A method for accuracy verification of a windshield test bench, the test bench and computer equipment

Abstract

This application discloses a method for verifying the accuracy of a windshield test bench, the test bench itself, and a computer device, relating to the field of railway vehicle simulation testing technology. The method includes: determining the theoretical spatial distance between at least four corresponding feature vertices on the windshield mounting plates at the ends of the left and right motion platforms under a target working condition; driving the windshield test bench to simulate the target working condition, causing it to move to a target pose; after reaching the target pose, measuring the actual spatial distance between the corresponding feature vertices; comparing the actual spatial distance with the theoretical spatial distance; and determining the positioning accuracy of the windshield test bench based on the comparison result. This application can achieve closed-loop self-checking of the positioning accuracy of a dual-end independently driven windshield test bench by measuring the actual spatial distance between feature vertices and comparing it with the theoretical spatial distance without relying on individual measurements of each motion axis, thereby improving verification efficiency and reliability.

Description

Technical Field

[0001] This application relates to the field of railway vehicle simulation testing technology, and in particular to a method for accuracy verification of a windshield test bench, the test bench, and computer equipment. Background Technology

[0002] The windshield is a critical component at the connection between adjacent carriages in a train, and its performance directly affects train operation safety and passenger comfort. When trains pass through complex tracks such as curves, reverse curves, superelevation, and inclines, multi-directional coupled relative motion occurs between adjacent carriages, causing the windshield to be subjected to complex alternating loads such as tension, compression, shear, and bending. This is the primary cause of windshield fatigue failure. Therefore, before the windshields leave the factory, fatigue performance tests must be conducted on the windshields using a windshield test bench to simulate carriage posture changes under actual track conditions.

[0003] Currently, some windshield test benches employ a six-degree-of-freedom single-end drive structure, where one end is fixed and the other simulates the movement of a carriage, only capable of replicating the relative posture of adjacent carriages. However, in actual operation, both carriages possess independent six-degree-of-freedom motion, making it difficult for the single-end drive method to accurately reproduce the stress and deformation state of the windshield under independent motion at both ends. To address this, existing research has proposed a twelve-degree-of-freedom windshield test bench, with a six-degree-of-freedom motion platform on each side, which can be independently controlled, thus more accurately simulating the absolute posture of two adjacent carriages.

[0004] For multi-degree-of-freedom test benches like these, accuracy verification is a crucial step in ensuring the reliability of test results. Existing accuracy verification methods typically employ the following two approaches: Firstly, the actual displacement or rotation angle of each motion axis (such as the X-axis, Y-axis, Z-axis, roll axis, pitch axis, and yaw axis) is measured and compared with the target command value. While this method can reflect single-axis drive errors, it cannot reflect the end-effector composite pose error caused by multi-axis coupling. Since the windshield is installed at the end of two motion platforms, its actual force and deformation depend on the relative spatial pose between the two windshield mounting plates. The simple superposition of single-axis errors often lacks a clear mapping relationship with the end-effector composite error, which can easily cause some coupling errors (such as the superposition of multiple small-angle rotations causing a large offset of the corner point of the windshield mounting plate) to be missed.

[0005] Secondly, the spatial coordinates of several feature vertices on the end effector (windshield mounting plate) are directly measured using equipment such as laser trackers, and then compared with theoretical coordinates. Although this method can directly obtain the end effector pose, the measurement process is complex, the equipment is expensive, and it is time-consuming, making it difficult to widely apply in rapid and repeatable verification on the production site.

[0006] In summary, existing technologies lack a precision verification method suitable for a twelve-DOF windshield test bench that can directly, quickly, and comprehensively reflect the spatial pose error of the end-mounted windshield plate. How to efficiently determine whether the test bench accurately reproduces the absolute pose of the vehicle under the target working condition, without having to measure each motion axis individually, has become a pressing technical problem in this field. Summary of the Invention

[0007] The purpose of this application is to provide a method, test bench, and computer equipment for the accuracy verification of a windshield test bench, which solves the technical problem that existing windshield test benches cannot directly, quickly, and comprehensively reflect the spatial pose error of the end windshield mounting plate in accuracy verification. Thus, without relying on individual measurement of each motion axis, it can achieve closed-loop self-check of the positioning accuracy of the dual-end independently driven windshield test bench by measuring the actual spatial distance between feature vertices and comparing it with the theoretical spatial distance, thereby improving verification efficiency and reliability.

[0008] To achieve the above objectives, this application provides the following solution: Firstly, this application provides a method for verifying the accuracy of a windshield test bench, the windshield test bench having two motion platforms, left and right, including: Determine the theoretical spatial distance between at least four corresponding feature vertices on the end windshield mounting plates of the left and right motion platforms under the target working condition; The windshield test bench is driven to simulate the target working condition, and moves to the target pose. After reaching the target pose, the actual spatial distance between the corresponding feature vertices is measured. The actual spatial distance is compared with the theoretical spatial distance, and the positioning accuracy of the windshield test platform is determined based on the comparison result.

[0009] Secondly, this application provides a windshield test bench, comprising: Left six-degree-of-freedom motion platform and right six-degree-of-freedom motion platform; Windshield mounting plates are installed at the ends of the left and right motion platforms. Each windshield mounting plate is provided with at least four feature vertices for measuring the distance. and a control unit, the control unit comprising: The acquisition module is used to determine the theoretical spatial distance between all corresponding feature vertices under the target working condition; The drive module is used to drive the motion platform to the target pose; The measurement module is used to obtain the actual spatial distance between the corresponding feature vertices; The determination module is used to compare the actual spatial distance with the theoretical spatial distance and output the positioning accuracy determination result based on the comparison result.

[0010] Thirdly, this application provides a computer device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the accuracy verification method for the windshield test bench described in any one of the above-mentioned methods.

[0011] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method for verifying the accuracy of a windshield test bench, the test bench itself, and computer equipment. By determining the theoretical spatial distance between at least four corresponding feature vertices on the windshield mounting plates at the ends of the left and right motion platforms under a target working condition, a standardized theoretical benchmark is provided for determining the positioning accuracy. Since the vehicle body posture varies significantly under different vehicle models and different track conditions, without a unified theoretical spatial distance as a reference, it is impossible to quantitatively determine whether the actual movement meets the standard. Therefore, this step solves the problem of the lack of a unified and quantitative positioning accuracy benchmark in the prior art, realizing standardized pre-calculation of the theoretical spatial distance under different vehicle models and different track conditions, and providing a reliable theoretical basis for subsequent accuracy determination.

[0012] By simulating the target working condition using a windshield test bench and measuring the actual spatial distance between the corresponding feature vertices, the actual relative position of the end-mounted windshield plate in space can be directly obtained. Existing single-axis measurement methods can only reflect the displacement or rotational deviation of each drive axis itself, failing to capture the comprehensive end-mounted pose error caused by coupling during multi-axis linkage. Directly measuring the distance between feature vertices, however, can capture the comprehensive influence of all degrees of freedom on the relative pose of the windshield plate in one go. Therefore, this step solves the problem that existing single-axis measurement methods cannot reflect the comprehensive end-mounted pose error caused by multi-degree-of-freedom coupling, achieving direct, rapid, and low-cost measurement of the windshield plate's spatial pose. It avoids the cumulative error omissions caused by axis-by-axis measurement and overcomes the drawbacks of high cost and complex operation of laser trackers for measuring absolute coordinates.

