Weld joint lossless tracking system and method

Abstract

The invention provides a welding seam lossless tracking system and method, relates to the technical field of welding seam position tracking, and is applied to a strip steel cold continuous rolling unit. The welding seam lossless tracking system comprises a visual collection module, a central processing server and a unit tracking control module, the visual collection module is used for collecting visual image data and sending the visual image data to the central processing server, the visual image data comprises first visual acquisition data, second visual acquisition data and third visual acquisition data; the central processing server is used for processing the visual image data based on a deep learning weld joint recognition algorithm to obtain an equipment control instruction and sending the equipment control instruction to the unit tracking control module; and the unit tracking control module is used for receiving the equipment control instruction and executing equipment control on the strip steel cold continuous rolling unit according to the equipment control instruction. According to the invention, high-precision and high-real-time automatic control of the cold continuous rolling unit is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of weld position tracking, in particular to a weld non-destructive tracking system and method. BACKGROUND

[0002] In the rolling process of a strip steel tandem cold rolling mill, by tracking the position of the weld in the mill, the rolling parameters can be adjusted in time to ensure the rolling quality. At present, the industry generally tracks by punching a round hole at the weld and setting a photoelectric probe at the key point of the production line to detect the hole. However, the punching behavior itself can cause damage to the base metal of the strip steel, stress concentration and work hardening. When this weak point is subjected to a large rolling force by the rolling mill, it is easy to cause a strip breakage accident. On the other hand, during the tracking of the weld, if temporary adjustment of the rolling parameters is needed, manual intervention is often required. Manual adjustment is highly dependent on the experience and subjective judgment of the operator, which can easily cause inconsistent control strategies and affect the stability of product quality. From the time the weld signal is detected to the time the operator receives the alarm information and manually modifies the process parameters, there is inevitably a time delay in the whole process. And the cold rolling production line is often in high-speed working condition, and the parameter adjustment cannot be timely effective. SUMMARY

[0003] The present application aims to solve at least one of the above problems.

[0004] To solve the above problems, the present application provides a weld non-destructive tracking system and method.

[0005] In a first aspect, the present application provides a weld non-destructive tracking system applied to a strip steel tandem cold rolling mill, the weld non-destructive tracking system comprising a visual acquisition module, a central processing server and a mill tracking control module, The visual acquisition module is configured to acquire visual image data and send the visual image data to the central processing server, wherein the visual image data comprises first visual acquisition data, second visual acquisition data and third visual acquisition data. The central processing server is configured to process the visual image data based on a deep learning weld identification algorithm to obtain a device control instruction and send the device control instruction to the mill tracking control module. The mill tracking control module is configured to receive the device control instruction and execute device control on the strip steel tandem cold rolling mill according to the device control instruction.

[0006] Optionally, the device control instruction comprises a weld quality control instruction, a weld correction control instruction and a rolling operation control instruction, and the processing of the visual image data to obtain a device control instruction comprises: the first visual acquisition data to obtain the weld appearance quality score; correcting the weld according to the second visual acquisition data to obtain the weld correction control instruction; obtaining a third visual acquisition weld position according to the third visual acquisition data; obtaining the rolling operation control instruction according to the third visual acquisition weld position to perform the rolling operation.

[0007] Optionally, the weld quality control instruction includes a welder rewelding instruction, a mill group tension and speed reduction instruction, and a mill group normal operation instruction, and the weld appearance quality score is obtained by comparing the weld image and a standard weld image. when the weld appearance quality score is less than or equal to a first threshold value, the welder rewelding instruction is obtained through the first visual acquisition weld position; when the weld appearance quality score is greater than the first threshold value and less than or equal to a second threshold value, the mill group tension and speed reduction instruction is obtained, wherein the mill group tension and speed reduction instruction is used to indicate that the strip cold continuous rolling mill group is controlled to reduce tension and speed according to a preset value; when the weld appearance quality score is greater than the second threshold value, the mill group normal operation instruction is obtained.

[0008] Optionally, the weld correction control instruction is obtained by correcting the weld according to the second visual acquisition data, including:

[0009] Optionally, the strip cold continuous rolling mill group includes a welder, an online loop, and a rolling mill connected in sequence, and the visual acquisition module includes a first visual acquisition unit, the first visual acquisition unit is arranged on a side of the welder close to the online loop, and the first visual acquisition unit is used to acquire the first visual acquisition data.

