Device and method for detecting size and out-of-roundness of train wheel set

Abstract

The invention discloses a device and a method for detecting the size and out-of-roundness of a train wheel set. The device comprises a vision measurement module, a data processing module and a data processing module, wherein the vision measurement module is used for measuring geometric dimension parameters of the train wheel set; the wheel out-of-roundness measurement module is used for measuring the out-of-roundness of the train wheels; the out-of-roundness data acquisition module is used for acquiring measurement data; the image imaging module I is used for detecting the rotation angle of the measured wheel at the wheel set geometric dimension measurement position; the image imaging module II is used for detecting the rotation angle of the detected wheel at the wheel out-of-roundness measurement position; the wheel detection module is used for detecting the arrival of wheels; an electrical control module; and a data processing module. The wheel set size detection system and the wheel out-of-roundness measurement system are fused, the wheel out-of-roundness measurement data of the wheel out-of-roundness measurement system are corrected by using the high-precision wheel diameter data of the wheel set size detection system, the measurement precision and robustness of the wheel out-of-roundness measurement system can be effectively improved, the operation steps are simple, the cost is low, and the method is suitable for popularization and application. The maintenance is easy.

Description

Technical Field

[0001] The invention belongs to the field of measurement technology, and in particular relates to a device and method for detecting the size and out-of-roundness of a train wheel set. Background Art

[0002] The wear state of train wheels is one of the important factors affecting train driving safety. Automatically measuring the relevant geometric dimension parameters of wheels by detection devices has received widespread attention in recent years, and the technology has gradually developed. At present, most wheelset size detection equipment uses visual measurement technology. For example, the Chinese invention patent, publication number CN102889861A, discloses a dynamic measurement device for the geometric parameters of train wheelsets. Two base-mounted beams are set on both sides of the track, and structured light vision sensors for measuring the geometric parameters of train wheelsets are installed on the beams. When the train passes, the structured light projector projects a light strip to the train wheelset. The camera captures the image of the light strip and calculates the cross-sectional contour data of the wheelset where the light strip is located based on the structured light vision model, and then calculates the geometric parameters of the wheelset. However, this type of visual technology method only detects the local area when the wheelset passes, and cannot cover the contour dimensions of all positions in a complete circle of the wheel.

[0003] The wheel out-of-roundness measuring device belonging to another type of wheelset measurement is usually composed of a measuring unit and a signal processing unit composed of two or more parallelogram mechanisms and relative displacement sensors. For example, the Chinese invention patent, publication number CN2921779Y, discloses an online dynamic detection device for wheel tread scratches and out-of-roundness, which can reflect the change in the position of the apex of the wheel rim by using the contact between the wheel and the contact rod of the parallelogram mechanism when the wheelset rotates, thereby forming a radial change curve of the wheel, and realizing the measurement of the wheel diameter change of a complete circle of the wheel, that is, the out-of-roundness parameter. However, when the train wheel contacts the contact rod of the parallelogram scratch mechanism, the wheel of the above-mentioned measuring device will bring greater impact and vibration to the conventional undamped mechanism, which may easily cause the plate to instantly detach from the wheel rim, lose the measurement reference, and cause overall deviation. For the mechanism with a damper, the damper will become loose after repeated impacts when the wheel passes, and needs to be replaced frequently. If the replacement is not timely, it will also affect the measurement accuracy. Summary of the invention

[0004] In order to solve the technical problems existing in the prior art, the purpose of the present invention is to provide a device and method for detecting the size and out-of-roundness of train wheelsets.

[0005] In order to achieve the above purpose and the above technical effect, the technical solution adopted by the present invention is:

[0006] A train wheel set size and out-of-roundness detection device, comprising:

[0007] Visual measurement module for measuring geometric parameters of train wheelsets;

[0008] Wheel out-of-roundness measurement module for measuring the out-of-roundness of train wheels;

[0009] Out-of-roundness data acquisition module for collecting measurement data;

[0010] Image imaging module I for detecting the rotation angle of the wheel under test at the wheelset geometry measurement position;

[0011] Image imaging module II for detecting the rotation angle of the wheel under test at the wheel out-of-roundness measurement position;

[0012] A wheel detection module for detecting the arrival of wheels, installed upstream in the direction of train travel;

[0013] An electrical control module, which is respectively connected to the visual measurement module, the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module, the image imaging module I and the image imaging module II;

[0014] The data processing module is respectively connected with the visual measurement module, the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module, the image imaging module I and the image imaging module II.