[0013] By comparing the actual spatial distance with the theoretical spatial distance and determining the positioning accuracy based on the comparison results, a quantitative mapping relationship from measured data to accuracy conclusions was established. Since the determination results can directly guide subsequent error compensation or mechanical adjustments, each run of the windshield test bench can be automatically verified and corrected in a closed loop. Therefore, this step solves the problem that existing accuracy verification methods cannot automatically and quantitatively determine whether the end-effector pose meets the standards, realizing closed-loop self-checking of the positioning accuracy of the windshield test bench. This provides a clear basis for subsequent error compensation or mechanical adjustments, significantly improving the reproducibility and reliability of windshield testing.

[0014] In summary, this application forms a complete and self-consistent accuracy verification scheme for the windshield test bench through the coordinated action of three steps: theoretical spatial distance acquisition, actual spatial distance measurement, and comparative judgment. This scheme avoids the defects of missing coupling errors in single-axis measurement and does not require separate measurement of each motion axis, thus achieving efficient, direct, and reliable determination of positioning accuracy. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the structure of a windshield test bench according to one embodiment of this application; Figure 2 This is a schematic diagram of the hardware architecture of a control system provided in an embodiment of this application; Figure 3 A flowchart illustrating a method for verifying the accuracy of a windshield test bench according to an embodiment of this application; Figure 4 A schematic diagram of the global coordinate system of the left and right test platforms provided in another embodiment of this application; Figure 5 This is a schematic diagram of the train's position at a curved entrance, provided in an embodiment of this application. Figure 6 This is a partial enlarged view of the curve entry point provided in an embodiment of this application; Figure 7 A schematic diagram of the geometric relationship of four points at the curve entry point provided in an embodiment of this application; Figure 8 This is a schematic diagram of the curve apex pose provided in an embodiment of this application; Figure 9 This is a schematic diagram of the geometric relationship between four points at the apex of a curve provided in an embodiment of this application; Figure 10 This is a schematic diagram of the train's reverse curve pose provided in an embodiment of this application; Figure 11 A geometric relationship diagram of four points of a reverse curve provided in an embodiment of this application; Figure 12 This is a schematic diagram of the position of a roll servo electric cylinder provided in one embodiment of this application.

[0017] Explanation of reference numerals in the attached figures: 1-Base assembly; 2-X-axis motion platform; 3-Y-axis motion platform; 4-Roll motion platform; 5-Pitch motion platform; 6-Yaw motion platform; 7-Vertical motion platform; 8-Windshield mounting plate; 9-Windshield to be tested. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be described in detail below with reference to the accompanying drawings and specific embodiments. The accuracy verification method, test bench, and control system of the windshield test bench provided in this application are particularly suitable for a twelve-degree-of-freedom dual-end independently driven windshield test bench, but can also be used for accuracy verification of other multi-degree-of-freedom test benches.

[0020] See Figure 1 The windshield test bench provided in this embodiment includes a left six-degree-of-freedom motion platform (hereinafter referred to as the left test bench) and a right six-degree-of-freedom motion platform (hereinafter referred to as the right test bench). The two motion platforms are arranged symmetrically in the center. Each motion platform, from bottom to top, consists of: a base assembly 1, an X-axis motion platform 2, a Y-axis motion platform 3, a roll motion platform 4 (around the X-axis), a pitch motion platform 5 (around the Y-axis), a yaw motion platform 6 (around the Z-axis), and a vertical motion platform 7 (Z-axis). The driving methods of each motion platform are as follows: the X-axis motion platform 2, the Y-axis motion platform 3, and the vertical motion platform 7 are driven by servo motors in conjunction with ball screws; the roll motion platform 4, the pitch motion platform 5, and the yaw motion platform 6 are driven by servo electric cylinders.

[0021] At the end of each motion platform (i.e., the front end of the vertical motion platform 7), a windshield mounting plate 8 is fixedly installed for fixing the windshield 9 to be tested. Preferably, the windshield mounting plate 8 is a rectangular rigid plate, with its four corner points as feature vertices; alternatively, at least four other feature vertices besides the corner points can be selected. For ease of description, this embodiment uses four corner points as an example.

[0022] See Figure 2 , Figure 2This is a schematic diagram of the hardware architecture of a control system provided in one embodiment of this application. The control system includes a control unit and a measuring device. The control unit consists of an industrial computer (host computer) and a motion controller (slave computer), which are connected via an industrial Ethernet network (such as the Modbus TCP protocol). The motion controller communicates with multiple servo drives via an EtherCAT bus. The servo drives drive corresponding servo motors or servo cylinders to achieve motion control of each axis of each motion platform.

[0023] The measuring device includes at least four rope displacement sensors (preferably four), which are respectively installed between corresponding feature vertices of the left and right windshield mounting plates 8. These sensors are used to measure the actual spatial distance between the feature vertices in real time and feed the measured values ​​back to the control unit. As a calibration method with higher precision, the measuring device can also employ a laser tracker.

[0024] The industrial control computer has a built-in or connected memory that stores a database of theoretical spatial distances for different train models under different operating conditions, or a calculation program for the absolute pose model. The industrial control computer executes the accuracy verification method provided in the embodiments of this application, obtains the theoretical spatial distance, drives the test bench to move to the target pose through the motion controller, collects the actual spatial distance data of the measuring device, compares and judges the data, and outputs the positioning accuracy result.

[0025] In practical applications, an industrial control computer can be an industrial control computer, an embedded controller, a programmable logic controller, or other devices with computing and control functions. A motion controller can be a standalone motion control card or a software motion controller integrated into the industrial control computer. Servo drives, servo motors, and draw-wire displacement sensors are all commercially available industrial automation products.

[0026] In addition, in some embodiments, the acquisition of theoretical spatial distance (such as the calculation of absolute pose model) can also be completed by a separate computing server, and the industrial control computer obtains the results from the server through the network. However, considering the real-time requirements of the test bench, local calculation or pre-storage is usually adopted.

[0027] In one exemplary embodiment, such as Figure 3 The flowchart shown illustrates a method for verifying the accuracy of a windshield test bench. This method can be executed independently by an industrial control computer (ICC) or collaboratively by an ICC and a motion controller. This embodiment applies this method to... Figure 2 Taking the control unit of the windshield test bench shown as an example, the control unit includes an industrial computer and a motion controller. The industrial computer acts as the host computer, responsible for acquiring the theoretical spatial distance, receiving and comparing measurement data; the motion controller acts as the slave computer, responsible for driving the left and right motion platforms to the target pose. Specifically, it includes the following steps: Step 10: Determine the theoretical spatial distance between at least four corresponding feature vertices on the end windshield mounting plates 8 of the left and right motion platforms under the target working condition; Step 20: Drive the windshield test bench to simulate the target working condition and move it to the target pose. After reaching the target pose, measure the actual spatial distance between the corresponding feature vertices. Step 30: Compare the actual spatial distance with the theoretical spatial distance, and determine the positioning accuracy of the windshield test platform based on the comparison result.

