High-precision double-seam welding robot

By constructing a virtual centerline and dynamically adjusting welding parameters using a high-precision double-seam welding robot, the problems of repetitive positioning error and asymmetrical thermal deformation in existing equipment have been solved, achieving high-precision and highly symmetrical double-sided welding, thus improving production efficiency and quality.

CN121798261APending Publication Date: 2026-04-07EZHOU KEBEI LASER CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-02-10
Publication Date
2026-04-07

AI Technical Summary

Technical Problem

Existing welding equipment cannot meet the requirements of high-precision, high-symmetry, and adaptive double-seam welding, and suffers from problems such as repeated positioning errors, asymmetrical thermal deformation, and high equipment costs.

Method used

A high-precision double-seam welding robot is adopted, which uses a rotary table and laser positioning module to construct a virtual center line. Combined with a vision module to monitor the weld gap, it can realize synchronous welding of a single robot. Furthermore, thermal strain compensation is performed by dynamically adjusting welding parameters, and a rapid online calibration system is established.

Benefits of technology

It achieves high-precision symmetrical welding of double-sided welds, eliminates repeated positioning errors, avoids weld penetration and excessive weld height, and improves equipment utilization and production quality consistency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121798261A_ABST
    Figure CN121798261A_ABST
Patent Text Reader

Abstract

The high-precision double-seam welding robot comprises a robot body, a control system and a rotating seat, the rotating seat is long, a welding gun is installed at the end, away from the robot body, of the rotating seat, and the rotating axis of the rotating seat is configured to be orthogonal to the welding face of a workpiece; when the rotary seat is in the 0-degree posture, the welding gun inclines outwards to point to a first welding seam on one side of the workpiece; when the rotary seat rotates by 180 degrees to a reverse posture, the welding gun inclines outwards to point to a second welding seam on the opposite side of the workpiece; and the control system is configured to obtain the spatial positions of the first welding seam and the second welding seam by using the laser positioning module, construct a geometric center line of the two welding seams as a virtual median line, and control the robot body to drive the rotation axis of the rotary seat to intersect with the virtual median line and the length center line to be parallel to the virtual median line. Single-machine synchronous high-precision welding of welding seams on the two sides can be achieved, and repeated positioning errors can be eliminated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the technical field of welding equipment for large workpieces, and in particular to a high-precision double-seam welding robot. Background Technology

[0002] With the rapid development of heavy equipment manufacturing, shipbuilding, pressure vessels, and other fields, higher requirements have been placed on the welding quality and efficiency of double-sided symmetrical welds in thick plate structural components. Existing double-seam welding technology mainly adopts the following two schemes: The first approach is a single-robot welding scheme. After the robot completes the first side weld, a positioner flips the workpiece or moves the robot to the opposite side to weld the second side. This scheme has a fundamental flaw: the two independent trajectory planning and positioning processes inevitably introduce repetitive positioning errors. For large workpieces exceeding 10 meters in length, the secondary positioning error can reach ±2-5mm, leading to defects such as asymmetrical weld penetration and weld bead misalignment. More seriously, the thermal deformation of the first side weld alters the actual spatial position of the second side weld, but the robot still welds according to the theoretical trajectory, resulting in incomplete penetration or burn-through rates as high as 15%-20%, leading to extremely high rework costs.

[0003] The second approach is a dual-robot synchronous welding scheme. While this solves the problem of repetitive positioning, the coordinated control of the two robots is extremely complex, and the two independent control systems struggle to achieve true motion coupling and synchronous compensation for thermal deformation. Even slight differences in welding speed and heat input can lead to irreversible torsional deformation of the workpiece due to asymmetric thermal strain on both sides. Furthermore, the dual-robot scheme increases equipment costs by over 80%, imposes stringent requirements on workshop space layout, and is economically inefficient.