[0010] Optionally, the visual acquisition module further includes a second visual acquisition unit, the second visual acquisition unit is arranged on a side of the online loop close to the rolling mill, and the second visual acquisition unit is used to acquire the second visual acquisition data. ​​​​​

[0011] Optionally, the vision acquisition module further includes a third vision acquisition unit, which is located on the side of the rolling mill near the online looper, and is used to acquire the third vision acquisition data.

[0012] Optionally, the vision acquisition module further includes a fourth vision acquisition unit, which is located on the side of the rolling mill away from the online looper. The fourth vision acquisition unit is used to acquire fourth vision acquisition data, wherein the fourth vision acquisition data is used to evaluate the quality score after rolling.

[0013] Optionally, the central processing server and the unit tracking and control module are communicatively connected.

[0014] In a second aspect, the present invention provides a non-destructive welding seam tracking method, based on the non-destructive welding seam tracking system described in the first aspect, applied to a strip cold rolling mill; the non-destructive welding seam tracking method includes: The visual acquisition module acquires visual image data and sends the visual image data to the central processing server. The visual image data includes first visual acquisition data, second visual acquisition data, and third visual acquisition data. The central processing server uses a deep learning weld seam recognition algorithm to process the visual image data to obtain equipment control commands, and then sends the equipment control commands to the unit tracking and control module. The unit tracking control module receives the equipment control command and performs equipment control on the strip cold rolling mill according to the equipment control command.

[0015] The beneficial effects of the weld non-destructive tracking system and method of the present invention are as follows: The weld non-destructive tracking system acquires visual image data by setting a vision acquisition module, eliminating the need to punch circular holes at the weld and avoiding damage to the steel. First, second, and third visual acquisition data are acquired separately, and multiple visual acquisitions effectively improve the accuracy of identifying the position and quality of the strip steel weld, ensuring subsequent high-precision control. The visual image data is directly transmitted to the central processing server for analysis and processing, generating equipment control commands and sending them to the unit tracking control module, shortening the response time of traditional manual intervention and achieving high-precision, high-real-time automatic control of the cold rolling mill unit. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of a weld non-destructive tracking system according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a strip cold rolling mill according to an embodiment of the present invention; Figure 3This is a flowchart illustrating a non-destructive welding seam tracking method according to an embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0018] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0019] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0020] It should be noted that the terms "one" and "more" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0021] To address the problems existing in the aforementioned related technologies, this embodiment provides a non-destructive tracking system and method for weld seams.

[0022] like Figure 1 As shown in the figure, an embodiment of the present invention provides a weld seam non-destructive tracking system applied to a strip steel cold continuous rolling mill; the weld seam non-destructive tracking system includes a vision acquisition module 10, a central processing server 20, and a mill tracking control module 30. The visual acquisition module 10 is used to acquire visual image data and send the visual image data to the central processing server, wherein the visual image data includes first visual acquisition data, second visual acquisition data and third visual acquisition data; The central processing server 20 is used to process the visual image data based on a deep learning weld seam recognition algorithm to obtain equipment control commands, and then send the equipment control commands to the unit tracking control module. The unit tracking control module 30 is used to receive the equipment control command and perform equipment control on the strip cold rolling mill according to the equipment control command.

[0023] Specifically, the vision acquisition module collects visual image data. After the welding machine, it records the first visual acquisition data, i.e., its precise initial mechanical coordinates, as the system reference. At the looper exit, it acquires second visual acquisition data to identify the weld seam. The actual identified position is compared with the expected position calculated by the system based on linear velocity integration, and subsequent tracking positions are corrected to eliminate the cumulative error from the welding machine to the looper exit, ensuring positioning accuracy before the mill. At the mill inlet, it acquires third visual acquisition data for final identification, generating a highly reliable signal directly used for mill control. The central processing server receives the visual image data from the vision acquisition module, runs a deep learning-based weld seam recognition algorithm, and processes all position information uniformly to achieve collaborative tracking. The central server integrates all information and displays the latest weld seam position across the entire line in real time on the operation interface. The unit tracking control module receives equipment control commands from the central server and executes equipment control on the strip cold continuous rolling mill unit.

[0024] In this embodiment, the weld non-destructive tracking system acquires visual image data through a vision acquisition module, eliminating the need to punch circular holes at the weld and avoiding damage to the steel. First, second, and third visual acquisition data are obtained separately; multiple visual acquisitions effectively improve the accuracy of identifying the location and quality of the strip weld, ensuring subsequent high-precision control. The visual image data is directly transmitted to the central processing server for analysis and processing, generating equipment control commands and sending them to the unit tracking control module, shortening the response time of traditional manual intervention and achieving highly real-time automatic control of the cold rolling mill.