[0028] Implementing steps 10 to 30 yielded the following beneficial effects: Step 10, by obtaining the theoretical spatial distance, provided a standardized theoretical benchmark for determining positioning accuracy, solving the problem of the lack of a unified and quantitative benchmark in existing technologies; Step 20, by measuring the actual spatial distance, directly obtained the actual relative positional relationship of the end-mounted windshield mounting plate 8, solving the problem that single-axis measurement cannot reflect the comprehensive pose error caused by multi-degree-of-freedom coupling, and realizing low-cost, fast, and direct end-mounted pose measurement; Step 30, by comparing and judging the actual spatial distance with the theoretical spatial distance, established a quantitative mapping from measured data to accuracy conclusions, realizing closed-loop self-checking of positioning accuracy, and providing a clear basis for error compensation or mechanical adjustment. The synergistic effect of these three steps forms a highly efficient, low-cost, and highly reliable windshield test bench accuracy verification scheme.

[0029] Furthermore, the method provided in this application is not limited to a twelve-DOF windshield test bench, but can also be applied to the end-effector accuracy verification of other multi-DOF motion platforms (such as a six-DOF Stewart platform, a nine-DOF hybrid robot, a spacecraft docking simulator, etc.). The feature vertices are not limited to the physical corner points of the windshield mounting plate 8, but can also be additionally set target points or reflective patches. The number of corresponding feature vertices is preferably four corner points, but theoretically, at least four non-collinear points can uniquely determine the relative pose. The acquisition of theoretical spatial distance is not limited to real-time calculation based on the absolute pose model, but can also adopt an offline calibration method (the standard distance is calibrated in advance by high-precision equipment and stored in a database, and directly retrieved from the table during the test), which is suitable for batch production line testing. The measurement of actual spatial distance is not limited to a rope displacement sensor or a laser tracker, but can also adopt binocular vision, structured light camera or ultrasonic ranging sensor. The measurement trigger can be automatic triggering or manual triggering. When the positioning accuracy is determined to be unqualified, in addition to automatically compensating motion parameters or prompting mechanical adjustment, machine learning can also be combined to perform trend analysis on multiple verification data to predict mechanical wear or zero drift and achieve predictive maintenance.

[0030] In a preferred embodiment of this application, a drawstring displacement sensor is used to measure the actual spatial distance. Compared to expensive equipment such as laser trackers, the drawstring displacement sensor is low-cost, easy to install, and can output data in real time without complex coordinate transformation. Therefore, while achieving the accuracy verification method of this application, it further reduces equipment costs and operational complexity, making it particularly suitable for rapid and repeatable verification on production lines.

[0031] The following detailed description of the implementation process of the accuracy verification method for the windshield test bench, with reference to specific embodiments, illustrates this process.

[0032] Step 10: Determine the theoretical spatial distance between at least four corresponding feature vertices on the end windshield mounting plates 8 of the left and right motion platforms under the target working condition.

[0033] Optionally, the control unit determines the theoretical spatial distance between at least four corresponding feature vertices on the windshield mounting plate 8 at the end of the left and right motion platforms under the target working condition. This embodiment provides two acquisition methods: (1) lookup table method: read the pre-stored theoretical spatial distance from the local database; (2) real-time calculation method: calculate based on the absolute pose model of the train and the geometric dimensions of the windshield mounting plate 8. The implementation process of the real-time calculation method (steps 101-104) is described in detail below.

[0034] In one embodiment, steps 101-104 specifically include: Step 101: Define basic parameters First, determine the basic parameters of the train and the track parameters, as detailed in Tables 1 and 2. These parameters are the basis for establishing the absolute pose model.

[0035] Table 1 Basic Parameters of the Train

[0036] Table 2 Track Parameters

[0037] It should be noted that the above train and track parameters are all known fixed values ​​or design values. Regarding the spacing between adjacent carriages... The value is a standard value (e.g., 800mm) on straight sections, and can be calculated in real time on curved sections based on the relationship between the train's travel distance and the track geometry. Specifically, when a train travels at speed v through a curve with radius R, the relative turning angle between adjacent carriages will affect the distance between them. With slight variations, this application uses a precise geometric model (as shown in the formula below) to calculate this variation, or the distance between adjacent carriages is directly input by the test personnel according to the standard operating condition manual. Those skilled in the art will understand that the acquisition of the theoretical spatial distance does not depend on real-time measurement, but is calculated based on a pre-established absolute pose model and geometric dimensions. Therefore, the method of this application can be repeatedly performed without external calibration equipment.

[0038] Meanwhile, the auxiliary angles are defined in Table 3: Table 3 Auxiliary Angles

[0039] Under curved conditions, the spacing between adjacent carriages It is not a constant value. In this application, ,in The standard spacing for straight line segments (e.g., 800mm). The formula for calculating the minute expansion and contraction caused by the coupler deflection when passing through a curve is as follows: , The angle between the line connecting the coupler and the horizontal line is shown in Table 3.

[0040] Step 102: Establish the test bench coordinate system and feature vertices like Figure 4 As shown, establish the global coordinate system for the left and right test platforms: Left test bench coordinate system , , , Coordinate system of the right test rig , , , It is centrally symmetrical. and opposite directions and opposite directions and They are in the same direction.

[0041] Definition of positive directions of each axis: X-axis: When viewed from the front of the test bench, the left side of the test bench is positive to the left, and the right side of the test bench is positive to the right; Y-axis: When viewed from the front of the test bench, the left side of the test bench facing forward (away from the observer) is positive, and the right side of the test bench facing backward (closer to the observer) is positive; Z-axis: positive upwards, negative downwards; Roll angle A: Around the X-axis, rightward roll is positive. Pitch angle B: around the Y-axis, with pitch being positive. Yaw angle C: Counterclockwise is positive At least four non-collinear feature vertices are selected on the left and right windshield mounting plates 8. This application preferably uses four corner points, denoted as: corner points A, B, C, and D of the left windshield mounting plate 8, and corresponding corner points E, F, G, and H of the right windshield mounting plate 8 (e.g., ...). Figure 1 As shown (corner points C and G are not displayed). The spatial distance between corresponding points is denoted as... , , , .

[0042] In addition, this application defines the equipment origin, the windshield origin, and the windshield mounting point. The equipment origin is the absolute reference of the global coordinate system of the left and right test benches and has a power-off memory function; the windshield origin is the starting reference position for operation and has X-axis, Y-axis, and Z-axis offsets from the equipment origin; the windshield mounting point is the fixed interface for the windshield and can be adjusted according to different windshield types.