[0004] For example, Chinese patent application CN202411261731.1 in the related technology proposes an automatic welding device for large workpieces, including a ground rail, a worktable, a cantilever, a welding robot arm, and a cleaning device. The ground rail includes a base frame, guide rails, and levelers. Levelers are installed at the bottom of the base frame, and parallel guide rails are installed at the top of both sides of the base frame. The worktable is slidably mounted on the guide rails via a slider. A servo geared motor is installed on the worktable, which drives the worktable to move on the ground rail. The bottom and sides of the base frame are fitted with protective plates to cover the ground rail. One end of the cantilever is fixed to the worktable, and the welding robot arm is installed at the other end of the cantilever. The worktable is equipped with a welding machine, a welding wire hopper, a water tank, and a control cabinet. A line laser sensor for identifying the position and direction of the weld is installed at the end of the welding robot arm.

[0005] This demonstrates that existing welding equipment cannot meet the demands of high-precision, highly symmetrical, and adaptive double-seam welding, severely hindering the improvement of quality in high-end equipment manufacturing. Therefore, a welding robot system capable of solving these problems is urgently needed. Summary of the Invention

[0006] To address the issue that existing welding equipment cannot meet the requirements of high-precision, high-symmetry double-seam welding, this application provides a high-precision double-seam welding robot.

[0007] This application provides a high-precision double-seam welding robot using the following technical solution: A high-precision double-seam welding robot includes a robot body, a control system, a rotary seat located at the end of the robot body, and a welding torch, a vision module, a laser positioning module, and a welding fume extraction module mounted on the rotary seat. The rotary seat is elongated and the welding torch is mounted at the end of the rotary seat away from the robot body. The rotation axis of the rotary seat is configured to be orthogonal to the welding surface of the workpiece. When the rotary table is in the 0-degree position, the welding torch is tilted outwards and points to the first weld seam on one side of the workpiece; when the rotary table is rotated 180 degrees to the reverse position, the welding torch is tilted outwards and points to the second weld seam on the opposite side of the workpiece. The control system is configured to: use the laser positioning module to obtain the spatial positions of the first weld and the second weld, construct the geometric center lines of the two welds as virtual midline, and control the robot body to drive the rotation axis of the rotary seat to intersect with the virtual midline and the length midline to be parallel to the virtual midline.

[0008] Furthermore, the control system is configured as follows: Before welding begins, the laser positioning module is controlled to sequentially scan multiple feature points P of the first weld seam. 11 ...P 1n Multiple feature points P of the second weld 21 ...P 2n And calculate multiple line segments P 11 P 21 ...P 1n P 2n midpoint P m1 ...P mn With multiple midpoints P m1 ...P mn Construct the virtual median; Throughout the welding process, the robot body is controlled to constrain the center line of the flange mounted on the robot body to the virtual center line in real time.

[0009] Furthermore, the vision module monitors the real-time gap width at the root of the second weld seam in front of the welding torch when the rotary table is in a 180-degree position for welding the second weld seam. The control system compares the real-time gap width with the initial gap width monitored by the vision module during the welding of the first weld, and calculates the thermal strain difference ΔW. like The control system increases the wire feeding speed and oscillation amplitude of the welding torch to increase the weld metal deposition cross-sectional area. like The control system reduces the wire feeding speed of the welding torch to prevent the weld reinforcement from exceeding the standard.

[0010] Furthermore, if the thermal strain difference ΔW calculated by the control system exceeds a preset threshold, the control system controls the welding torch to stop feeding wire and controls the welding fume extraction module to increase the extraction power.

[0011] Furthermore, the control system is also configured to: At time t during the welding process of the second weld, the virtual median line is translated and rotated along the half-value vector direction of ΔW to generate the dynamic virtual median line at time t, and the robot body is controlled to correct the position of the rotary seat.

[0012] Furthermore, the welding robot also includes: The robot body moves on the guide rail. The base is used to install the guide rail and to set up tooling for fixing the workpiece to be welded. Two calibration blocks are provided and distributed at both ends of the guide rail. Each calibration block has two calibration points with known distances. The control system is also configured to: before each welding operation, control the rotary table to move so that the laser positioning module is aligned with two calibration points on the same calibration block, so as to correct and update the virtual median coordinate system.

