A method and system for fault detection of sintering machine trolley wheels

Abstract

This application relates to the field of metallurgical sintering, and provides a method and system for fault detection of wheels on sintering machine trolleys. The method determines whether the sintering machine trolley has wheels, and whether the wheel to be tested is a front or rear wheel, obtaining an image of the front or rear wheel to be tested. A coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis. Multiple planes are marked between the infrared laser and the wheel to be tested, perpendicular to the laser direction, and the imaging position of the laser line emitted by the infrared laser in the image to be tested, as well as the corresponding coordinate points, are obtained. The coordinate points of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted. This method can promptly detect trolley wheel faults and provide alarm feedback, reducing the time the trolley operates with faults and reducing the workload of manual inspection in this part of the process.

Description

Technical Field

[0001] This application relates to the field of metallurgical sintering, and in particular to a fault detection method and system for the wheels of a sintering machine trolley. Background Technology

[0002] Sintering is a crucial step in the steel production process. The general process involves mixing iron ore powder, lime powder, and coal powder in a specific ratio to form a mixture, which is then calcined to obtain sinter with sufficient strength and particle size. The sintering machine is the main equipment in the steel production process, and its proper operation directly affects the stability of the entire production process. The sintering machine trolley, as a core component of the sintering machine, plays a vital role in determining the machine's operating rate.

[0003] like Figure 1 The diagram shows a structural schematic of a sintering machine trolley in the prior art, including: a frame 01, a side panel 02, wheels 03, and a track 04; wherein the frame 01 is mounted on the wheels 03, the side panel 03 is mounted on the frame 01, and the wheels 03 can drive the entire sintering machine trolley to move on the track 04 to ensure that the sintering process proceeds in an orderly manner.

[0004] During the sintering process, the wheels 03 of the trolley may buckle due to prolonged operation under high temperatures, pressure from the frame 01, or friction between the wheels 03 and the track 04. This can cause the connection between the wheels 03 and the frame 01 to loosen, resulting in the wheels 03 running off course. If this misalignment goes unnoticed for a long time, the wheels 03 may wobble and fall off. Due to collisions, the wheels 03 may also fall off directly. If the wheel misalignment goes unnoticed for a long time, the wear on the wheels 03 will gradually become severe, eventually leading to their scrapping. In severe cases, the wheels 03 may even detach from the track 04, causing the sintering machine trolley to malfunction. If the wheels 03 fall off, the machine must be stopped immediately and the trolley replaced, which seriously affects normal production.

[0005] To prevent wheel 03 malfunctions from affecting sintering production, abnormal wheel 03 detection is extremely important. In existing technologies, abnormal detection of wheel 03 malfunctions on sintering machine trolleys in steel plants mainly relies on manual inspection or monitoring by central control personnel via industrial television. Manual inspection is affected by the inspection interval, resulting in delays and making it difficult to detect wheel slippage in a timely manner. Central control personnel monitoring wheel 03 via industrial television also suffers from timeliness issues, and manual judgment makes it difficult to identify abnormal conditions such as wheel swaying or camber. Summary of the Invention

[0006] In order to detect wheel misalignment and wheel drop of sintering machine trolley in a timely manner, this application provides a fault detection method and system for sintering machine trolley wheels.

[0007] This application provides a fault detection method for the wheels of a sintering machine trolley, the fault detection method for the wheels of the sintering machine trolley includes:

[0008] Preliminary images of the sintering machine trolley wheels and their corresponding steel stamp numbers were obtained using a visible light camera during the first round of inspection.

[0009] Based on the preliminary images of the sintering machine trolley wheels detected in the first round, after one sampling time, preliminary images of the sintering machine trolley wheels detected in the second round are obtained. The preliminary images of the sintering machine trolley wheels detected in the first round are compared with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary images of the sintering machine trolley wheels detected in the second round, the sintering machine trolley stops running. If the wheels are present in the preliminary images of the sintering machine trolley wheels detected in the second round, the preliminary images of the sintering machine trolley wheels detected in the second round are stored as the wheel image to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested.

[0010] Based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, the image of the wheel to be tested and the steel stamp number of the trolley to be tested are compared to determine whether the wheel to be tested is a front wheel or a rear wheel, and the image of the front wheel to be tested or the image of the rear wheel to be tested are obtained.