[0043] Step 103: Establish an absolute pose model for typical operating conditions. Based on the target working conditions (such as curve entrance, curve exit, curve apex, curve apex, reverse curve, outer rail superelevation, climbing, etc.), and based on the train's basic parameters, track parameters, and the geometric positional relationship between the carriage and the track, an absolute pose model of the left and right test rigs is established (this model can be an analytical formula based on classical mechanics and geometric relationships, or a numerical model calibrated through finite element simulation or experiments; this application does not limit this). That is, the six-degree-of-freedom parameters of the left and right test rigs in the global coordinate system are solved. , , , , , The following example illustrates the modeling process using the curve inlet condition (other conditions are similar): The train travels from left to right; the left test platform is stationary, while the right test platform simulates the movement of the preceding carriage. (Based on...) Figure 5 , Figure 6 The geometric relationships shown are as follows: Calculate the angle between the line connecting the axles of two adjacent carriages and the horizontal line. : Right test stand yaw angle : Among them, yaw angle Negative values ​​indicate right skewness; Y-axis displacement of the right test bench : in, This refers to the axle spacing between the two carriages after deflection. ; Angle between the line connecting the coupler and the horizontal line : X-axis displacement of the right test bench : Left test bench at rest: , , All other degrees of freedom are 0.

[0044] The derivation of other working conditions (curve exit, arc apex, arc apex, reverse curve, outer rail superelevation, and climbing) is similar to that of curve entry. For specific formulas, please refer to steps 1041-1043 and application examples below, which will not be repeated here.

[0045] Step 104: Calculate the theoretical spatial distance After obtaining the absolute poses of the left and right test stands, combine them with the geometric dimensions (width) of the windshield mounting plate 8. ,length (1) Calculate the coordinates of the feature vertices in the 8-coordinate system of the windshield mounting plate, and calculate the theoretical spatial distance between the corresponding feature vertices.

[0046] like Figure 7 As shown, taking the curved entrance condition as an example, assuming the left windshield mounting plate 8 is fixed and the right windshield mounting plate 8 is displaced ( , ) and yaw angle Of the four corner points, the corner points A, B, C, and D of the left windshield mounting plate correspond to the corner points E, F, G, and H of the right windshield mounting plate, while I and J are the midpoints of the end face of the carriage.

[0047] for (Distance from corner point A of the left windshield mounting plate to corner point E of the right windshield mounting plate): First, calculate the auxiliary length, auxiliary angle, and distance: Similarly, for (Distance from corner point B of the left windshield mounting plate to corner point F of the right windshield mounting plate): Due to symmetry, , .

[0048] For linear tension and compression conditions, only the spacing between adjacent carriages needs to be changed. The value of is taken, and all other pose parameters are zero. The four-point spacing is directly determined by . and mounting plate width calculate: .

[0049] In actual use, the operator selects the target working condition from the human-machine interface, and the system automatically calls the corresponding formula to calculate the theoretical spatial distance (real-time calculation); for standard models and common working conditions, the pre-stored values ​​can also be read directly by looking up the table.

[0050] Furthermore, the absolute pose model can be used to establish analytical formulas for various working conditions, including curve entry, curve exit, curve apex, curve apex, reverse curve, superelevation of outer rail, and climbing. For example, the analytical formula for the curve entry condition has been given in steps 103-104 above; the curve exit condition can be directly derived from symmetry; similar formulas can be established for curve apex, apex, and reverse curve conditions based on corresponding geometric diagrams, as detailed in steps 1041-1043; the turning angle formulas for superelevation of outer rail and climbing conditions have also been exemplified in step 103. Those skilled in the art can obtain the analytical formulas for the remaining working conditions without creative effort based on the above examples.

[0051] In one embodiment, steps 1041-1043 specifically include: Step 1041, Curve Crest Working Condition according to Figure 8 , Figure 9 The geometric diagram shown is used to calculate the intermediate variables. , , ): Yaw angle of left test stand : X-axis displacement of the left test bench : Y-axis displacement of the left test bench : Right test stand position: , , .

[0052] Four-point spacing (assuming the width of the windshield mounting plate is 8) ): Step 1042, Curved Bottom Working Condition The base and apex of the arc are symmetrical about the X-axis, therefore, it is only necessary to change the formula for the apex. , , , Invert: Four-point spacing exchange: , , , .

[0053] Step 1043, Reverse Curve (S-shaped) Working Condition according to Figure 10 , Figure 11 The geometric diagram shown has two opposing curves with the same radius, and the length of the straight line segment in the middle is... Let K be the intersection of the line connecting the midpoints of the two car ends, the line connecting the couplers, and the straight segment, with K being their midpoint. Let g be the distance between the lines connecting the midpoints of the car ends. and This is an intermediate variable. Calculate: Yaw angle (equal on both left and right test benches): Y-axis displacement (equal displacement on left and right test benches): in, , , , ; X-axis displacement (equal displacement on left and right test benches): Four-point spacing (left and right symmetrical) , ): Furthermore, the feature vertices are the four corner points of the windshield mounting plate 8, and the theoretical spatial distance is four values. , , , For symmetrical working conditions (such as curve inlet and outlet), where some distances are equal, calculations can be simplified.

[0054] Step 20: Drive the windshield test bench to simulate the target working condition and move it to the target pose. After reaching the target pose, measure the actual spatial distance between the corresponding feature vertices.

[0055] Optionally, the control unit converts the absolute pose of the left and right test platforms under the target working condition into position commands for each servo axis (linear axes directly use displacement values, while rotary axes are first converted into electric cylinder displacements), and sends this to the motion controller via the Modbus TCP protocol. The motion controller, based on axis group control, synchronously drives 12 servo axes via the EtherCAT bus, enabling the left and right motion platforms to move from the current pose to the target pose. The motion controller monitors the actual position and command deviation of each axis in real time, and determines the target position when all axis deviations are less than a set threshold.

[0056] Once in position, the control unit automatically triggers the pull-cord displacement sensors to measure the spatial distance between corresponding feature vertices on the left and right windshield mounting plates 8. Each pull-cord displacement sensor outputs the pull-cord length, which, after analog-to-digital conversion and filtering, yields the actual spatial distance. , , , (If four corner points are used). The measurement data is uploaded to the industrial control computer for use in step 30. If the motion times out or the measurement fails, the system will issue an alarm and stop the test.

[0057] Furthermore, a pull-string displacement sensor is used to measure the actual spatial distance. Specifically, the fixed ends of the pull-string displacement sensor are installed at the four corner points of the left windshield mounting plate 8, and the free ends of the pull strings are fixed to the corresponding corner points of the right windshield mounting plate 8. The pull-string displacement sensor outputs the extension length of the pull string, which is the spatial distance between the corresponding feature vertices. The pull-string displacement sensor has a range of 1500mm and an accuracy of ±0.1mm, which meets the test requirements. As an alternative, a laser tracker can be used to measure the spatial coordinates of each feature vertex sequentially and then calculate the distance between the two points, but this is more expensive and suitable for calibration applications.