[0013] Furthermore, the welding torch is located between the vision module and the welding fume extraction module. When the rotary seat is in the 0-degree and 180-degree positions, the vision module is located in front of the welding torch in the direction of travel during the welding process.

[0014] Furthermore, a three-axis adjustment mechanism is installed at the end of the rotary seat away from the robot body, and an angle adjustment mechanism is installed at the output end of the three-axis adjustment mechanism. The welding torch is installed at the output end of the angle adjustment mechanism.

[0015] In summary, the beneficial technical effects of this application are as follows: 1. Achieve single-machine synchronous high-precision welding of double-sided welds and eliminate repeated positioning errors; by switching the attitude of the long rotary seat at 180° and constraining the virtual center line, the welding task of double-sided welds is decoupled into continuous path tracking of a single robot body, avoiding the cumulative error of ±2 to 5 mm caused by two positioning in the traditional solution, and the symmetry of the double-sided welds can be controlled within ±0.3 mm, effectively solving the core defects of complex dual-robot coordination and low repeated positioning accuracy of single robot in the existing technology; 2. By leveraging a real-time thermal deformation sensing and welding parameter adaptive adjustment mechanism, quality control issues can be avoided. This application monitors the change in the width ΔW of the second weld gap and dynamically adjusts the wire feeding speed and oscillation amplitude to achieve real-time compensation for thermal deformation. Compared to existing technologies that can only preset fixed process parameters, this application achieves closed-loop control of thermal strain during the welding process through "measurement-feedback-adjustment," effectively avoiding defects such as weld burn-through and excessive weld reinforcement, thus overcoming the limitations of passive execution in traditional welding robots. 3. A dynamic virtual median line compensation algorithm is proposed to realize intelligent correction of welding trajectory; the compensation method of translating and rotating the virtual median line along the direction of ΔW half-value vector decouples the complex three-dimensional thermal deformation into a single-direction correction amount, which has low computational cost but high compensation accuracy; this "single-to-double" compensation idea solves the technical problem of synchronous compensation of thermal deformation on both sides. 4. Establish a rapid online calibration system to ensure the long-term accuracy and stability of the system; using calibration blocks at both ends of the guide rail, coordinate system drift is automatically corrected before each operation, without the need for external measuring equipment, and the calibration accuracy can reach ±0.05mm; compared with the existing technology that requires manual calibration every shift, this application significantly improves the equipment utilization rate and effectively suppresses the long-term accuracy decay caused by mechanical wear, temperature drift and other factors, ensuring the quality consistency of mass production. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the swivel seat in the 0-degree position according to an embodiment of this application; Figure 2 This is a front view of the overall structure of an embodiment of this application; Figure 3 This is a structural schematic diagram of the rotary seat in the 180-degree position according to an embodiment of this application; Figure 4 This is a control logic diagram of the control system according to an embodiment of this application.

[0017] Explanation of reference numerals in the attached figures: 1. Robot body; 2. Rotary seat; 31. Welding torch; 32. Vision module; 33. Laser positioning module; 34. Welding fume extraction module; 41. First weld; 42. Second weld; 5. Virtual median; 61. Base; 62. Guide rail; 7. Calibration block; 71. Calibration point; 81. Three-axis adjustment mechanism; 82. Angle adjustment mechanism. Detailed Implementation

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

[0019] This application discloses a high-precision double-seam welding robot. (Refer to...) Figure 1 , Figure 2 and Figure 3 It includes a robot body 1, a control system, a rotary base 2 located at the end of the robot body 1, and a welding torch 31, a vision module 32, a laser positioning module 33, and a welding fume extraction module 34 mounted on the rotary base 2. Among them, the robot body 1, the vision module 32, the laser positioning module 33, and the welding fume extraction module 34 are all existing technologies, which can be fully implemented by those skilled in the art, and need not be elaborated.