[0015] The invention also discloses a method for detecting the size and out-of-roundness of a train wheelset, which comprises two steps of offline angle calibration and on-site real-time processing, and obtains a wheelset geometric size measurement result and an out-of-roundness measurement result with a clear wheel position.

[0016] Furthermore, the step of offline angle calibration includes:

[0017] Step 1: After installing and debugging each module on site, place the wheel at the measurement position, that is, the position where the wheel triggers the camera to be photographed during actual measurement;

[0018] Step 2: Mark a horizontal line passing through the wheel center on the side of the wheel;

[0019] Step 3: Turn on the laser of the visual measurement module, and mark the center line of the laser line projected on the side of the wheel according to the laser line appearing on the side of the wheel, and extend it to intersect with the horizontal line described in step 2;

[0020] Step 4: measuring the angle α between the horizontal line and the laser line as a correction value to obtain the specific position of the wheel during the dimension measurement.

[0021] Furthermore, the step of on-site real-time processing includes:

[0022] Step 1: Use the wheel detection module to detect whether the vehicle is coming;

[0023] Step 2: Use the electrical control module to control the laser of the visual measurement module to project laser;

[0024] Step 3: Using the wheel detection module to determine in real time whether the wheel has reached the shooting position of the visual measurement module according to the sensor information;

[0025] Step 4: Determine whether the wheel under test has reached the shooting position. If so, go to step 5; otherwise, go to step 3.

[0026] Step 5: Use the electrical control module to synchronously trigger the camera of the visual measurement module and the camera of the image imaging module I to collect images;

[0027] Step 6: Determine whether the wheel has reached the initial measurement position for wheel out-of-roundness. If so, proceed to step 7; otherwise, continue to determine.

[0028] Step 7: Use the electrical control module to trigger the out-of-roundness data acquisition module to collect out-of-roundness data and the camera of the image imaging module II to collect images;

[0029] Step 8: The data processing module calculates the wheel rotation angle θ1 according to the image collected by the camera of the image imaging module I, and obtains the wheel position θ when the current wheelset size is measured according to the correction value of the offline calibration. TW ,θ TW =θ1+α;

[0030] Step 9: The data processing module receives the image data collected by the camera of the visual measurement module and calculates to obtain the wheel position θ TW Wheel set geometric dimension parameters;

[0031] Step 10: Calculate the wheel rotation angle θ2 according to the image collected by the camera of the image imaging module II through the data processing module to obtain the initial measurement wheel position θ of the current out-of-roundness NR ,θ NR =θ2+90°;

[0032] Step 11: The data processing module initially measures the wheel position θ based on the out-of-roundness NR The collected and processed out-of-round wheel diameter data is assigned positions, that is, each wheel diameter value corresponds to a specific absolute position of the wheel;

[0033] Step 12: Compare θ TW The difference between the wheel diameter value measured by the position wheelset size and the wheel diameter value measured by the out-of-roundness;

[0034] Step 13: When θ TWWhen the difference Δr between the wheel diameter value measured by the position wheelset size and the wheel diameter value measured by the out-of-roundness is greater than the set threshold, the wheel diameter value measured by the out-of-roundness is corrected based on the wheelset size measurement value, that is, each out-of-roundness measurement wheel diameter value is added with Δr; when the difference Δr is less than or equal to the set threshold, no correction is performed. Finally, the wheelset geometric size measurement result and the out-of-roundness measurement result with a clear wheel position are obtained.

[0035] Furthermore, in step 1 of the on-site real-time processing, the wheel detection module includes at least one sensor deployed upstream of the detection position, and when the sensor continuously senses the arrival of at least two wheels, it is determined that a train is approaching.

[0036] Furthermore, in step three of the on-site real-time processing, a sensor located at the shooting position of the visual measurement module and included in the wheel detection module is used to directly detect whether each wheel has reached the shooting position; when the sensor cannot be installed at the shooting position of the visual measurement module due to limited conditions, at least two sensors are installed at a position close to the upstream of the shooting position of the visual measurement module, and the speed of the train is calculated according to the distance between the sensors and the time when the sensors sense the wheels, and the delay time for the wheels to reach the shooting position of the visual measurement module is calculated according to the speed and the distance between the sensor and the shooting position of the visual measurement module.