[0058] In a specific example, the above driving and measurement process can be further broken down into the following sub-steps (steps 201 to 203): Step 201: The control unit issues motion commands. The control unit (industrial computer + motion controller) determines the theoretical pose of the left and right test benches under the target working condition obtained in step 1. , , , , , ), generating motion commands for each servo axis.

[0059] Step 2011: Generation of motion commands A1) Linear motion axes (X, Y, Z): Target displacement , , It can be used directly as a position command.

[0060] The motion speed and acceleration are planned according to the preset axis motion parameters.

[0061] A2) Rotational motion axes (roll, pitch, yaw): Since the driving element is a servo electric cylinder, the target rotation angle needs to be converted into a displacement command for the servo electric cylinder.

[0062] For example, for the roll motion platform 4, such as Figure 12 As shown, the fixed fulcrum L is the moving end of the servo cylinder, and M is the roll shaft bearing housing; N0 is the initial position of the servo cylinder, N1 is the extreme position when the servo cylinder is extended, and N2 is the shortest position when the servo cylinder is retracted. The roll angle is known. Elongation of the servo electric cylinder Mapping relationship: in, This refers to the distance from the moving end of the servo electric cylinder to the roll shaft bearing housing. This refers to the extension length of the servo electric cylinder when it is in its initial position. The distance between the servo electric cylinder and the roll shaft bearing housing when the servo cylinder is in its initial position is a fixed dimension. The initial angle (which can be obtained from the design drawings).

[0063] The control unit will turn the corner according to the above formula (or a pre-calibrated lookup table). This is converted into servo cylinder displacement commands. Pitch and yaw are similar.

[0064] A3) Exercise parameter settings: The control unit also sets the motion speed, acceleration, deceleration, and smoothing time for each axis to ensure synchronization. This test bench is designed to ensure that each cycle of motion does not exceed 25 seconds.

[0065] Step 2012, Instruction Issuance Process The industrial control computer (host computer) packages the target position, speed, acceleration, and other parameters of each axis into a custom communication frame. This frame is then sent to the motion controller (slave computer) via the Modbus TCP protocol.

[0066] The motion controller parses the communication frames and calls motion control commands (such as MC_MoveAbsolute, MC_MoveVelocity, etc.) to drive the servo driver.

[0067] Step 202: Servo system drives the movement of the test bench. Step 2021, Servo driver and motor execution The motion controller periodically sends position commands to each servo drive via the EtherCAT bus. The servo drives employ closed-loop control (position loop, speed loop, current loop) to drive the servo motor (or servo cylinder) to move according to the commands. For the vertical axis (Z-axis) and rotary axes requiring braking, the servo drive also needs to control the on / off state of the brake relay to prevent slippage in the event of power failure.

[0068] Step 2022, Multi-axis synchronous control The two six-degree-of-freedom motion platforms, with a total of 12 axes, need to move synchronously to accurately reproduce the target pose.

[0069] The motion controller uses axis group function (as described above in this application) to set the 6 axes of the left test bench as one axis group and the 6 axes of the right test bench as another axis group.

[0070] The DNC_MoveDirectAbsolute command is used to start the axis group movement, ensuring that all axes start simultaneously and reach the target position at the same time.

[0071] Step 2023: Determining the Position of Movement The motion controller reads the actual position of each servo drive in real time (via EtherCAT bus). When the deviation between the actual position and the target position of all axes is less than the preset positioning accuracy (e.g., ±0.01mm), the motion is considered complete. If a timeout occurs or some axes fail to reach their target position, an alarm is triggered.

[0072] Step 203: Measure the actual spatial distance After reaching the target pose, the control unit activates the measurement module to obtain the actual spatial distance between the corresponding feature vertices on the windshield mounting plate 8.

[0073] Step 2031, Measurement Equipment and Installation This embodiment uses a pull-wire displacement sensor (such as the BLS-S-1500-R type). The installation method is as follows (taking four corner points as an example): Fix the housing (or fixed end) of the pull rope displacement sensor at the four corner points of the left windshield mounting plate 8.

[0074] Fix the free end of the pull rope at the corresponding corner point of the right windshield mounting plate 8.

[0075] The pull rope extends in a straight line along the space between the two plates, and the encoder inside the pull rope displacement sensor outputs the length of the pull rope (i.e., the distance between the two points) in real time.

[0076] For cases with at least four feature vertices, the corresponding number of rope displacement sensors are also arranged.

[0077] An alternative approach (such as using a laser tracker, but only for initial calibration and not as a routine verification method) involves setting up the laser tracker in front of the test bench, sequentially measuring the 3D coordinates of each feature vertex, and then calculating the Euclidean distance between the corresponding points. This method is more expensive but more accurate and suitable for calibration applications.

[0078] As a preferred embodiment of this application, a pull-string displacement sensor is used to measure the actual spatial distance. This enables rapid automatic verification after each test with extremely low equipment and maintenance costs, fully meeting the cycle time requirements of mass production lines.

[0079] As a supplementary or verification method in extreme cases (such as absolute calibration during the initial installation of the test bench or single sampling inspections with special accuracy requirements), a laser tracker can be used to sequentially measure the three-dimensional coordinates of each feature vertex and then calculate the Euclidean distance between the corresponding points. It should be noted that laser trackers are expensive, complex to operate, and time-consuming to measure, making them unsuitable as routine calibration tools on production lines. Therefore, using them as an alternative in this application is only for infrequent calibration scenarios and does not affect the substantial contribution of this application in solving the technical problem of low-cost and rapid calibration using a drawstring displacement sensor.

[0080] Step 2032, Measurement Triggering and Data Acquisition Automatic triggering: After the motion controller determines that the motion is in place, it triggers the data acquisition of the rope displacement sensor or the measurement command of the laser tracker through digital output (DO).

[0081] Manual triggering: The operator triggers the measurement by clicking the "Measure" button on the human-machine interface.

[0082] The measuring unit (such as a PLC analog input module or a dedicated counter) reads the distance values ​​from each pull-rope displacement sensor and converts them into actual spatial distance in millimeters. , , , (For the four corner points).

[0083] If the output of the rope displacement sensor is an analog signal (4-20mA or 0-10V), it needs to be converted from analog to digital and calibrated.

[0084] Step 2033, Data Verification and Filtering To eliminate noise, samples can be taken 3-5 times consecutively, and the average value can be used as the final measured distance. If a measurement value deviates significantly (e.g., exceeds the theoretical spatial distance ±50mm), the measurement is considered abnormal, and the operator is prompted to check the connection of the rope displacement sensor.

[0085] In special operating conditions, such as climbing: due to Z-axis displacement and pitch angle, the cable displacement sensor may not be horizontal, but the cable is always stretched in a straight line, which does not affect the distance measurement. Ensure that the cable displacement sensor has a sufficient range (in this embodiment, the cable displacement sensor has a range of 1500mm, which meets the measurement requirements). For outer rail superelevation conditions: the roll angle is large, and it is necessary to check whether the cable interferes with the edge of the windshield mounting plate 8 to avoid friction causing measurement errors. For reverse curve conditions: the yaw angles of the left and right test benches are equal, the relative rotation angle is zero, but there is still a large Y-axis displacement; sufficient space should be reserved for the installation of the cable displacement sensor.