[0020] Specifically, the robot body 1 is a five-axis robot, the rotary seat 2 is long and the welding torch 31 is installed at the end of the rotary seat 2 away from the robot body 1. The rotation axis of the rotary seat 2 is configured to be orthogonal to the welding surface of the workpiece. The rotary seat 2 is installed on the rotary servo of the sixth axis of the robot body 1 through the mounting flange. The rotary servo has a built-in encoder, which can accurately identify the rotation angle of the rotary seat 2.

[0021] When the rotary seat 2 is in a 0-degree position, the welding torch 31 tilts outwards towards the first weld seam 41 on one side of the workpiece, as shown. Figure 1 As shown; when the rotary seat 2 rotates 180 degrees to the reverse posture, the welding torch 31 tilts outwards towards the second weld seam 42 on the opposite side of the workpiece, as shown. Figure 3 As shown; it should be noted that the workpiece mainly refers to large thick plate workpieces such as H-beams and box girders with parallel double weld seams.

[0022] The control system is configured to: use the laser positioning module 33 to obtain the spatial position of the first weld 41 and the second weld 42, construct the geometric center line of the two welds as the virtual center line 5, and control the robot body 1 to drive the rotation axis of the rotary seat 2 to intersect with the virtual center line 5 and the length center line to be parallel to the virtual center line 5.

[0023] Specifically, refer to Figure 1 and Figure 4 The control system is configured as follows: Before welding begins, the laser positioning module 33 is controlled to sequentially scan multiple feature points P of the first weld seam 41. 11 ...P 1n More than 42 feature points P of the second weld 21 ...P 2n And calculate multiple line segments P11 P 21 ...P 1n P 2n midpoint P m1 ...P mn With multiple midpoints P m1 ...P mn Construct a virtual median line 5; Throughout the welding process, the robot body 1 controls the center line of the flange mounted on the rotary seat 2 on the robot body 1 to be constrained in real time to the virtual center line 5, and makes the length direction of the rotary seat 2 parallel to the virtual center line 5.

[0024] Furthermore, when the rotary table 2 is in a 180-degree position while welding the second weld seam 42, the vision module 32 monitors the real-time gap width at the root of the second weld seam 42 in front of the welding torch 31. The control system compares the real-time gap width with the initial gap width monitored by the vision module 32 when the first weld 41 is welded, and calculates the thermal strain difference ΔW. like The control system increases the wire feeding speed and oscillation amplitude of the welding torch 31 to increase the weld metal deposition cross-sectional area. like The control system reduces the wire feeding speed of the welding torch 31 to prevent the weld reinforcement from exceeding the standard.

[0025] Furthermore, if the thermal strain difference ΔW calculated by the control system exceeds the preset threshold, the control system controls the welding torch 31 to stop feeding wire and controls the welding fume extraction module 34 to increase the extraction power.

[0026] In addition, the control system is also configured as follows: At time t during the welding process of the second weld 42, the virtual median line 5 is translated and rotated along the half-value vector direction of ΔW to generate the dynamic virtual median line 5 at time t, and the robot body 1 is controlled to correct the position of the rotary seat 2. Specifically, the control system translates the virtual median line 5 by a distance of |ΔW / 2| along the direction of change of ΔW (i.e., the normal direction perpendicular to the weld tangent).

[0027] Therefore, before welding, the control system drives the robot body 1, so that the laser positioning module 33 on the rotary seat 2 scans the first weld 41 sequentially along the direction of the first weld 41 with a step size Δx (preferably 20-50mm), and obtains n feature points P. 11 (x 11 ,y 11 ,z 11 ...P 1n (x 1n ,y 1n ,z 1nSubsequently, the control system controls the rotary table 2 to rotate 180°, sweeping across the second weld 42 at the same time length, and obtaining P. 21 …P 2n Each feature point contains three-dimensional coordinates and information about the weld root gap width W.

[0028] Then, the control system calculates the midpoint P of the corresponding point pair in real time. mi = (P 1i + P 2i ) / 2, i=1…n; In practice, vector operations are used: P mi = ((x 1i +x 2i ) / 2, (y 1i +y 2i ) / 2, (z 1i +z 2i ) / 2), forming the midpoint sequence P m1 …P mn Then, a cubic spline curve is fitted to the midpoint sequence to generate a continuous and smooth virtual median line 5, whose parametric equation is represented by L(u), u∈[0,1]. This virtual median line 5 is stored in the controller memory of the control system as a reference for subsequent path planning.