[0037] Furthermore, in step six of the on-site real-time processing, the wheel out-of-roundness measurement module and the data acquisition module are used to determine whether the wheel has reached the initial wheel out-of-roundness measurement position. When the wheelset rotates, the wheel rim is pressed by the steel plate of the wheel out-of-roundness measurement module, and the steel plate is displaced downward by pressure. The displacement sensor of the wheel out-of-roundness measurement module outputs corresponding perception data reflecting the radial displacement change of the wheelset during rotation. The data acquisition module digitally samples the analog quantity of the displacement sensor according to the set sampling frequency. When the displacement sensor value is greater than the set threshold, it is detected that the wheel has reached the initial measurement position.

[0038] Furthermore, in step eight of the on-site real-time processing, determining the rotation angle of the wheel by detecting and identifying the features of the wheel axle includes the following steps:

[0039] a: Detect the axle end cap in the wheel image and obtain the axle end cap image;

[0040] b: Detect characters in the shaft end cover image and obtain the corresponding character frame;

[0041] c: Get the center horizontal line of each character box as the character horizontal line;

[0042] d: Calculate the angle between the horizontal direction of the character and the horizontal direction of the image as the character rotation angle, and use the character rotation angle as the wheel rotation angle.

[0043] Furthermore, in step 11 of the on-site real-time processing, assuming that the train passes at a constant speed during the out-of-roundness measurement, the position corresponding to each out-of-roundness wheel diameter value starting from the first measurement data is θ NR +360° / N, where N is the number of out-of-roundness wheel diameter values ​​measured this time;

[0044] If the speed of the train changes greatly when passing through the out-of-roundness measurement system, a speed measurement device is installed, and the out-of-roundness data is linearly interpolated according to the measured speed value to obtain the out-of-roundness measurement wheel diameter value with uniform spatial distribution, and then the position is allocated.

[0045] Furthermore, in the step twelve of the on-site real-time processing, firstly according to the position θ TW Search for the out-of-roundness measurement value position when θ NR +i*360° / N=θ TW When , the i-th out-of-roundness measurement value is θ TW The measured value of the position and then calculate θ TW The difference Δr between the wheel diameter value measured by the position wheelset size and the wheel diameter value measured by the out-of-roundness.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention integrates the wheelset size detection system and the wheel out-of-roundness measurement system, and uses the high-precision wheel diameter data of the wheelset size detection system to calibrate the wheel out-of-roundness measurement data of the wheel out-of-roundness measurement system, which can effectively improve the measurement accuracy and robustness of the wheel out-of-roundness measurement system.

[0048] The present invention adopts a conventional industrial camera to realize the registration of wheelset size measurement data and out-of-roundness measurement data, which can effectively realize the alignment and fusion of data, has simple operation steps, low cost and easy maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a structural schematic diagram of a train wheel set size and out-of-roundness detection device of the present invention;

[0050] Figure 2 The figure is a diagram of the installation position of the camera of the image imaging module I of the present invention;

[0051] Figure 3 is a flow chart of the present invention;

[0052] Figure 4 An image taken by a camera facing the wheel of the present invention;

[0053] Figure 5 The image of the shaft end cover of the present invention, wherein: Figure 5 A is a schematic diagram of a character box. Figure 5B is a schematic diagram of the character rotation angle. DETAILED DESCRIPTION

[0054] The present invention is described in detail below so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0055] A brief summary of one or more aspects is given below to provide a basic understanding of these aspects. This summary is not an exhaustive overview of all conceived aspects, and is neither intended to identify the key or critical elements of all aspects nor to define the scope of any or all aspects. Its only purpose is to give some concepts of one or more aspects in a simplified form as a prelude to a more detailed description that will be given later.