[0025] Optionally, the equipment control commands include weld quality control commands, weld correction control commands, and rolling operation control commands. The process of processing the visual image data to obtain the equipment control commands includes: The weld appearance quality is determined based on the first visual acquisition data to obtain the weld quality control instruction. Weld seam correction is performed based on the second visual acquisition data to obtain the weld seam correction control command; The location of the weld seam is obtained based on the third visual acquisition data. The rolling operation control command is obtained based on the weld position acquired by the third vision, so as to carry out the rolling operation.

[0026] Specifically, image recognition is performed on the first visual acquisition data to determine the weld appearance quality and obtain weld quality control instructions. Second visual acquisition data is used for further recognition and weld correction, resulting in weld correction control instructions. These instructions correct the weld position stored in the central processing server to obtain a more accurate actual weld position. Third visual acquisition data is used for final weld recognition, generating a high-precision trigger signal to control the rolling mill equipment to perform operations related to weld passage. Only weld recognition is required; weld quality is not judged. After weld recognition, this point is used as the final weld position to obtain rolling operation control instructions, executing weld rolling, FGC, and other rolling operations.

[0027] In some more specific embodiments, the weld seam image recognition algorithm employs a semantic segmentation or object detection deep learning model. Its input is the original image acquired by the visual unit, and its output is the precise pixel coordinates of the weld seam in the image, which are then converted into mechanical coordinates through calibration.

[0028] In this optional embodiment, image recognition and processing are performed using first-vision acquisition data, second-vision acquisition data, and third-vision acquisition data respectively, ensuring the accuracy of the final weld position identification. This allows for more precise achievement of each objective, thereby improving the overall system response speed and control precision.

[0029] Optionally, the weld quality control instructions include welding machine re-welding instructions, unit tension and speed reduction instructions, and unit normal operation instructions. The step of judging the weld appearance quality from the first visually acquired data to obtain the weld quality control instructions includes: The weld image and the weld position acquired by the first visual acquisition data are obtained based on the first visual acquisition data. The weld appearance quality score is obtained by comparing the weld image with a standard weld image. When the weld appearance quality score is less than or equal to the first threshold, the welding machine re-welds the weld position by the first visual acquisition. When the weld appearance quality score is greater than the first threshold and less than or equal to the second threshold, the unit tension reduction and speed reduction command is obtained, wherein the unit tension reduction and speed reduction command is used to control the strip cold rolling mill to reduce the tension and speed according to preset values; When the weld appearance quality score is greater than the second threshold, the unit is given a normal operation command.

[0030] Specifically, the initial image of the weld is acquired through first-vision data acquisition, the precise initial position of the weld is identified and recorded, and the weld appearance quality is judged simultaneously. Weld appearance quality mainly includes visual appearance aspects such as weld continuity and weld fullness. The weld image and a standard weld image are input into a pre-trained neural network model to extract image features, compare feature conditions, and score the weld image to obtain a weld appearance quality score R. When R is less than or equal to 30, the first threshold is 30, the weld quality level is 0 (poor weld quality), and a re-welding command is activated to re-weld the steel. When the R score is greater than 30 but less than or equal to 60, the second threshold is 60, the weld quality level is 1 (relatively poor weld quality), and a weld protection program is activated, reducing the tension and speed of the unit. By controlling and reducing the speed of each frame motor, the entire production line operates at a lower linear speed (e.g., from 800 m / min to 400–600 m / min). Tension is established by the inlet / outlet tension rollers, looper, and speed differences between frames. Tension reduction is achieved by adjusting the torque setpoint of the tension rollers and temporarily relaxing the target value of the tension closed-loop control. When R is greater than 60, the weld quality level is 2, indicating good weld quality, and the unit operates normally, receiving the normal operation command.

[0031] In this optional embodiment, a quantifiable weld appearance quality score is generated by comparing the actual weld image with a standard weld image, replacing the traditional method that relies on manual visual inspection or simple threshold judgment, significantly improving the objectivity and consistency of quality assessment. A graded response mechanism is established to enhance the system's intelligent decision-making capabilities; if the weld defect is severe, a re-welding command is triggered, eliminating high-risk defects at the source. When the weld quality is moderate, proactively implementing tension and speed reduction measures can significantly reduce the risk of strip steel breakage due to insufficient weld strength as it passes through the rolling mill.