[0029] Therefore, during the welding process, the control system calculates the coordinates (x, y) of the center point F of the flange at the end of the robot body 1 (i.e., the mounting center of the rotary seat 3) in real time through inverse kinematics calculations. F , y F , z F The position error vector e is generated by taking the difference between the coordinates of the nearest point on the virtual median line L(u) and the coordinates of the nearest point on the virtual median line L(u), ignoring the z-axis value in the vertical direction.

[0030] And a PID closed-loop control law is adopted: ; in: , , This is the gain coefficient. The robot's speed command is corrected. The sampling period T is preferably 10ms to ensure that the real-time tracking accuracy of the x and y axes of the installation flange center point F is within ±0.1mm, so that the rotation axis of the rotary seat 2 is constrained to the virtual center line in real time.

[0031] Therefore, during the welding process of the first weld 41, the control system controls the robot body 1 to constrain the rotation axis of the rotary seat 2 to the virtual center line 5 in real time and to walk along the first weld 41. On the other hand, the control system also controls the vision module 32 to collect the initial gap width of the unwelded part of the first weld 41. Since the workpiece corresponding to this application is a thick plate structure with symmetrical welds on both sides, the initial gap width of the unwelded part of the first weld 41 can be regarded as the initial gap width of the second weld 42 in the cold state.

[0032] During the welding process of the second weld 42, the thermal deformation generated by the first weld 41 during the welding process will inevitably change the actual spatial position of the second weld 42. Therefore, the vision module 32 monitors the real-time gap width of the unwelded part of the second weld 42 in real time, and then the control system compares it with the initial gap width monitored by the vision module 32 when the first weld 41 is welded. The resulting thermal strain difference value ΔW can be used to characterize the gap change of the second weld 42 under the influence of thermal strain, and at the same time predict the local failure of the virtual median line 5 calculated before welding.

[0033] For example, when This indicates that the gap of the second weld 42 has widened. The control system increases the wire feed speed of the welding torch 31 and simultaneously increases the oscillation amplitude of the welding torch 31 to ensure that the cross-sectional area of ​​the deposited metal matches the increased bevel volume. At the same time, the control system also corrects the virtual median line 5 in a direction away from the second weld 42, specifically by shifting it away from the second weld 42. | distance.

[0034] And when This indicates that the gap of the second weld 42 has narrowed, and the control system reduces the wire feed speed of the welding torch 31 to prevent excessive weld reinforcement. Simultaneously, the control system also corrects the virtual center line 5 towards the direction closer to the second weld 42, specifically by shifting it towards the direction closer to the second weld 42. | distance.

[0035] Furthermore, when ΔW exceeds the preset threshold, the control system determines that thermal deformation is out of control or the workpiece clamping is abnormal, and immediately outputs a wire-stop signal to interrupt the wire feeding of the welding torch 31. At the same time, the control system also increases the suction power of the welding fume extraction module 34 to 150% of the rated value through the PWM signal. On the one hand, it can quickly remove the fume to protect the optical path of the vision module 32 and prevent weld oxidation; on the other hand, it can also accelerate the heat exchange in the weld area and accelerate the cooling of the welding heat accumulation area by means of the suction airflow.

[0036] Therefore, the long rotary seat 2 in the above scheme achieves "one gun for two uses" by rotating 180°, transforming the welding task of the double-sided weld seam into a single trajectory tracking problem of the single-sided robot body 1, fundamentally eliminating the complexity of dual-robot coordination and repetitive positioning errors.

[0037] Furthermore, the virtual median line 5, as a mathematical abstraction of geometric coupling, binds the two spatially separated weld seams into a unified path reference. This allows the robot body 1 to ensure that the rotation axis of the rotary seat 2 intersects the virtual median line 5 in real time and that the length centerline of the rotary seat 2 is parallel to the virtual median line 5 in real time, thus ensuring that the welding torch 31 automatically aligns with the weld seams on both sides in both postures, reducing the dimensionality and computational complexity of path planning.