[0056] like Figure 1-2 As shown, a train wheel set size and out-of-roundness detection device comprises:

[0057] Visual measurement module for measuring geometric parameters of train wheelsets;

[0058] Wheel out-of-roundness measurement module for measuring the out-of-roundness of train wheels;

[0059] Out-of-roundness data acquisition module for collecting measurement data;

[0060] Image imaging module I for detecting the rotation angle of the wheel under test at the wheelset geometry measurement position;

[0061] Image imaging module II for detecting the rotation angle of the wheel under test at the wheel out-of-roundness measurement position;

[0062] A wheel detection module for detecting the arrival of wheels, installed upstream in the direction of train travel;

[0063] An electrical control module, which is respectively connected to the visual measurement module, the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module, the image imaging module I and the image imaging module II;

[0064] The data processing module is respectively connected to the visual measurement module, the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module, the image imaging module I and the image imaging module II, and is used to receive the image data collected by the camera of the visual measurement module, the image data collected by the camera of the image imaging module I, the image data collected by the camera of the image imaging module II, and the wheel diameter data collected by the out-of-roundness data acquisition module and perform calculation processing to obtain the wheelset geometric size parameters and out-of-roundness data of the corresponding angle of the wheel, and perform correction processing on the out-of-roundness data.

[0065] In the present invention, the visual measurement module and the image imaging module I constitute a wheelset size detection system, and the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module and the image imaging module II constitute a wheel out-of-roundness measurement system.

[0066] The visual measurement module preferably uses a structured light vision sensor, which includes a camera and a structured light laser. The structured light laser preferably uses a line projection method. When a train wheelset passes, the structured light laser projects at least one line light strip onto the wheel, forming characteristic light strips on the outer surface and side of the wheel. The camera is controlled by a trigger pulse to capture an image of the wheel containing the laser light strip, and the geometric size parameters of the wheels of the wheelset are obtained through three-dimensional measurement calculation of structured light vision.

[0067] The wheel out-of-roundness measurement module is implemented using the existing contact measurement method. It consists of a measurement unit and a signal processing unit composed of two or more sets of parallelogram mechanisms and relative displacement sensors. When the wheelset rotates, the wheel rim is pressed by the steel plate of the parallelogram mechanism, and the steel plate is displaced downward by pressure. The displacement sensor outputs the corresponding perception data reflecting the radial displacement change of the wheelset during rotation, that is, the out-of-roundness parameters are calculated through the radial runout curve.

[0068] The data acquisition module can be implemented through an embedded data acquisition board or a PLC industrial control module, that is, the displacement sensor analog quantity is digitally sampled according to the set sampling frequency. At the same time, the data acquisition module includes a camera trigger unit. When the displacement sensor value is greater than the set threshold, that is, when the wheel is detected to reach the measurement module, a shooting trigger signal is generated and transmitted to the image imaging module II. Preferably, the trigger signal is a differential signal.

[0069] The structures of the imaging module I and the imaging module II are the same, both of which include a camera and a fill light, and can receive an external trigger source to trigger the photo. The camera installation position of the imaging module I is as follows: Figure 2 As shown in the figure, the camera is facing the wheel at the trigger position of the visual measurement module. The camera imaging plane is as parallel to the wheel surface, i.e. the rail, as possible. At the same time, the camera optical axis is as parallel to the wheel axle as possible. The camera field of view includes the entire wheel and the upper edge of the track. The fill light ensures the image quality at night or in poor light conditions, and facilitates the detection and recognition of features in the image to obtain the wheel rotation angle.

[0070] The camera of the imaging module II is located at the initial measurement position of the wheel out-of-roundness measurement module, that is, the position where the wheel rim just touches the steel plate. Like the camera of the imaging module I, the imaging plane of the camera is as parallel to the wheel surface, that is, the rail, as possible. At the same time, the optical axis of the camera is as parallel to the wheel axle as possible, and the camera field of view includes the entire wheel and the upper edge of the track.

[0071] The wheel detection module is used to detect the arrival of the wheels. Preferably, a magnetic steel sensor is used to sense the wheels. Preferably, multiple sets of magnetic steel sensors can be used to determine the arrival of the train and detect the moment when each wheel pair reaches the sensing position. When it is not convenient to install magnetic steel sensors on the visual measurement module, the train speed can be calculated by using multiple magnetic steel sensors with known distances between each other, and the trigger delay time is calculated by the distance between the magnetic steel sensor and the wheel measurement position of the visual measurement module to trigger the camera. In addition to using magnetic steel sensors, other types of sensors such as photoelectric sensors can also be used to detect the arrival of wheels.