[0032] Optionally, weld seam correction is performed based on the second visual acquisition data to obtain the weld seam correction control command, including: The location of the weld seam is obtained based on the second visual acquisition data; The positional deviation is obtained by acquiring the weld position through the second visual acquisition and by calculating the weld position through the system. The weld seam correction control command is obtained through the positional deviation.

[0033] Specifically, the weld seam is re-identified using data acquired through a second vision system. Based on the deviation between the cumulative running length of the strip and its initial position, the positional error of the weld seam before entering the rolling mill is calibrated. This weld seam position calibration resets the weld seam position calculated in the system, which is based on the number of rotations of the tension roll. This weld seam identification corrects the weld seam position in the control system. The system's calculated weld seam position is based on the weld seam position acquired through the first vision system and the strip's moving speed.

[0034] Optionally, such as Figure 2 As shown, the strip cold rolling mill includes a welding machine, a looper, and a rolling mill connected in sequence; the vision acquisition module includes a first vision acquisition unit, which is located on the side of the welding machine near the looper, and is used to acquire the first vision acquisition data.

[0035] Specifically, the first vision acquisition unit, located behind the welding machine, is used to image the weld immediately after welding is completed, obtain its absolute initial position, and at the same time make a preliminary assessment of the weld formation quality.

[0036] Optionally, such as Figure 2 As shown, the vision acquisition module further includes a second vision acquisition unit, which is located on the side of the online looper near the rolling mill. The second vision acquisition unit is used to acquire the second vision acquisition data.

[0037] Specifically, the second vision acquisition unit, located at the outlet of the looper, is used for secondary identification of the weld after it leaves the looper and before entering the rolling mill. Its core function is to perform final calibration of the weld's position before it enters the rolling mill, eliminating the cumulative error in weld position calculation caused by strip slippage and stretching during the distance from the welding mill to the looper outlet. Simultaneously, it sends preset parameters to the rolling mill from this location. Optionally, such as Figure 2 As shown, the vision acquisition module also includes a third vision acquisition unit, which is located on the side of the rolling mill near the looper. The third vision acquisition unit is used to acquire the third vision acquisition data.

[0038] Specifically, the third vision acquisition unit, located at the mill inlet, is used for final precise positioning of the weld before it enters the first stand. The trigger signal here has the highest priority and is used to precisely initiate or fine-tune specific rolling procedures such as speed reduction and tension adjustment in the mill area, ensuring the weld safely passes through the mill.

[0039] Optionally, such as Figure 2 As shown, the vision acquisition module further includes a fourth vision acquisition unit, which is located on the side of the rolling mill away from the online looper. The fourth vision acquisition unit is used to acquire fourth vision acquisition data, which is used to evaluate the quality score after rolling.

[0040] Specifically, the fourth vision acquisition unit, located at the mill exit, is used to inspect the surface of the strip (including the weld area) after it passes through the mill. This can both assess the quality after rolling and verify the accuracy of the tracking system.

[0041] In some more specific embodiments, the first vision acquisition unit, the second vision acquisition unit, the third vision acquisition unit, and the fourth vision acquisition unit all include an industrial camera, a lens, and a lighting system, and all vision units within the system are synchronized through a unified encoder signal and a network clock to ensure the consistency of position information calculation.

[0042] Optionally, the central processing server and the unit tracking and control module are communicatively connected.

[0043] Specifically, the central processing server has a built-in weld seam recognition algorithm. This algorithm is based on a deep learning model and is trained to focus on recognizing the visual features of the weld seam from the natural texture of the strip steel substrate, thus achieving weld seam location without any physical markers.

[0044] like Figure 3 As shown, an embodiment of the present invention provides a non-destructive welding seam tracking method, based on the non-destructive welding seam tracking system described above, applied to a strip cold rolling mill; the non-destructive welding seam tracking method includes: Step 310: The visual acquisition module acquires visual image data and sends the visual image data to the central processing server. The visual image data includes first visual acquisition data, second visual acquisition data, and third visual acquisition data. Step 320: The central processing server processes the visual image data based on a deep learning weld seam recognition algorithm to obtain equipment control commands, and sends the equipment control commands to the unit tracking control module. Step 330: The unit tracking control module receives the equipment control command and performs equipment control on the strip cold rolling mill according to the equipment control command.

[0045] The weld non-destructive tracking method of this embodiment is used to implement the weld non-destructive tracking system as described above. Its advantages over the prior art are the same as those of the weld non-destructive tracking system compared to the prior art, and will not be repeated here.