[0086] Step 30: Compare the actual spatial distance with the theoretical spatial distance, and determine the positioning accuracy of the windshield test platform based on the comparison results.

[0087] The control unit will use the theoretical spatial distance obtained in step 10. The actual spatial distance measured in step 20 Compare one by one and calculate the absolute deviation. Preset tolerance (For example, 0.5mm) can be set by the operator according to the test requirements. If all If the positioning accuracy is satisfactory, the human-machine interface will display "satisfactory" and the result will be stored in the database; if any If so, it is deemed unqualified.

[0088] If the system fails to meet the requirements, it first attempts automatic error compensation: adjusting the target displacement of the corresponding motion axis according to the sign and magnitude of the deviation, and then re-executing steps 20 and 30. If the system still fails to meet the requirements after compensation or exceeds the maximum number of compensation attempts, an alarm is issued, prompting the operator to check the mechanical structure (such as the installation of the rope displacement sensor, limit switches, transmission clearance, etc.). After the operator completes the adjustment, the system can be recalibrated.

[0089] In the automated cyclic test, the decision result of step 30 serves as the basis for whether to continue to the next segment or the next cycle. All deviation data are recorded and used to generate test reports and perform equipment status trend analysis, as detailed in steps 301-307.

[0090] In one embodiment, the process of steps 301-307 includes: Step 301, Data Preparation After step 20 is completed, the control unit (industrial computer) has obtained: the theoretical spatial distance (from step 10): denoted as , , , Actual spatial distance (from the measurement in step 20): denoted as... , , , All distance units are uniformly set to millimeters (mm), with accuracy retained to one or two decimal places (depending on the accuracy of the cable displacement sensor, for example, the accuracy of the cable displacement sensor is ±0.1mm).

[0091] Step 302, Deviation Calculation The control unit calculates the absolute deviation of the distance to each corresponding feature vertex one by one: , Simultaneously, the relative deviation can be calculated: These deviation values ​​will be displayed on the human-machine interface and stored in the database. For example: Theoretical value: , , , ; Measured value: , , , ; deviation: , , , .

[0092] Step 303: Preset tolerance settings The tolerance (i.e. the maximum allowable deviation) is determined based on the accuracy requirements of the windshield test and the repeatability of the equipment.

[0093] Step 3031, Source of Tolerance The clearance between the windshield and the windshield mounting plate 8 is typically ±1mm, therefore the four-point spacing deviation should not exceed 0.5mm to 1mm. In this test bench, the servo motor control accuracy can reach 0.01mm, the ball screw lead is 12mm, and the overall positioning accuracy is better than ±0.05mm. However, considering the error of the cable displacement sensor (±0.1mm), mechanical clearance, etc., controlling the total deviation within ±0.5mm is reasonable. The dimensional tolerances for the windshield interface in TB / T 3094 "Windshields for Locomotives and Rolling Stock" can also be referenced.

[0094] Step 3032, Tolerance Setting Method Fixed tolerance (e.g.) ); or condition-related tolerances. For certain large deformation conditions (such as reverse curves), the tolerance can be appropriately relaxed to 1.0 mm; for precision positioning (such as windshield mounting point calibration), it can be tightened to 0.2 mm. Alternatively, user-defined tolerances can be used (for example, operators can input the tolerances of each feature vertex in the human-machine interface according to the test requirements).

[0095] Step 304, Decision Logic The control unit compares each From the preset tolerance : If all If the positioning accuracy is satisfactory, then the positioning accuracy is deemed acceptable.

[0096] If any If so, the positioning accuracy is deemed unqualified.

[0097] For example, If 1.3mm > 0.5mm, it is considered unqualified, and no other items need to be checked. The judgment result is displayed in text (qualified / unqualified) and color (green / red) through the human-machine interface, and is also written to the log and database.

[0098] Step 305: Follow-up processing when non-conforming When a non-compliance is determined, the control unit selects one or a combination of the following operations based on the preset non-compliance handling strategy configuration: B1) Automatic error compensation: The control unit corrects the target displacement of the corresponding motion axis according to the sign and magnitude of the deviation and re-executes the test.

[0099] For example, for the curve inlet condition: Observed > and < This indicates that the actual X-axis position of the right test bench is too large or the yaw angle is too large. Incremental PID (Proportional-Integral-Derivative, closed-loop control algorithm) or simple proportional compensation is used to correct the decrease in the target displacement of the right test bench's X-axis. (For example, reduce by 0.65mm), while slightly correcting the yaw angle. Repeat steps 20 and 30 until it passes or the maximum number of compensations is reached (e.g., 3 times). The compensation value must not exceed the mechanical hard limit or safety range; if it still fails after compensation, stop the test and trigger an alarm.

[0100] B2) Prompt for mechanical adjustment: If the automatic compensation is ineffective or the system is not configured with automatic compensation, a prompt message is sent through the human-machine interface, such as: "The in-place accuracy is unqualified. Please check the following possible reasons: whether the rope displacement sensor is loose or worn, whether the bolts of the windshield mounting plate 8 are tightened, whether the limit switch of the moving platform is offset, and whether there is clearance in the mechanical transmission components (lead screw, guide rail)." After the operator completes the mechanical adjustment, manually click the "Re-verify" button and repeat steps 20 and 30.

[0101] B3) Record and skip (only applicable to non-critical tests): For some pre-tests or commissioning phases, it can be configured to "record unqualified but continue", that is, the deviation data is stored in the database and the test is not stopped. However, it is not recommended for formal fatigue tests.

[0102] Step 306, Data recording and reporting when qualified When it is determined to be qualified, the control unit stores the working condition name, theoretical space distance, actual space distance, each deviation value, determination result, and timestamp of this test in the test result database. In the automatic test, the counter is incremented by 1 and the next segment or the next cycle continues. If it is a single verification, "Qualified" is displayed and the test ends.

[0103] Step 307, Integration in batch / cyclic tests In the automatic fatigue test, the determination result of step 30 affects the test process: Step 30 is executed every time the movement reaches the position.

[0104] If a certain determination is unqualified, the system can pause the test, record the current cycle number and working condition, and continue after manual intervention or automatic compensation.

[0105] If it is unqualified for multiple consecutive times (such as 3 times), the entire test is stopped and an alarm is given.

[0106] All determination results are recorded in the database, and a test report is finally generated, including statistical information such as the pass rate, maximum deviation, and out-of-tolerance working conditions.

[0107] Optionally, the windshield test bench is a twelve-degree-of-freedom test bench, and each of the left and right moving platforms has six degrees of freedom. The above accuracy verification method is also applicable to left and right double-end independent control: The theoretical space distance has separately considered the absolute poses of the left and right moving platforms, and the measured space distance is the direct distance between the corresponding points of the left and right windshield mounting plates 8. Therefore, the comparison result comprehensively reflects the motion errors of all degrees of freedom of the double ends. Even if there is a slight deviation in a certain axis on one side, as long as it ultimately affects the relative pose of the windshield mounting plate 8, it will be reflected in the four-point spacing, avoiding missed inspections.

