0129] In an embodiment of the present application, the vibration mode of the vibration table is two-dimensional vibration, and the bridge response data includes the vertical displacement of the bridge. Correspondingly, the bridge response data receiving module 401 is specifically configured to:

[0130] According to the first period, receive the vertical displacement of the bridge output by the bridge model based on the linear interpolation algorithm;

[0131] wherein the linear interpolation algorithm is represented as: m is an integer and sequentially takes values from 0 to B / A; k represents different sampling time points; z ba (k) represents the vertical displacement of the bridge output by the bridge model at the kth sampling time point; represents the vertical displacement of the bridge at the kth sampling time point obtained through the linear interpolation algorithm; B represents the time step of the bridge model without interpolation; and A represents the period of the bridge response data output by the bridge model after interpolation.

[0132] In an embodiment of the present application, the vibration mode of the vibration table is two-dimensional vibration, and the current group of bridge response data is specifically the vertical displacement of the bridge at n different positions at the same time; n is a positive integer not less than 2.

[0133] Correspondingly, the fitting module 403 is specifically configured to:

[0134] Each time the current group of bridge response data is received, fitting of the one-dimensional linear regression model is performed based on the n vertical displacements of the bridge in the current group to obtain the fitted one-dimensional linear regression model, and the fitted one-dimensional linear regression model is represented as: wherein x t represents the center position of the vibration table, and z t ​This represents the vertical displacement of the center position of the shaking table at that moment. and All of these are model parameters of the univariate linear regression model at that moment, determined through fitting.

[0135] In one specific embodiment of the present invention, the attitude control module 404 is specifically used for:

[0136] Based on this set of bridge response data and the corresponding fitted univariate linear regression model, a signal carrying z-axis data was sent to the shaking table. to and pitch to The attitude control commands are used to adjust the current attitude of the vibration table;

[0137] Among them, z to This represents the vertical displacement of the center position of the shaking table, which is superimposed with track irregularities, and z to =z t +z ir , z ir For track irregularities, pitch to This represents the rotation angle of the vibration table around the y-axis, and

[0138]

[0139] In one specific embodiment of the present invention, the vibration mode of the vibration table is three-dimensional vibration, and the bridge response data includes the vertical displacement of the bridge and the rotation angle of the bridge about the x-axis. Accordingly, the bridge response data receiving module 401 is specifically used for:

[0140] According to the first cycle, the vertical displacement of the bridge and the rotation angle of the bridge around the x-axis are output by the bridge model based on the linear interpolation algorithm.

[0141] The linear interpolation algorithm is expressed as follows: m is an integer and takes values ​​from 0 to B / A; k represents different sampling time points; z ba (k) represents the vertical displacement of the bridge output by the bridge model at the k-th sampling time point; roll ba (k) represents the angle of rotation of the bridge around the x-axis output by the bridge model at the k-th sampling time point; This represents the result obtained through a linear interpolation algorithm at the 1st... The vertical displacement of the bridge at any given moment; This represents the result obtained through a linear interpolation algorithm at the 1st... B represents the bridge's rotation angle around the x-axis at time t; B represents the time step of the bridge model without interpolation; A represents the period of the bridge model's output bridge response data after interpolation.

[0142] In one specific embodiment of the present application, the vibration mode of the vibration table is three-dimensional vibration, the vibration table is connected by a vibration units in the x-axis direction, the bridge response data at the same time is specifically: the vertical displacement of the bridge at q different positions at the time, and the rotation angle of the bridge around the x-axis at the q different positions; q is a positive integer not less than 2; the one-dimensional linear regression model comprises a one-dimensional linear regression sub-model;

[0143] Correspondingly, the fitting module 403 is specifically configured to:

[0144] Each time a set of bridge response data at the same time is received, the q bridge vertical displacements in the set are divided into the corresponding one-dimensional linear regression sub-model according to the different positions, and the fitting of each one-dimensional linear regression sub-model is performed to obtain the a one-dimensional linear regression sub-models after fitting, and the jth one-dimensional linear regression sub-model after fitting is represented as: Wherein, x tj represents the center position of the jth vibration unit, z tj represents the vertical displacement of the center position of the jth vibration unit at the time, And are model parameters of the jth one-dimensional linear regression sub-model at the time determined by fitting, j is a positive integer and 1≤j≤a.

[0145] In one specific embodiment of the present application, the attitude control module 404 is specifically configured to:

[0146] Based on the set of bridge response data and the corresponding one-dimensional linear regression model after fitting, the attitude control module 404 sends the attitude control instruction carrying z to1 to z toa , pitch to1 to pitch toa , roll t1 to roll ta to the vibration table to adjust the current attitude of the vibration table.