[0072] After the electrical control module obtains the sensor signal, it turns on the laser in advance, and directly triggers the response signal of the detected wheel or calculates the vehicle speed based on multiple sets of magnetic steel signals to delay the triggering of the camera of the visual measurement module and the cameras of the image imaging module I and the image imaging module II to collect images at the same time. In order to ensure the accuracy of the wheel size measurement value and its corresponding wheel rotation angle, the same external trigger is used to synchronously trigger the camera of the visual measurement module and the cameras of the image imaging module I and the image imaging module II.

[0073] The data processing module receives the image data collected by the camera of the image imaging module I, and calculates the specific position of the wheel during the size measurement (described by the angle value relative to the axle end feature) according to the result of the wheel axle end feature recognition; the data processing module also receives the image data collected by the camera of the visual measurement module as the input of the train wheel contour structured light three-dimensional reconstruction calculation, and calculates the geometric size parameters of the wheel at the specific position; the data processing module also receives the image data collected by the camera of the image imaging module II, calculates the starting position of the wheel for the out-of-roundness measurement (described by the angle value relative to the axle end feature), and receives the wheel diameter data collected by the out-of-roundness data acquisition module, and corresponds the 360° position (described by the angle value relative to the axle end feature) to the out-of-roundness parameter according to the starting position of the measured wheel. According to the wheel position corresponding to the wheelset size measurement value, the data processing module finds the wheel diameter value of the position in the out-of-roundness measurement data and compares it with the wheel diameter value of the wheelset size measurement value. If there is a certain deviation between the two, the out-of-roundness measurement data is corrected based on the wheel diameter value of the wheelset size measurement.

[0074] like Figure 3 As shown, a method for detecting the size and out-of-roundness of a train wheelset includes two steps: offline angle calibration and on-site real-time processing.

[0075] Offline angle calibration includes the following steps:

[0076] Step 1: After installing and debugging each module on site, place the wheel at the measurement position, that is, the position where the wheel triggers the camera to be photographed during actual measurement;

[0077] Step 2: Mark a horizontal line passing through the wheel center on the side of the wheel;

[0078] Step 3: Turn on the laser of the visual measurement module, mark the center line of the laser line according to the laser line that appears on the side of the wheel, and extend it to intersect with the horizontal line of the wheel in step 2, such as Figure 2 As shown;

[0079] Step 4: Measure the angle α between the horizontal wheel line and the laser line. You can stick a target point on the line, use a laser tracker to track the target point, measure the three-dimensional coordinate values ​​of each point, and then perform plane and straight line fitting. Then calculate the angle between the two straight lines after fitting. You can also use a digital inclinometer or other equipment to measure.

[0080] Real-time processing includes the following steps:

[0081] Step 1: Use the wheel detection module to detect the arrival of the vehicle. The wheel detection module includes at least one sensor deployed at the far end of the visual measurement module track, i.e., upstream of the detection position, to sense the approaching vehicle. Preferably, a magnetic steel sensor is used to sense the wheels. When at least two wheels are continuously sensed to arrive, it is determined that a train has arrived. In addition to using magnetic steel sensors, other sensors such as photoelectric sensors can also be used to detect the arrival of a train;

[0082] Step 2: Using an electrical control module to control the laser to project laser light;

[0083] Step 3: Using the wheel detection module to determine in real time whether the wheel has reached the shooting position of the visual measurement module according to the sensor information;

[0084] Preferably, the magnetic steel sensor included in the wheel detection module and located at the shooting position of the visual measurement module is used to directly detect whether each wheel has reached the shooting position; when the sensor cannot be installed at the shooting position of the visual measurement module due to limited conditions, two or more sensors, preferably magnetic steel sensors, can be installed at a position close to the upstream of the shooting position of the visual measurement module, and the speed of the train is calculated according to the distance between the magnetic steel sensors and the time when the magnetic steel sensors sense the wheels, and the delay time for the wheels to reach the shooting position of the visual measurement module is calculated according to the speed and the distance between the magnetic steel sensor and the shooting position of the visual measurement module;

[0085] Step 4: Determine whether the wheel under test has reached the shooting position. If so, go to step 5; otherwise, go to step 3.