[0011] A coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray that intersects the origin and is perpendicular to the ground as the Y-axis.

[0012] Multiple planes are marked between the infrared laser instrument and the wheel under test, perpendicular to the direction of the laser, and the imaging position of the laser line emitted by the infrared laser instrument in the image under test, as well as the corresponding coordinate point, are obtained.

[0013] The coordinates of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted.

[0014] Optionally, the sampling time can be obtained through a sampling time model, which is:

[0015]

[0016] L represents the wheel spacing. v is the sampling time. i Let be the machine speed during the i-th sampling. The threshold is close to 0.

[0017] Optionally, in the step of comparing the preliminary images of the sintering machine trolley wheels detected in the first round with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels, the following method is used, including:

[0018] Preprocessing is performed on the preliminary images of the sintering machine trolley wheels from the first round of inspection and the preliminary images of the sintering machine trolley wheels from the second round of inspection to obtain preprocessed data.

[0019] By performing edge detection on the preprocessed data, the coordinates of each pixel in the preliminary image are obtained.

[0020] Based on the coordinates of each pixel in the preliminary image, the presence or absence of a wheel is determined using the Hough circle detection method.

[0021] Optionally, the specific steps for determining whether the wheel is tilted based on the slope of the fitted straight line are as follows:

[0022] If the slope is 0 or less than a threshold close to 0, then the wheel does not tilt.

[0023] If the slope is too large, the wheel will tilt.

[0024] Optionally, the step of determining whether the wheel is tilted further includes:

[0025] Based on the distance range [P3, P4] from the wheel to the infrared laser, the centrally symmetrical point P0 is obtained.

[0026] Determine whether P0P3 and P0P4 are vertically symmetrical about the center point P0.

[0027] Within the P0P3 range, n points are sampled at equal intervals, denoted as: Within the P0P4 range, n points are sampled at equal intervals, denoted as:

[0028] The wheel tilt error e is calculated using the following model:

[0029]

[0030] in, express The value of the point's coordinates in the x-direction.

[0031] The larger the tilt error 'e', ​​the more severe the wheel tilt. When 'e' exceeds a certain threshold, it indicates that the wheel is misaligned.

[0032] In a second aspect, this application provides a fault detection system for the wheels of a sintering machine trolley, the system comprising an acquisition module, a steel stamp number identification and storage module, a front and rear wheel determination module, and a wheel tilt determination module.

[0033] The acquisition module is used to acquire preliminary images of the wheels of the sintering machine trolley in the first round of detection, as well as the corresponding steel stamp number of the sintering machine trolley, through a visible light camera.

[0034] The steel stamp number identification and storage module is used to acquire a preliminary image of the sintering machine trolley wheels for the second round of detection after one sampling time, based on the preliminary image of the sintering machine trolley wheels detected in the first round. The module compares the preliminary image of the sintering machine trolley wheels detected in the first round with the preliminary image of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels detected in the second round, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheels detected in the second round, the preliminary image of the sintering machine trolley wheels detected in the second round is stored as the wheel image to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested.

[0035] The front and rear wheel determination module is used to compare the image of the wheel to be tested and the steel stamp number of the trolley to be tested based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, and determine whether the wheel to be tested is a front wheel or a rear wheel, thereby obtaining the image of the front wheel to be tested or the image of the rear wheel to be tested.

[0036] The wheel tilt determination module is used to establish a coordinate system with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis; to mark multiple planes between the infrared laser and the wheel under test in a direction perpendicular to the laser, and to obtain the imaging position of the laser line emitted by the infrared laser in the image under test, as well as the corresponding coordinate points; to fit the coordinate points of the wheel to obtain a fitted straight line, and to determine whether the wheel is tilted based on the slope of the fitted straight line.

[0037] Optionally, the sampling time can be obtained through a sampling time model, which is:

[0038]

[0039] L represents the wheel spacing. v is the sampling time. i Let be the machine speed during the i-th sampling. The threshold is close to 0.

[0040] Optionally, in the step of comparing the preliminary images of the sintering machine trolley wheels detected in the first round with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels, the following method is used, including:

[0041] Preprocessing is performed on the preliminary images of the sintering machine trolley wheels from the first round of inspection and the preliminary images of the sintering machine trolley wheels from the second round of inspection to obtain preprocessed data.