[0147] Wherein, z to1 represents the vertical displacement of the center position of the first vibration unit superimposed with the track irregularity, z toa represents the vertical displacement of the center position of the a vibration unit superimposed with the track irregularity, and z toj =z tj +z ir , z irj is the track irregularity.

[0148] pitch to1 represents the rotation angle of the first vibration unit around the y-axis, pitch toa represents the rotation angle of the a vibration unit around the y-axis, and

[0149] roll t1 denotes the rotation angle of the first vibration unit around the x-axis, roll ta denotes the rotation angle of the a-th vibration unit around the y-axis.

[0150] In one specific embodiment of the present application, q = 2a + 1, and the q different positions specifically include: two end positions of each of the a vibration units, and a center position of each of the a vibration units;

[0151] Correspondingly, roll 2j-1 denotes the rotation angle of the bridge around the x-axis at the 2j-1-th position, roll 2j+1 denotes the rotation angle of the bridge around the x-axis at the 2j+1-th position, roll tj denotes the rotation angle of the j-th vibration unit around the x-axis.

[0152] It is also noted that the aforementioned terms, such as first and second, are used only to distinguish one entity or operation from another, and do not necessarily require or imply these entities or operations to be in any physical or logical order. Furthermore, the terms "comprise", "comprising", or any other variant thereof are intended to cover non-exclusive inclusions, such that processes, methods, articles, or apparatuses that comprise a list of elements are not required to only include those elements, but can include other elements not expressly listed or inherent to such processes, methods, articles, or apparatuses. Without further limitation, an element preceded by "comprises a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element.

[0153] Those skilled in the art will further appreciate that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be embodied in electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various examples were described above generally in terms of their functionality, without referring to the corresponding acts in a computing device. Whether such functionality is implemented in hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled artisans can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0154] The principles and implementation manners of the present application are described by using specific examples in the present application, and the above examples are only used to help understand the technical solutions of the present application and the core ideas thereof. It should be noted that, for those skilled in the art, without departing from the principles of the present application, the present application can be improved and modified in several ways, and these improvements and modifications also fall within the protection scope of the present application.

Claims

1. A method of testing a train passing over a bridge, characterized in that, The method comprises the following steps: receiving bridge response data output by a bridge model; configuring a linear regression model for representing vertical displacement of different bridge position points; fitting the linear regression model based on a group of the bridge response data received at the same time to obtain a fitted linear regression model; sending corresponding attitude control instructions to the vibration table to adjust the current attitude of the vibration table based on the group of the bridge response data and the corresponding fitted linear regression model; detecting force at a specified position by a sensor during the process of the train passing through the vibration table and taking the detected data as input parameters of the bridge model.

2. The method of claim 1, wherein, The vibration mode of the vibration table is two-dimensional vibration or three-dimensional vibration.

3. The method of claim 2, wherein, The vibration mode of the vibration table is two-dimensional vibration, and the bridge response data comprises bridge vertical displacement, and correspondingly, the bridge response data received by the bridge model comprises: receiving bridge vertical displacement output by the bridge model based on a linear interpolation algorithm according to a first period; The linear interpolation algorithm is expressed as follows: m is an integer and takes values ​​from 0 to B / A; k represents different sampling time points; This represents the vertical displacement of the bridge output by the bridge model at the k-th sampling time point; This represents the result obtained through the linear interpolation algorithm at the 1st... B represents the vertical displacement of the bridge at time t; B represents the time step of the bridge model without interpolation; A represents the period of the bridge response data output by the bridge model after interpolation.

4. The method of claim 2, wherein, The vibration mode of the vibration table is two-dimensional vibration, and the group of the bridge response data at the same time specifically comprises bridge vertical displacement at n different positions at the time; n is a positive integer not less than 2. Correspondingly, the fitting of the linear regression model based on the group of the bridge response data received at the same time to obtain a fitted linear regression model comprises: Whenever a group of the bridge response data at the same time is received, based on the n bridge vertical displacements of the group, fitting of the one-dimensional linear regression model is performed to obtain a fitted one-dimensional linear regression model, and the fitted one-dimensional linear regression model is expressed as: ; wherein, denotes the center position of the vibration table, denotes the vertical displacement of the center position of the vibration table at the time point, and are model parameters of a linear regression model of the first order determined by fitting at the time point.