[0108] In one embodiment, four corner points are collected as feature vertices, and a drawstring displacement sensor is used. The operator selects the curve entrance condition, and the system automatically retrieves the theoretical spatial distance under this condition from the database. , , , . After clicking "Start", the left and right test benches move to the target pose. After reading the actual spatial distance, comparison is made and "Qualified" is displayed. The entire process takes no more than 5 seconds.

[0109] In another exemplary embodiment, in a typical fatigue cycle test, the test needs to simulate a straight track, a curve entrance, the bottom of the curve arc, a reverse curve, the top of the curve arc, and a curve exit in sequence. The operator selects this cycle on the automatic test interface and inputs the number of cycles as 10 times. The system pre-calculates the theoretical spatial distance under each condition and stores it. During the test process, after each condition is completed, the system automatically conducts a four-point spacing verification; if the in-place accuracy of a certain condition is unqualified, the system pauses the test, gives an audible and visual alarm, and automatically performs error compensation (correcting the target position of subsequent conditions), and continues the test after compensation. After all cycles are completed, an accuracy verification report is generated.

[0110] In another exemplary embodiment, on the same test bench, single-axis measurement (reading the encoder positions of each servo motor) and the four-point spacing measurement in the present application are respectively used for comparison. The top of the curve arc condition is set, and a deviation of 0.1 is artificially introduced to the right yaw axis. The single-axis measurement shows that all axis deviations are within 0.05, and it is determined to be qualified; but in the four-point spacing and a deviation of 1.2 mm appears, and it is determined to be unqualified. It can be seen that the four-point spacing method can capture the end pose error amplified due to multi-axis coupling, while the single-axis measurement cannot detect it.

[0111] Through the above specific embodiments, the present application has achieved the following beneficial effects: By obtaining the theoretical spatial distance, a standardized theoretical benchmark is provided for the determination of in-place accuracy, solving the problem in the prior art of lacking a unified and quantified in-place accuracy benchmark, and realizing the standardized pre-calculation of the theoretical spatial distance under different vehicle models and different line conditions.

[0112] By driving the test bench to simulate the target condition and measuring the actual spatial distance, the actual relative position relationship of the end windshield mounting plate 8 can be directly obtained, solving the problem that the existing single-axis measurement method cannot reflect the end comprehensive pose error caused by multi-degree-of-freedom coupling, realizing the direct, fast, and low-cost measurement of the spatial pose of the windshield mounting plate 8, avoiding the missed detection of cumulative errors caused by axis-by-axis measurement, and at the same time overcoming the defects of high cost and complex operation of the laser tracker for measuring absolute coordinates.

[0113] By comparing the actual spatial distance with the theoretical spatial distance and determining the accuracy, a quantitative mapping relationship from measured data to accuracy conclusions was established. This solved the problem that existing accuracy verification methods cannot automatically and quantitatively determine whether the end pose meets the standard. It also realized the closed-loop self-check of the positioning accuracy of the windshield test bench, providing a clear basis for subsequent error compensation or mechanical adjustment, and significantly improving the reproducibility accuracy and reliability of the windshield test.

[0114] In summary, the accuracy verification method, test bench, and control system of the windshield test bench provided in this application can efficiently and accurately perform end-effector comprehensive pose verification on a twelve-degree-of-freedom test bench, and have high industrial practical value.

[0115] To more clearly illustrate the technical solution of this application, a complete application example is given below using a 25T bus windshield test as an example. The train parameters, track parameters, geometric derivations, and numerical results in this example are all derived from actual engineering data, but are for illustrative purposes only and do not constitute any limitation on this application.

[0116] Step 1: Train and track parameters This example uses the following parameters:

[0117] To facilitate subsequent derivation, an auxiliary angle is defined:

[0118] Step 2: Definition of the test bench coordinate system and calibration points The coordinate system of the left and right motion platforms is centrally symmetrical: the X-axis is positive on the left and positive on the right (opposite directions), the Y-axis is positive on the left and negative on the right, and the Z-axis is positive upwards; the roll angle (A-axis) is positive when rolling to the right, the pitch angle (B-axis) is positive when pitching downwards, and the yaw angle (C-axis) is positive counterclockwise. Four corner points are selected as feature vertices on the left and right windshield mounting plates 8, and the theoretical spatial distance between corresponding points is denoted as... , , , .

[0119] Step 3: Position and four-point spacing of the curve entry condition This embodiment uses the curve entry condition as an example for explanation. In this condition, the left test bench simulates the following vehicle in the straight section and remains stationary; the right test bench simulates the preceding vehicle that has just entered the curve and performs the corresponding six-degree-of-freedom motion.

[0120] Under this condition, the left test stand is stationary, while the right test stand is moving. The yaw angle of the right test stand is... (Negative values ​​indicate clockwise) Y-axis displacement X-axis displacement Calculate using the following formula: First, calculate the auxiliary angle. : Right test stand yaw angle : Y-axis displacement of the right test bench : A positive value indicates that the right test platform moves along the positive Y-axis, consistent with the coordinate system definition.

[0121] Angle between the line connecting the coupler and the horizontal line : X-axis displacement of the right test bench : Four-point spacing calculation: like Figure 7 As shown, the four corner points A, B, C, and D of the left windshield mounting plate, and the corresponding corner points E, F, G, and H of the right windshield mounting plate. (Distance from top left corner) and Taking the distance from the bottom left corner as an example, we can obtain the following from the Law of Cosines: in, (Width of right windshield mounting plate) Similarly, it can be calculated .

[0122] Substitution , , , Calculations show that: Step 4: Position and four-point spacing of the curve exit condition The exit and entry conditions of the curve are symmetrical about the origin of the coordinate system. In the entry condition, the right test bench moves (simulating the preceding vehicle entering the curve), and in the exit condition, the left test bench moves (simulating the following vehicle leaving the curve). Based on the sign relationship between the left and right test bench coordinate systems, their pose parameters are the same; therefore: Yaw angle of left test stand (Counterclockwise) Y-axis displacement of the left test bench (Negative) X-axis displacement of the left test bench (Also in the positive direction, due to symmetry, the front and back directions remain unchanged.) Four-point spacing: Because the geometry is perfectly symmetrical, the distance between corresponding points is the same as that at the inlet, i.e. , (Interchange corresponding corner points).

[0123] Step 5: Position and four-point spacing of the curve apex working condition Under this condition, the two carriages are symmetrically distributed at the apex of the curve, and the coupler line is horizontal. (Left test bench yaw angle) The calculation formula is: in: Substituting the parameters, we can calculate the result. (Counterclockwise), right test stand yaw angle (Clockwise).

[0124] X-axis displacement of the left and right motion platform: Y-axis displacement of the left and right motion platform: The distance between the four points of the left and right movement platforms (assuming the width of the windshield mounting plate) ): The deviation from the precise value is within an acceptable range.