[0042] By performing edge detection on the preprocessed data, the coordinates of each pixel in the preliminary image are obtained.

[0043] Based on the coordinates of each pixel in the preliminary image, the presence or absence of a wheel is determined using the Hough circle detection method.

[0044] Optionally, the specific steps for determining whether the wheel is tilted based on the slope of the fitted straight line are as follows:

[0045] If the slope is 0 or less than a threshold close to 0, then the wheel does not tilt.

[0046] If the slope is too steep, the wheel will tilt.

[0047] Optionally, the step of determining whether the wheel is tilted further includes:

[0048] Based on the distance range [P3, P4] from the wheel to the infrared laser, the centrally symmetrical point P0 is obtained;

[0049] Determine whether P0P3 and P0P4 are vertically symmetrical about the center point P0.

[0050] Within the P0P3 range, n points are sampled at equal intervals, denoted as: Within the P0P4 range, n points are sampled at equal intervals, denoted as:

[0051] The wheel tilt error e is calculated using the following model:

[0052]

[0053] in, express The value of the point's coordinates in the x-direction.

[0054] The larger the tilt error 'e', ​​the more severe the wheel tilt. When 'e' exceeds a certain threshold, it indicates that the wheel is misaligned.

[0055] As can be seen from the above technical solution, this application provides a fault detection method and system for sintering machine trolley wheels. A preliminary image of the sintering machine trolley wheels in the first round of detection, along with the corresponding sintering machine trolley steel stamp number, is acquired using a visible light camera. Based on the preliminary image of the sintering machine trolley wheels in the first round of detection, a preliminary image of the sintering machine trolley wheels in the second round of detection is acquired after one sampling time. The preliminary images of the sintering machine trolley wheels in the first round of detection and the preliminary images of the sintering machine trolley wheels in the second round of detection are compared to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels in the second round of detection, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheels in the second round of detection, the preliminary image of the sintering machine trolley wheels in the second round of detection is stored as the image of the wheel to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested.

[0056] Based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, the image of the wheel to be tested and the steel stamp number of the trolley to be tested are compared to determine whether the wheel to be tested is a front wheel or a rear wheel, and the image of the front wheel to be tested or the image of the rear wheel to be tested is obtained; a coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis; multiple planes are marked between the infrared laser and the wheel to be tested in the direction perpendicular to the laser, and the imaging position of the laser line emitted by the infrared laser in the image to be tested, as well as the corresponding coordinate points, are obtained; the coordinate points of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted. The wheel fault detection method provided in this application can detect trolley wheel faults in a timely manner and provide alarm feedback, reducing the time the trolley runs with faults and reducing the workload of manual inspection in this part of the work. The equipment is relatively simple to install, has low maintenance costs, requires few supporting devices, and is stable and reliable. Attached Figure Description

[0057] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a schematic diagram of the structure of an existing sintering machine trolley;

[0059] Figure 2 This is a positional diagram of the visible light camera and the infrared laser in the embodiments of this application;

[0060] Figure 3 This is a schematic diagram of the calibration plane in the embodiments of this application;

[0061] Figure 4 This is a simulated measurement image for an embodiment of this application;

[0062] Figure 5 This is a schematic diagram of infrared laser beams under normal and tilted wheel conditions.

[0063] Figure 6 Based on Figure 5 The graph after symmetry treatment of the line equations in the diagram;

[0064] Figure 7 A flowchart illustrating a fault detection method for the wheels of a sintering machine trolley provided in an embodiment of this application;

[0065] Figure 8 A framework diagram of a fault detection system for the wheels of a sintering machine trolley provided in an embodiment of this application.

[0066] in:

[0067] 01 Frame, 02 Sideboard, 03 Wheels, 04 Track, 1 Visible light camera, 2 Infrared laser. Detailed Implementation

[0068] The embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described below do not represent all embodiments consistent with this application. They are merely examples of systems and methods consistent with some aspects of this application as detailed in the claims.

[0069] like Figure 2 The image shown is a positional diagram of the visible light camera and the infrared laser in an embodiment of this application.

[0070] To illustrate the implementation scenarios of the embodiments of this application, such as Figure 2 The diagram shown is a positional diagram of the visible light camera and the infrared laser in an embodiment of this application. The infrared laser is positioned directly opposite the frame 01 on both sides of the frame 01. The visible light camera 1 and the infrared laser 2 are set at a 45° angle and are 1-3m away from the trolley.