5. The method of claim 4, wherein, sending corresponding attitude control instructions to the vibration table to adjust the current attitude of the vibration table based on the group of the bridge response data and the corresponding fitted linear regression model comprises: Based on the bridge response data and the corresponding fitted one-dimensional linear regression model described in this group, the attitude control instruction carrying and is sent to the vibration table to adjust the current attitude of the vibration table; wherein, represents vertical displacement of the center position of the vibration table to which the track irregularity amount is superimposed, and , is the track irregularity amount, represents the rotation angle of the vibration table around the y-axis, and .

6. The method of claim 2, wherein, The vibration mode of the vibration table is three-dimensional vibration, and the bridge response data comprises bridge vertical displacement and bridge rotation angle around the x-axis, and correspondingly, the bridge response data received by the bridge model comprises: receiving bridge vertical displacement and bridge rotation angle around the x-axis output by the bridge model based on a linear interpolation algorithm according to a first period; The linear interpolation algorithm is expressed as follows: m is an integer and takes values ​​from 0 to B / A; k represents different sampling time points; This represents the vertical displacement of the bridge output by the bridge model at the k-th sampling time point; This represents the angle of rotation of the bridge around the x-axis output by the bridge model at the k-th sampling time point; This represents the result obtained through the linear interpolation algorithm at the 1st... The vertical displacement of the bridge at any given moment; This represents the result obtained through the linear interpolation algorithm at the 1st... B represents the angle of rotation of the bridge around the x-axis at time t; B represents the time step of the bridge model without interpolation; A represents the period of the bridge response data output by the bridge model after interpolation.

7. The method of claim 2, wherein, The vibration mode of the vibration table is three-dimensional vibration, and the vibration table is connected by a vibration unit in the x-axis direction, and the group of the bridge response data at the same time specifically comprises bridge vertical displacement at q different positions at the time and bridge rotation angle around the x-axis at the q different positions; q is a positive integer not less than 2; the linear regression model comprises a linear regression sub-model; Correspondingly, the fitting of the linear regression model based on the group of the bridge response data received at the same time to obtain a fitted linear regression model comprises: Whenever a set of bridge response data at the same time is received, the q bridge vertical displacements in the set are divided into corresponding one-dimensional linear regression sub-models according to the different positions, and fitting of each one-dimensional linear regression sub-model is performed to obtain a fitted a one-dimensional linear regression sub-models, and the jth one-dimensional linear regression sub-model after fitting is represented as: ; wherein, represents the center position of the jth vibration unit, represents the vertical displacement of the center position of the jth vibration unit at the moment, and are model parameters of the jth one-dimensional linear regression sub-model at the moment determined by fitting, j is a positive integer and 1≤j≤a.

8. The method of claim 7, wherein, sending corresponding attitude control instructions to the vibration table to adjust the current attitude of the vibration table based on the group of the bridge response data and the corresponding fitted linear regression model comprises: Based on the bridge response data and the corresponding fitted one-dimensional linear regression model described in this group, the attitude control instruction carrying To , To , To the attitude of the current vibration table; wherein represents a vertical displacement of the center position of the a-th vibration unit superimposed with the track irregularity amount, represents a vertical displacement of the center position of the a-th vibration unit superimposed with the track irregularity amount, and , is a track irregularity amount of the j-th vibration unit, is a vertical displacement of the center position of the j-th vibration unit at the time point, is a vertical displacement of the center position of the j-th vibration unit superimposed with the track irregularity amount; represents a rotation angle of the first vibration unit around the y-axis, represents a rotation angle of the a-th vibration unit around the y-axis, and ; This represents the rotation angle of the first vibrating element around the x-axis. This represents the rotation angle of the a-th vibration element around the y-axis.

9. The method of claim 8, wherein, q=2a+1, and the q different positions specifically comprise two end positions of each vibration unit in the a vibration units and a center position of each vibration unit in the a vibration units. Correspondingly, , denotes the rotation angle of the bridge around the x-axis at the 2j-1th position; denotes the rotation angle of the bridge around the x-axis at the 2j+1th position, denotes the rotation angle of the jth vibration unit around the x-axis.

10. A system for testing a train passing over a bridge, characterized by The method comprises the following steps: The bridge response data receiving module receives bridge response data output by the bridge model; The linear regression model configuration module is configured to configure a linear regression model for representing vertical displacement of different bridge position points; The fitting module is configured to fit the linear regression model based on a group of the bridge response data at the same time to obtain a fitted linear regression model; The attitude control module is configured to send a corresponding attitude control instruction to the vibration table to adjust the current attitude of the vibration table based on the group of the bridge response data and the corresponding fitted linear regression model; The feedback recording module is configured to detect force at a specified position by a sensor as an input parameter of the bridge model and record the detection data during the train passing through the vibration table.

Images