[0086] Step 5: Use the electrical control module to synchronously trigger the camera of the visual measurement module and the camera of the image imaging module I to collect images using the same external trigger signal;

[0087] Step six, determine whether the wheel has reached the initial measurement position of the wheel out-of-roundness. If the wheel has reached it, go to step seven, otherwise keep judging. This judgment is realized by the wheel out-of-roundness measurement module and the data acquisition module. Specifically, the out-of-roundness measurement module is composed of a measurement unit and a signal processing unit composed of two or more sets of parallelogram mechanisms and relative displacement sensors. When the wheelset rotates, when the wheel rim is pressed by the steel plate of the parallelogram mechanism, the steel plate is displaced downward by pressure. The displacement sensor outputs the corresponding perception data reflecting the radial displacement change of the wheelset during rotation. The data acquisition module digitally samples the analog quantity of the displacement sensor according to the set sampling frequency. When the displacement sensor value is greater than the set threshold, it is detected that the wheel has reached the initial position of the measurement module.

[0088] Step 7: Use the electrical control module to trigger the out-of-roundness data acquisition module to collect data and the camera of the image imaging module II to collect images. Preferably, the trigger signal is a differential signal to resist interference;

[0089] Step 8: The data processing module calculates the wheel rotation angle θ1 (with the horizontal position of the characters on the wheel axle end cover sign as the angle 0° position) based on the image collected by the camera of the image imaging module I, and then corrects the wheel position of the current wheelset size measurement according to the offline calibration angle: θ TW =θ1+α;

[0090] Step 9: The data processing module receives the image data collected by the camera of the visual measurement module and calculates the wheel position θ TW The calculation process includes extracting the coordinates of the center of the laser light strip from the wheel image, and then calculating the wheel section profile data from the center of the light strip combined with the calibration parameters of the structured light 3D measurement model, and calculating various geometric size parameters based on the profile data. This calculation method is well known and will not be described in detail.

[0091] Step 10: The data processing module calculates the wheel rotation angle θ2 using the same method based on the image captured by the camera of the image imaging module II. Since the out-of-roundness measurement position is at the wheel rim that touches the steel plate of the measurement mechanism, which is located at the bottom of the wheel and is 90° away from the horizontal direction, the current out-of-roundness initial measurement wheel position is: θ NR =θ2+90°;

[0092] Step 11: The data processing module initially measures the wheel position θ based on the out-of-roundness NR The collected and processed out-of-roundness wheel diameter data is assigned positions, that is, each wheel diameter value corresponds to a specific absolute wheel position (expressed as an angle value). Assuming that the train passes at a constant speed during the out-of-roundness measurement, the position corresponding to each out-of-roundness wheel diameter value starting from the first measurement data is: θ NR+360° / N, N is the number of out-of-roundness wheel diameter values ​​measured this time; if the speed of the train passing through the out-of-roundness measurement system changes greatly, a speed measurement device can be installed, and the out-of-roundness data can be linearly interpolated according to the measured speed value to obtain the out-of-roundness measurement wheel diameter value with uniform spatial distribution, and then the position allocation can be performed according to the above method;

[0093] Step 12: Compare θ TW The difference between the wheel diameter value measured by the position wheel set size and the wheel diameter value measured by the out-of-roundness. TW Search for the out-of-roundness measurement value position when θ NR +i*360° / N=θ TW When , the i-th out-of-roundness measurement value is θ TW The measured value of the position is used to calculate the difference Δr between the two wheel diameters;

[0094] Step 13: When the difference Δr between the two wheel diameter values ​​is greater than the set threshold, it is considered that there is a systematic error in the out-of-roundness measurement system, and correction is performed based on the wheelset size measurement value, that is, each out-of-roundness measurement wheel diameter value is added with Δr. When the difference Δr is less than or equal to the set threshold, no correction is performed. Finally, the wheelset geometric size measurement result and out-of-roundness measurement result with a clear wheel position are obtained.

[0095] The image taken by the camera facing the wheel is as follows Figure 4 As shown, the shaft end cover on the wheel axle is fixed, as shown in FIG. Figure 4 In the area shown in the red box, the rotation angle of the wheel is determined by detecting and identifying the features of the wheel axle. Preferably, the characters on the axle end cap sign are used as features, and the method of detecting and identifying characters is adopted. The rotation angle of the characters is used as the wheel rotation angle, and the horizontal direction is defined as 0° and the counterclockwise rotation is defined as the positive angle. The specific steps are as follows:

[0096] a: In Figure 4 Detect the axle end cap in the wheel image and obtain the axle end cap image. This can be achieved using existing deep learning detection networks, such as the YOLO series network, SSD network, etc.