[0071] Figure 7 This is a flowchart illustrating a method for detecting faults in the wheels of a sintering machine trolley, as provided in an embodiment of this application.

[0072] The first aspect of this application provides a method for detecting faults in the wheels of a sintering machine trolley, the method comprising:

[0073] S101 uses a visible light camera to acquire preliminary images of the wheels of the sintering machine trolley during the first round of inspection, as well as the corresponding steel stamp number of the sintering machine trolley.

[0074] S102, based on the preliminary images of the sintering machine trolley wheels detected in the first round, after one sampling time, a preliminary image of the sintering machine trolley wheels detected in the second round is obtained. The preliminary images of the sintering machine trolley wheels detected in the first round are compared with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels detected in the second round, the sintering machine trolley stops running. If the wheels exist in the preliminary image of the sintering machine trolley wheels detected in the second round, the preliminary image of the sintering machine trolley wheels detected in the second round is stored as the wheel image to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested.

[0075] The sampling time can be obtained through a sampling time model, which is as follows:

[0076]

[0077] L represents the wheel spacing. Let be the sampling time, and vi be the machine speed during the i-th sampling. The threshold is close to 0.

[0078] In the step of comparing the preliminary images of the sintering machine trolley wheels detected in the first round with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels, the following method is used, including:

[0079] Preprocessing is performed on the preliminary images of the sintering machine trolley wheels from the first round of inspection and the preliminary images of the sintering machine trolley wheels from the second round of inspection to obtain preprocessed data.

[0080] By performing edge detection on the preprocessed data, the coordinates of each pixel in the preliminary image are obtained.

[0081] Based on the coordinates of each pixel in the preliminary image, the presence or absence of a wheel is determined using the Hough circle detection method. Image preprocessing primarily includes image filtering and color level adjustments to facilitate subsequent edge detection; the Laplacian operator is used to segment edges; the Laplacian operator is a type of operator that uses second-order differentiation, essentially measuring gradient divergence; Hough circle detection is mainly used to detect circles in the edge image; finally, wheel detection determines whether the detected Hough circle is a wheel, primarily based on the circle's size and vertical position.

[0082] The circle is expressed as equation (1), and the polar coordinates are expressed as equation (2).

[0083] (xa) 2 +(yb) 2 =r 2 (1)

[0084] x = a + r cos θ

[0085] y = b + r sinθ (2)

[0086] The principle of Hough circle detection is as follows: In a circular coordinate system, within the three-dimensional coordinate system formed by the abr and circle coordinates (A and B), a single point defines a unique circle. In the Cartesian xy-coordinate system, all circles passing through a given point are mapped to a three-dimensional curve in the Abr coordinate system. All circles passing through all non-zero pixels in the xy-coordinate system constitute many three-dimensional curves in the Abr coordinate system. Since the equations of all points on the same circle in the xy-coordinate system are identical, they map to the same point in the Abr coordinate system. Therefore, this point should have a total of [number] intersections with curves in the Abr coordinate system. By judging the cumulative number of intersections at each point in the Abr coordinate system, points exceeding a certain threshold are considered circles.

[0087] S103, Based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, compare the image of the wheel to be tested and the steel stamp number of the trolley to be tested to determine whether the wheel to be tested is a front wheel or a rear wheel, and obtain the image of the front wheel to be tested or the image of the rear wheel to be tested.

[0088] S104. Establish a coordinate system with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray that intersects the origin and is perpendicular to the ground as the Y-axis.

[0089] Multiple planes are marked between the infrared laser instrument and the wheel under test, perpendicular to the direction of the laser, and the imaging position of the laser line emitted by the infrared laser instrument in the image under test, as well as the corresponding coordinate point, are obtained.

[0090] The coordinates of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted.

[0091] In this method, the line structured light uses a standard plane to calibrate the distance from near to far. The calibration data is generated based on the distance range of the object being measured, with the calibration plane facing the infrared laser. If the distance range of the object being measured is [Lmin, Lmax], then data is recorded every 10mm within this range.

[0092] like Figure 3 As shown, this is a schematic diagram of the calibration plane in an embodiment of this application. L0, L1, ..., Ln are the imaging positions of the laser lines emitted by the infrared laser instrument in the image to be measured after the corresponding distance planes are set. In the subsequent measurement process, the distance value of any point in the laser line can be obtained by interpolation through the recorded data.