[0046] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A non-destructive welding seam tracking system, characterized in that, Applied to cold continuous strip rolling mills; the weld non-destructive tracking system includes a vision acquisition module, a central processing server, and a mill tracking control module. The visual acquisition module is used to acquire visual image data and send the visual image data to the central processing server, wherein the visual image data includes first visual acquisition data, second visual acquisition data and third visual acquisition data; The central processing server is used to process the visual image data based on a deep learning weld seam recognition algorithm to obtain equipment control commands, and then send the equipment control commands to the unit tracking control module. The unit tracking control module is used to receive the equipment control instructions and perform equipment control on the strip cold rolling mill according to the equipment control instructions.

2. The weld non-destructive tracking system according to claim 1, characterized in that, The equipment control commands include weld quality control commands, weld correction control commands, and rolling operation control commands. The process of processing the visual image data to obtain the equipment control commands includes: The weld appearance quality is determined based on the first visual acquisition data to obtain the weld quality control instruction. Weld seam correction is performed based on the second visual acquisition data to obtain the weld seam correction control command; The location of the weld seam is obtained based on the third visual acquisition data. The rolling operation control command is obtained based on the weld position acquired by the third vision, so as to carry out the rolling operation.

3. The weld non-destructive tracking system according to claim 2, characterized in that, The weld quality control instructions include welding machine re-welding instructions, unit tension and speed reduction instructions, and unit normal operation instructions. The step of judging the weld appearance quality from the first visually acquired data to obtain the weld quality control instructions includes: The weld image and the weld position acquired by the first visual acquisition data are obtained based on the first visual acquisition data. The weld appearance quality score is obtained by comparing the weld image with a standard weld image. When the weld appearance quality score is less than or equal to the first threshold, the welding machine re-welds the weld position by the first visual acquisition. When the weld appearance quality score is greater than the first threshold and less than or equal to the second threshold, the unit tension reduction and speed reduction command is obtained, wherein the unit tension reduction and speed reduction command is used to control the strip cold rolling mill to reduce the tension and speed according to preset values; When the weld appearance quality score is greater than the second threshold, the unit is given a normal operation command.

4. The weld non-destructive tracking system according to claim 2, characterized in that, Weld seam correction is performed based on the second visual acquisition data to obtain the weld seam correction control command, including: The location of the weld seam is obtained based on the second visual acquisition data; The positional deviation is obtained by acquiring the weld position through the second visual acquisition and by calculating the weld position through the system. The weld seam correction control command is obtained through the positional deviation.

5. The weld non-destructive tracking system according to claim 1, characterized in that, The strip cold rolling mill includes a welding machine, a looper, and a rolling mill connected in sequence; the vision acquisition module includes a first vision acquisition unit, which is located on the side of the welding machine near the looper, and is used to acquire the first vision acquisition data.

6. The weld non-destructive tracking system according to claim 5, characterized in that, The vision acquisition module further includes a second vision acquisition unit, which is located on the side of the online looper close to the rolling mill. The second vision acquisition unit is used to acquire the second vision acquisition data.

7. The weld non-destructive tracking system according to claim 5, characterized in that, The vision acquisition module also includes a third vision acquisition unit, which is located on the side of the rolling mill near the looper. The third vision acquisition unit is used to acquire the third vision acquisition data.

8. The weld non-destructive tracking system according to claim 5, characterized in that, The vision acquisition module further includes a fourth vision acquisition unit, which is located on the side of the rolling mill away from the online looper. The fourth vision acquisition unit is used to acquire fourth vision acquisition data, which is used to evaluate the quality score after rolling.

9. The weld non-destructive tracking system according to claim 1, characterized in that, The central processing server and the unit tracking and control module are connected in communication.

10. A non-destructive welding seam tracking method, based on the non-destructive welding seam tracking system according to any one of claims 1 to 9, characterized in that, Applications in cold continuous strip rolling mills; The non-destructive welding tracking method includes: The visual acquisition module acquires visual image data and sends the visual image data to the central processing server. The visual image data includes first visual acquisition data, second visual acquisition data, and third visual acquisition data. The central processing server uses a deep learning weld seam recognition algorithm to process the visual image data to obtain equipment control commands, and then sends the equipment control commands to the unit tracking and control module. The unit tracking control module receives the equipment control command and performs equipment control on the strip cold rolling mill according to the equipment control command.

Images