[0125] Step 6: Position and four-point spacing in the curved bottom working condition The bottom arc condition and the top arc condition are symmetrical about the X-axis, therefore the yaw angle and Y-axis displacement are reversed, while the X-axis displacement remains unchanged: Yaw angle of left test stand: Yaw angle of right test stand: Y-axis displacement of the left test bench: Y-axis displacement of the right test bench: X-axis displacement: Four-point spacing: Step 7: Position and four-point spacing in the reverse curve (S-shaped) working condition This operating condition consists of two reverse curves and a straight insertion section. Based on geometric relationships, the yaw angles of the left and right test benches are equal and in the same direction (viewed along the positive Z-axis, counterclockwise is positive, and clockwise is negative): Substitute parameters ( ) Calculated .

[0126] Y-axis displacement of the left and right motion platforms (taking the left test platform as an example): in, , , , ; Substituting the parameters, we can calculate the result. .

[0127] X-axis displacement of the left and right motion platform: Substituting the parameters, we can calculate the result. .

[0128] Four-point spacing (equal under symmetrical working conditions): Step 8, Extra-high outer rail working condition The roll angle of the left and right motion platform is determined by the height difference between the inner and outer rails. and track gauge Decide: (The maximum superelevation for high-speed lines can be taken as 175mm, corresponding to approximately 7°). Under this condition, the four-point spacing is mainly affected by the roll angle. The formula can be derived from the geometric relationship of the windshield mounting plate 8, which is omitted here.

[0129] Step 9, Position during uphill operation According to the "High-Speed ​​Railway Design Code", the maximum gradient is taken as 35‰, then the pitch angle is: Z-axis displacement: in, The longitudinal distance from the coupler center to the bogie center (approximately 8000 mm in this example) is calculated as follows: The change in the distance between the four points is small under this condition, but it can still be verified using this method.

[0130] Step 10: Example of a complete verification process Taking the curve inlet condition as an example, the complete verification process is as follows: Step S1: The control unit calculates the theoretical spatial distance according to the above formula.

[0131] Substitute the corrected pose parameters ( , , , , )have to: , Step S2: The control unit sends commands to the right test platform: X-axis displacement 2.0mm, Y-axis displacement 209.1mm, yaw angle -6.84°, driving the right test platform to move. After reaching the desired position, the cable displacement sensor reads the actual spatial distance and measures: , , , Step S3: Calculate the deviation: , , , The preset tolerance is ±0.5mm. If the tolerance is exceeded, the product is deemed unqualified.

[0132] The system automatically performs iterative compensation on motion commands: first, it determines the direction of the error based on the sign of the deviation, for example... and This indicates that the absolute value of the actual yaw angle of the right test stand is too large and the X-axis displacement is too large. Therefore, according to the preset proportional coefficient (for example, taking...), The absolute values ​​of the X-axis displacement and yaw angle commands are reduced by half of the X-axis compensation value. In this example, the calculated X-axis compensation value is -0.65mm (i.e., the X-axis displacement is changed to 1.45mm), and the yaw angle compensation value is +0.1°. The compensated motion commands are reissued and the motion is driven. After re-measurement, the deviation meets the tolerance requirements and is judged to be qualified. The system records the compensation value for subsequent use.

[0133] The system reissues the compensated motion commands to the motion platform, driving the right test platform to move back into position. The actual spatial distance is then remeasured, yielding: , Meet the tolerance requirements, determine it as qualified, and record the compensation value for subsequent tests.

[0134] Taking the 25T passenger car as an example above, the calculation formulas for the absolute pose, the theoretical formula for the four-point spacing, and the specific numerical results under typical working conditions such as the curve entrance, curve exit, curve apex, curve bottom, reverse curve, superelevation of the outer rail, and climbing are given in detail, and the complete verification process is demonstrated. Those skilled in the art should understand that for other vehicle types (such as CRH (China Railway High-speed, i.e., Harmony Express) series high-speed multiple units) or other line parameters, only the train parameters and track parameters need to be adjusted accordingly, and the calculation can be performed according to the same geometric model without creative labor.

[0135] In an exemplary embodiment, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented. The processor and the memory can be integrated in an industrial control computer.

[0136] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0137] In this article, specific examples are used to elaborate on the principles and implementation manners of the present application. The descriptions of the above embodiments are only used to help understand the method of the present application and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for verifying the accuracy of a windshield test bench, wherein the windshield test bench has two moving platforms, left and right, characterized in that, include: Determine the theoretical spatial distance between at least four corresponding feature vertices on the end windshield mounting plates of the left and right motion platforms under the target working condition; The windshield test bench is driven to simulate the target working condition, and moves to the target pose. After reaching the target pose, the actual spatial distance between the corresponding feature vertices is measured. The actual spatial distance is compared with the theoretical spatial distance, and the positioning accuracy of the windshield test platform is determined based on the comparison result.

2. The accuracy verification method for the windshield test bench according to claim 1, characterized in that, The theoretical spatial distance is obtained by: calculating it based on the absolute pose model of the train under typical operating conditions and the geometric dimensions of the windshield mounting plate; or by reading it from a database that pre-stores theoretical spatial distances under different operating conditions.

3. The accuracy verification method for the windshield test bench according to claim 2, characterized in that, The typical operating conditions include one or more of the following: curve entry, curve exit, curve apex, curve apex, reverse curve, outer rail superelevation, and climbing.

4. The accuracy verification method for the windshield test bench according to claim 1, characterized in that, The at least four corresponding feature vertices are four points, namely the four corner points of the windshield mounting plate.

5. The accuracy verification method for the windshield test bench according to claim 1, characterized in that, The actual spatial distance was measured using a rope displacement sensor.

6. The accuracy verification method for the windshield test bench according to claim 1, characterized in that, The determination of positioning accuracy based on the comparison results includes: Calculate the deviation of each corresponding distance. If all deviations are less than the preset tolerance, the positioning accuracy is deemed qualified; otherwise, it is deemed unqualified.

7. The accuracy verification method for the windshield test bench according to claim 6, characterized in that, When a failure is detected, error compensation is performed on the motion parameters of the control system until the preset tolerance requirements are met.

8. The accuracy verification method for the windshield test bench according to any one of claims 1-7, characterized in that, The windshield test bench is a twelve-degree-of-freedom test bench, with each of the left and right motion platforms having six degrees of freedom.

9. A windshield test bench, characterized in that, include: Left six-degree-of-freedom motion platform and right six-degree-of-freedom motion platform; Windshield mounting plates are installed at the ends of the left and right motion platforms. Each windshield mounting plate is provided with at least four feature vertices for measuring the distance. and a control unit, the control unit comprising: The acquisition module is used to determine the theoretical spatial distance between all corresponding feature vertices under the target working condition; The drive module is used to drive the motion platform to the target pose; The measurement module is used to obtain the actual spatial distance between the corresponding feature vertices; The determination module is used to compare the actual spatial distance with the theoretical spatial distance and output the positioning accuracy determination result based on the comparison result.

10. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the accuracy verification method for the windshield test bench according to any one of claims 1-8.

Images