[0093] like Figure 4 The image shown is a simulated measurement image. It is assumed that 10 calibration planes at different distances were set up, with distance values ​​L1, L2, ..., L9. Figure 4 The perpendicular bisector represents the position of the laser line left on the corresponding plane; the curve represents the position of the laser line in the image during measurement.

[0094] Figure 4Let P1 and P2 be two points on the laser line. P1 lies between calibration planes L6 and L7, with image distance c from P1 to L6 and image distance d from P1 to L7. Then the distance from point P1 to the device is d1.

[0095]

[0096] According to equation (3), we can obtain Figure 4 The ranging information for the entire laser line is obtained. A coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis. Then the coordinates of point P1 are (x1, y1), where x1 = d1; y1 is the ordinate of point P1 on the image.

[0097] The data for the wheel section is selected based on the distance L between the wheel and the infrared laser emitter. All points in the image represent the wheel portion. Alternatively, the wheel position can be determined based on the detected position in the image. The portion of the wheel illuminated by the infrared laser is the scanned portion of the wheel. The wheel interval is denoted as [P3, P4].

[0098] Based on the distance points of the wheel section, a straight line is fitted, that is, a straight line is fitted to the measured cross-section of the wheel in the XY plane. The equation of the straight line is:

[0099] x=k*y+b (4)

[0100] Normally, the wheel is perpendicular to the ground, which means the slope is 0. Therefore, when the slope k is less than a certain threshold close to 0, it means that the wheel is in a normal state; when the slope k is too large, it means that the wheel is misaligned.

[0101] The degree of wheel tilt can be determined by the symmetry of the infrared scan lines in the image.

[0102] Based on the wheel's interval, the central symmetric point P0 can be determined. Then, it can be calculated whether P0P3 and P0P4 are vertically symmetric about the central point P0. First, the P0P4 segment is mapped symmetrically about the horizontal line containing P0. Then, the distance between the sampling points on the two lines is calculated at equal intervals vertically.

[0103] like Figure 6 As shown, it is based on Figure 5 The graph after symmetric processing of the line equation in P0P3 is denoted as follows: n points are sampled at equal intervals within the range of P0P3. Similarly, sampling n points at equal intervals within the P0P4 range is denoted as follows:

[0104] The wheel tilt error e is calculated as follows:

[0105]

[0106] in express The value of the point's coordinates in the x-direction.

[0107] The larger the tilt error 'e', ​​the more severe the wheel tilt. When 'e' exceeds a certain threshold, it indicates that the wheel is misaligned.

[0108] As can be seen from the above technical solutions, this application provides a fault detection method for the wheels of a sintering machine trolley. A preliminary image of the sintering machine trolley wheels in the first round of detection, along with the corresponding steel stamp number of the sintering machine trolley, is acquired using a visible light camera. Based on the preliminary image of the sintering machine trolley wheels in the first round of detection, a preliminary image of the sintering machine trolley wheels in the second round of detection is acquired after one sampling time. The preliminary images of the sintering machine trolley wheels in the first round of detection and the preliminary images of the sintering machine trolley wheels in the second round of detection are compared to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels in the second round of detection, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheels in the second round of detection, the preliminary image of the sintering machine trolley wheels in the second round of detection is stored as the image of the wheel to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested.

[0109] Based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, the image of the wheel to be tested and the steel stamp number of the trolley to be tested are compared to determine whether the wheel to be tested is a front wheel or a rear wheel, and the image of the front wheel to be tested or the image of the rear wheel to be tested are obtained; a coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis; multiple planes are marked between the infrared laser and the wheel to be tested in the direction perpendicular to the laser, and the imaging position of the laser line emitted by the infrared laser in the image to be tested, as well as the corresponding coordinate points, are obtained; the coordinate points of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted. The wheel fault detection method provided in this application embodiment can detect trolley wheel faults in a timely manner and provide alarm feedback, reducing the time the trolley runs with faults and reducing the workload of manual inspection in this part of the work. The equipment is relatively simple to install, has low maintenance costs, requires few supporting devices, and is stable and reliable.

[0110] See Figure 8 A framework diagram of a fault detection system for the wheels of a sintering machine trolley provided in an embodiment of this application.