[0097] In this step, in order to improve the angle calculation accuracy, the wheel image can be distorted and corrected to meet the conditions that the camera is facing the wheel and the rail is in a horizontal state in the image. Specifically, the Zhang Zhengyou calibration method can be used. After the camera is fixed, the track is used as the horizontal direction of the world coordinate system. Targets are placed multiple times at the position where the wheel appears to obtain multiple calibration images, and the internal and external parameters of the camera are obtained. The wheel image is corrected according to the internal and external parameters of the camera obtained by calibration. This method is well known and will not be described in detail.

[0098] b: Character detection is performed in the shaft end cover image, and the following is obtained: Figure 5The character box shown in the dotted line of A can be implemented using existing deep learning character detection networks, such as DBNet, EAST, TextBoxes++, etc.;

[0099] c: Get the center horizontal line of each character box as the character horizontal line;

[0100] d: Calculate the angle between the character horizontal direction and the image horizontal direction as the character rotation angle, such as Figure 5 B. The character rotation angle is used as the wheel rotation angle. The counterclockwise direction is a positive angle.

[0101] Parts or structures not specifically described in the present invention may adopt existing technologies or existing products and will not be described in detail here.

[0102] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A train wheel set size and out-of-roundness detection device, characterized in that: include: Visual measurement module for measuring geometric parameters of train wheelsets; Wheel out-of-roundness measurement module for measuring the out-of-roundness of train wheels; Out-of-roundness data acquisition module for collecting measurement data; Image imaging module I for detecting the rotation angle of the wheel under test at the wheelset geometry measurement position; Image imaging module II for detecting the rotation angle of the wheel under test at the wheel out-of-roundness measurement position; A wheel detection module for detecting the arrival of wheels, installed upstream in the direction of train travel; An electrical control module, which is respectively connected to the visual measurement module, the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module, the image imaging module I and the image imaging module II; The data processing module is respectively connected with the visual measurement module, the wheel out-of-roundness measurement module, the out-of-roundness data acquisition module, the image imaging module I and the image imaging module II.

2. A method for detecting the size and out-of-roundness of a train wheel set, characterized in that: The method includes two steps: offline angle calibration and on-site real-time processing, and obtains the wheelset geometric dimension measurement results and out-of-roundness measurement results with clear wheel positions.

3. A method for detecting the size and out-of-roundness of a train wheel set according to claim 2, characterized in that: The step of offline angle calibration includes: Step 1: After installing and debugging each module on site, place the wheel at the measurement position, that is, the position where the wheel triggers the photo taking during actual measurement; Step 2: Mark a horizontal line passing through the wheel center on the side of the wheel; Step 3: Turn on the laser of the visual measurement module, and mark the center line of the laser line projected on the side of the wheel according to the laser line appearing on the side of the wheel, and extend it to intersect with the horizontal line described in step 2; Step 4: measuring the angle α between the horizontal line and the laser line as a correction value to obtain the specific position of the wheel during the dimension measurement.

4. A method for detecting the size and out-of-roundness of a train wheel set according to claim 2, characterized in that: The steps of on-site real-time processing include: Step 1: Use the wheel detection module to detect whether the vehicle is coming; Step 2: Use the electrical control module to control the laser of the visual measurement module to project laser; Step 3: Using the wheel detection module to determine in real time whether the wheel has reached the shooting position of the visual measurement module according to the sensor information; Step 4: Determine whether the wheel under test has reached the shooting position. If so, go to step 5; otherwise, go to step 3. Step 5: Use the electrical control module to synchronously trigger the camera of the visual measurement module and the camera of the image imaging module I to collect images; Step 6: Determine whether the wheel has reached the initial measurement position for wheel out-of-roundness. If so, proceed to step 7; otherwise, continue to determine. Step 7: Use the electrical control module to trigger the out-of-roundness data acquisition module to collect out-of-roundness data and the camera of the image imaging module II to collect images; Step 8: The data processing module calculates the wheel rotation angle θ1 according to the image collected by the camera of the image imaging module I, and obtains the wheel position θ when the current wheelset size is measured according to the correction value of the offline calibration. TW ,θ TW =θ1+α; Step 9: The data processing module receives the image data collected by the camera of the visual measurement module and calculates to obtain the wheel position θ TW Wheel set geometric dimension parameters; Step 10: Calculate the wheel rotation angle θ2 according to the image collected by the camera of the image imaging module II through the data processing module to obtain the initial measurement wheel position θ of the current out-of-roundness NR ,θ NR =θ2+90°; Step 11: The data processing module initially measures the wheel position θ based on the out-of-roundness NR The collected and processed out-of-round wheel diameter data is assigned positions, that is, each wheel diameter value corresponds to a specific absolute position of the wheel; Step 12: Compare θ TW The difference between the wheel diameter value measured by the position wheelset size and the wheel diameter value measured by the out-of-roundness; Step 13: When θ TW When the difference Δr between the wheel diameter value measured by the position wheelset size and the wheel diameter value measured by the out-of-roundness is greater than the set threshold, the wheel diameter value measured by the out-of-roundness is corrected based on the wheelset size measurement value, that is, each wheel diameter value measured by the out-of-roundness is added with Δr; when the difference Δr is less than or equal to the set threshold, no correction is performed; finally, the wheelset geometric size measurement result and the out-of-roundness measurement result with a clear wheel position are obtained.