[0111] Corresponding to the aforementioned embodiment of a fault detection method for sintering machine trolley wheels, this application also provides an embodiment of a fault detection system for sintering machine trolley wheels. For details not disclosed in the monitoring method provided in this application embodiment, please refer to the detection system provided in this application embodiment. The fault detection system for sintering machine trolley wheels includes: an acquisition module, a steel stamp number identification and storage module, a front and rear wheel determination module, and a wheel tilt determination module.

[0112] The acquisition module is used to acquire preliminary images of the wheels of the sintering machine trolley in the first round of detection, as well as the corresponding steel stamp number of the sintering machine trolley, through a visible light camera.

[0113] The steel stamp number identification and storage module is used to acquire a preliminary image of the sintering machine trolley wheels for the second round of detection after one sampling time, based on the preliminary image of the sintering machine trolley wheels detected in the first round. The module compares the preliminary image of the sintering machine trolley wheels detected in the first round with the preliminary image of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels detected in the second round, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheels detected in the second round, the preliminary image of the sintering machine trolley wheels detected in the second round is stored as the wheel image to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested.

[0114] The front and rear wheel determination module is used to compare the image of the wheel to be tested and the steel stamp number of the trolley to be tested based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, and determine whether the wheel to be tested is a front wheel or a rear wheel, thereby obtaining the image of the front wheel to be tested or the image of the rear wheel to be tested.

[0115] The wheel tilt determination module is used to establish a coordinate system with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis; to mark multiple planes between the infrared laser and the wheel under test in a direction perpendicular to the laser, and to obtain the imaging position of the laser line emitted by the infrared laser in the image under test, as well as the corresponding coordinate points; to fit the coordinate points of the wheel to obtain a fitted straight line, and to determine whether the wheel is tilted based on the slope of the fitted straight line.

[0116] The preliminary images of the sintering machine trolley wheels detected in the first round, along with the corresponding steel stamp number of the sintering machine trolley, are input into the acquisition module. The acquisition module then transmits these images and steel stamp numbers to the steel stamp number recognition and storage module. Based on the preliminary images of the sintering machine trolley wheels detected in the first round, the steel stamp number recognition and storage module acquires preliminary images of the sintering machine trolley wheels detected in the second round after one sampling period. The module compares the preliminary images of the sintering machine trolley wheels detected in the first round with those of the second round to determine whether the sintering machine trolley has wheels. If the preliminary image of the sintering machine trolley wheels detected in the second round shows no wheels, the sintering machine trolley stops running. If the preliminary image of the sintering machine trolley wheels detected in the second round shows wheels present, the preliminary image of the sintering machine trolley wheels detected in the second round is stored as the image of the wheel to be tested, and the wheel to be tested is... The corresponding sintering machine trolley steel stamp number is stored as the steel stamp number of the trolley to be tested. The steel stamp number identification and storage module transmits the image of the wheel to be tested and the steel stamp number of the trolley to be tested to the front and rear wheel determination module. The front and rear wheel determination module compares the image of the wheel to be tested and the steel stamp number of the trolley to be tested to determine whether the wheel to be tested is a front wheel or a rear wheel, and obtains the image of the front wheel to be tested or the image of the rear wheel to be tested. The front and rear wheel determination module transmits the image of the front wheel to be tested or the image of the rear wheel to be tested to the wheel tilt determination module, and establishes a coordinate system with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis. Multiple planes are marked between the infrared laser and the wheel to be tested in the direction perpendicular to the laser, and the imaging position of the laser line emitted by the infrared laser in the image to be tested and the corresponding coordinate points are obtained. The coordinate points of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted.

[0117] As can be seen from the above technical solutions, the embodiments of this application provide a fault detection method and system for the wheels of a sintering machine trolley. A preliminary image of the sintering machine trolley wheels in the first round of detection, along with the corresponding sintering machine trolley stamp number, is acquired using a visible light camera. Based on the preliminary image of the sintering machine trolley wheels in the first round of detection, a preliminary image of the sintering machine trolley wheels in the second round of detection is acquired after one sampling time. The preliminary images of the sintering machine trolley wheels in the first round of detection and the preliminary images of the sintering machine trolley wheels in the second round of detection are compared to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels in the second round of detection, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheels in the second round of detection, the sintering machine trolley stops running. A preliminary image of the sintering machine trolley wheel is stored as the image of the wheel to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the sintering machine trolley to be tested. Based on the image of the wheel to be tested and the steel stamp number of the sintering machine trolley to be tested, the image of the wheel to be tested and the steel stamp number of the sintering machine trolley to be tested are compared to determine whether the wheel to be tested is a front wheel or a rear wheel, and the image of the front wheel to be tested or the image of the rear wheel to be tested is obtained. A coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis. Multiple planes are marked between the infrared laser and the wheel to be tested in the direction perpendicular to the laser, and the imaging position of the laser line emitted by the infrared laser in the image to be tested, as well as the corresponding coordinate point, are obtained. The coordinate points of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted. The wheel fault detection method provided in this application embodiment can detect trolley wheel faults in a timely manner and provide alarm feedback, reducing the time the trolley runs with faults and reducing the amount of manual inspection work for this part. The equipment is relatively simple to install, has low maintenance costs, requires fewer supporting devices, and is stable and reliable.

[0118] Similar parts between the embodiments provided in this application can be referred to mutually. The specific implementation methods provided above are only a few examples under the overall concept of this application and do not constitute a limitation on the scope of protection of this application. For those skilled in the art, any other implementation methods extended from the solution of this application without creative effort shall fall within the scope of protection of this application.

Claims

1. A method for fault detection of sintering machine trolley wheels, characterized in that, include: Preliminary images of the sintering machine trolley wheels and their corresponding steel stamp numbers were obtained using a visible light camera during the first round of inspection. Based on the preliminary images of the sintering machine trolley wheels detected in the first round, after one sampling time, a preliminary image of the sintering machine trolley wheels detected in the second round is obtained. The preliminary images of the sintering machine trolley wheels detected in the first round are compared with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheels detected in the second round, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheels detected in the second round, the preliminary image of the sintering machine trolley wheels detected in the second round is stored as the wheel image to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested. Based on the image of the wheel to be tested and the steel stamp number of the trolley to be tested, the image of the wheel to be tested and the steel stamp number of the trolley to be tested are compared to determine whether the wheel to be tested is a front wheel or a rear wheel, and the image of the front wheel to be tested or the image of the rear wheel to be tested are obtained. A coordinate system is established with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray that intersects the origin and is perpendicular to the ground as the Y-axis. Between the infrared laser instrument and the wheel under test, multiple planes are marked in the direction perpendicular to the laser, and the imaging position of the laser line emitted by the infrared laser instrument in the image under test, as well as the corresponding coordinate point, are obtained. The coordinates of the wheel are fitted to obtain a fitted straight line, and the slope of the fitted straight line is used to determine whether the wheel is tilted. The specific steps for determining whether the wheel is tilted based on the slope of the fitted straight line are as follows: If the slope is 0 or less than a threshold close to 0, then the wheel does not tilt; If the slope is too steep, the wheel will tilt.

2. The method for fault detection of sintering machine trolley wheels according to claim 1, characterized in that, The sampling time can be obtained through a sampling time model, which is as follows: L represents the wheel spacing. v is the sampling time. i Let be the machine speed during the i-th sampling. The threshold is close to 0.

3. The method for fault detection of sintering machine trolley wheels according to claim 1, characterized in that, In the step of comparing the preliminary images of the sintering machine trolley wheels detected in the first round with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels, the following method is used, including: Preprocessing is performed on the preliminary images of the sintering machine trolley wheels from the first round of inspection and the preliminary images of the sintering machine trolley wheels from the second round of inspection to obtain preprocessed data; By performing edge detection on the preprocessed data, the coordinates of each pixel in the preliminary image are obtained; Based on the coordinates of each pixel in the preliminary image, the presence or absence of a wheel is determined using the Hough circle detection method.

4. The method for fault detection of sintering machine trolley wheels according to claim 3, characterized in that, The step of determining whether the wheel is tilted also includes: Based on the distance range [P3, P4] from the wheel to the infrared laser, the centrally symmetrical point P0 is obtained; Determine whether P0P3 and P0P4 are vertically symmetrical about the center point P0; Within the P0P3 range, n points are sampled at equal intervals, denoted as: Within the P0P4 range, n points are sampled at equal intervals, denoted as: The wheel tilt error e is calculated using the following model: in, express The x-coordinate of the point; The larger the tilt error 'e', ​​the more severe the wheel tilt. When 'e' exceeds a certain threshold, it indicates that the wheel is misaligned.