5. A method for detecting the size and out-of-roundness of a train wheel set according to claim 4, characterized in that: In step 1, the wheel detection module includes at least one sensor deployed upstream of the detection position. When the sensor continuously senses the arrival of at least two wheels, it is determined that a train is approaching.

6. A method for detecting the size and out-of-roundness of a train wheel set according to claim 4, characterized in that: In step three, the sensor located at the shooting position of the visual measurement module and included in the wheel detection module is used to directly detect whether each wheel has reached the shooting position; when the sensor cannot be installed at the shooting position of the visual measurement module due to limited conditions, at least two sensors are installed at a position close to the upstream of the shooting position of the visual measurement module, and the speed of the train is calculated according to the distance between the sensors and the time when the sensors sense the wheels, and the delay time for the wheels to reach the shooting position of the visual measurement module is calculated according to the speed and the distance between the sensor and the shooting position of the visual measurement module.

7. A method for detecting the size and out-of-roundness of a train wheel set according to claim 4, characterized in that: In step six, the wheel out-of-roundness measurement module and the data acquisition module are used to determine whether the wheel has reached the initial measurement position of the wheel out-of-roundness. When the wheelset rotates, the wheel rim is pressed by the steel plate of the wheel out-of-roundness measurement module, and the steel plate is displaced downward by pressure. The displacement sensor of the wheel out-of-roundness measurement module outputs corresponding perception data reflecting the radial displacement change of the wheelset during rotation. The data acquisition module digitally samples the analog quantity of the displacement sensor according to the set sampling frequency. When the displacement sensor value is greater than the set threshold, it is detected that the wheel has reached the initial measurement position.

8. A method for detecting the size and out-of-roundness of a train wheel set according to claim 4, characterized in that: In step eight, the rotation angle of the wheel is determined by detecting and identifying the features of the wheel axle, including the following steps: a: Detect the axle end cap in the wheel image and obtain the axle end cap image; b: Detect characters in the shaft end cover image and obtain the corresponding character frame; c: Get the center horizontal line of each character box as the character horizontal line; d: Calculate the angle between the horizontal direction of the character and the horizontal direction of the image as the character rotation angle, and use the character rotation angle as the wheel rotation angle.

9. A method for detecting the size and out-of-roundness of a train wheel set according to claim 4, characterized in that: In step 11, assuming that the train is passing at a constant speed when the out-of-roundness is measured, the position corresponding to each out-of-roundness wheel diameter value starting from the first measurement data is θ NR +360° / N, where N is the number of out-of-roundness wheel diameter values ​​measured this time; If the speed of the train changes greatly when passing through the out-of-roundness measurement system, a speed measurement device is installed, and the out-of-roundness data is linearly interpolated according to the measured speed value to obtain the out-of-roundness measurement wheel diameter value with uniform spatial distribution, and then the position is allocated.

10. A method for detecting the size and out-of-roundness of a train wheel set according to claim 4, characterized in that: In step 12, first press position θ TW Search for the out-of-roundness measurement value position when θ NR +i*360° / N=θ TW When , the i-th out-of-roundness measurement value is θ TW The measured value of the position and then calculate θ TW The difference Δr between the wheel diameter value measured by the position wheelset size and the wheel diameter value measured by the out-of-roundness.

Images