5. A fault detection system for the wheels of a sintering machine trolley, characterized in that, The fault detection system for the wheels of a sintering machine trolley is used to execute the fault detection method for the wheels of a sintering machine trolley as described in any one of claims 1-4, and includes an acquisition module, a steel stamp number identification and storage module, a front and rear wheel determination module, and a wheel tilt determination module. The acquisition module is used to acquire preliminary images of the sintering machine trolley wheels in the first round of detection, as well as the corresponding sintering machine trolley steel stamp number, through a visible light camera. The steel stamp number identification and storage module is used to obtain a preliminary image of the sintering machine trolley wheel in the second round of detection after one sampling time, based on the preliminary image of the sintering machine trolley wheel detected in the first round. The module compares the preliminary image of the sintering machine trolley wheel detected in the first round with the preliminary image of the sintering machine trolley wheel detected in the second round to determine whether the sintering machine trolley has wheels. If the trolley has no wheels in the preliminary image of the sintering machine trolley wheel detected in the second round, the sintering machine trolley stops running. If the wheels are present in the preliminary image of the sintering machine trolley wheel detected in the second round, the preliminary image of the sintering machine trolley wheel detected in the second round is stored as the wheel image to be tested, and the steel stamp number of the sintering machine trolley corresponding to the wheel to be tested is stored as the steel stamp number of the trolley to be tested. The front and rear wheel determination module is used to compare the image of the wheel to be tested and the steel stamp number of the trolley to be tested according to the image of the wheel to be tested and the steel stamp number of the trolley to be tested, and determine whether the wheel to be tested is a front wheel or a rear wheel, and obtain the image of the front wheel to be tested or the image of the rear wheel to be tested. The wheel tilt determination module is used to establish a coordinate system with the ray emitted by the infrared laser as the X-axis, the emission point as the origin, and the ray intersecting the origin and perpendicular to the ground as the Y-axis; to mark multiple planes between the infrared laser and the wheel under test in a direction perpendicular to the laser, and to obtain the imaging position of the laser line emitted by the infrared laser in the image under test, as well as the corresponding coordinate points; to fit the coordinate points of the wheel to obtain a fitted straight line, and to determine whether the wheel is tilted based on the slope of the fitted straight line. The specific steps for determining whether the wheel is tilted based on the slope of the fitted straight line are as follows: If the slope is 0 or less than a threshold close to 0, then the wheel does not tilt; If the slope is too steep, the wheel will tilt.

6. The fault detection system for sintering machine trolley wheels according to claim 5, characterized in that, The sampling time can be obtained through a sampling time model, which is as follows: L represents the wheel spacing. v is the sampling time. i Let be the machine speed during the i-th sampling. The threshold is close to 0.

7. A fault detection system for sintering machine trolley wheels according to claim 5, characterized in that, In the step of comparing the preliminary images of the sintering machine trolley wheels detected in the first round with the preliminary images of the sintering machine trolley wheels detected in the second round to determine whether the sintering machine trolley has wheels, the following method is used, including: Preprocessing is performed on the preliminary images of the sintering machine trolley wheels from the first round of inspection and the preliminary images of the sintering machine trolley wheels from the second round of inspection to obtain preprocessed data; By performing edge detection on the preprocessed data, the coordinates of each pixel in the preliminary image are obtained; Based on the coordinates of each pixel in the preliminary image, the presence or absence of a wheel is determined using the Hough circle detection method.

8. The fault detection system for sintering machine trolley wheels according to claim 4, characterized in that, The step of determining whether the wheel is tilted also includes: Based on the distance range [P3, P4] from the wheel to the infrared laser, the centrally symmetrical point P0 is obtained; Determine whether P0P3 and P0P4 are vertically symmetrical about the center point P0; Within the P0P3 range, n points are sampled at equal intervals, denoted as: Within the P0P4 range, n points are sampled at equal intervals, denoted as: The wheel tilt error e is calculated using the following model: in, express The x-coordinate of the point; The larger the tilt error 'e', ​​the more severe the wheel tilt. When 'e' exceeds a certain threshold, it indicates that the wheel is misaligned.

Images