[0038] In addition, the amount of thermal deformation is inferred by monitoring the gap change of the second weld 42 through the vision module 32, and the deformation state of both sides is inferred by the information of one side. Then, forward compensation is performed through the dynamic virtual median line 5. This compensation strategy of "using one side to replace the two sides" avoids the high cost of independent monitoring of both sides and the problem of data synchronization.

[0039] Furthermore, to eliminate coordinate system drift caused by long-term operation, referencing Figure 1 , Figure 2 and Figure 3 Welding robots also include: The base 61 and the guide rail 62 are used to install the guide rail 62 and to set up tooling for fixing the workpiece to be welded. The robot body 1 walks on the guide rail 62. The calibration block 7 has two parts, which are distributed at both ends of the guide rail 62. The calibration block 7 has two calibration points 71 with known distances. The control system is also configured to: before each welding operation, control the rotary table 2 to move so that the laser positioning module 33 is aligned with two calibration points 71 on the same calibration block 7, so as to correct and update the coordinate system of the virtual center line 5.

[0040] Furthermore, the welding torch 31 is located between the vision module 32 and the welding fume extraction module 34. When the rotary seat 2 is in the 0-degree and 180-degree positions, the vision module 32 is always located in front of the welding torch 31 during the welding process. This ensures that the vision module 32 can collect weld information in advance, realizing the predictive control of "detecting first, then welding"; while the fume extraction module is always located behind or to the side to avoid interfering with the optical path of the vision module 32.

[0041] Therefore, before each welding operation, the control system moves the robot body 1 to the near end of the guide rail 62 and controls the rotary seat 2 to rotate to a 0-degree position. The laser positioning module 33 emits a cross laser, and manual or automatic fine-tuning is performed to align the laser center with calibration point A 71 on the calibration block 7. The current joint angle θ of the robot body 1 is recorded. A Then, the robot body 1 is controlled to translate, so that the cross laser is aligned with calibration point B 71 on the same calibration block 7, and the joint angle θ is recorded. B According to θ A θ BThe actual position of the laser positioning module 33 in the basic coordinate system of the robot body 1 is calculated and compared with the theoretical position to obtain the coordinate system error ΔT. Then, the coordinate transformation matrix of the virtual median line 5 is corrected using ΔT to update the coordinate system parameters of the current welding operation. The remote calibration block 7 repeats the above process to verify the calibration accuracy. This ensures the long-term accuracy and stability of the system without the need for additional measuring equipment.

[0042] In addition, it should be noted that, referring to Figure 1 and Figure 2 A three-axis adjustment mechanism 81 is installed at the end of the rotary base 2 furthest from the robot body 1. An angle adjustment mechanism 82 is installed at the output end of the three-axis adjustment mechanism 81, and the welding torch 31 is installed at the output end of the angle adjustment mechanism 82. The three-axis adjustment mechanism 81 can be a combination of three manual slides for manual operation, or it can be a three-axis robotic arm for program control. Correspondingly, the angle adjustment mechanism 82 can be a conventional manual adjustment structure or a servo motor with high-precision deceleration and braking functions. It should be noted that the existence of the three-axis adjustment mechanism 81 and the angle adjustment mechanism 82 is for precisely adjusting the spatial position and tilt angle of the welding torch 31. The aforementioned vision module 32 should be installed together with the welding torch 31 at the output end of the angle adjustment mechanism 82.

[0043] Alternatively, in other feasible embodiments, when the dynamic virtual median line 5 is corrected according to the thermal strain difference ΔW, it is achieved by controlling the operation of the three-axis adjustment mechanism 81 composed of a three-axis robotic arm and the angle adjustment mechanism 82 composed of a servo motor, so as to meet the requirements of high-precision fine adjustment.

[0044] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," "third," and similar terms used in this application specification and claims do not indicate any order, quantity, or importance, but are merely used to distinguish different components. The terms "an" or "a" and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms "comprising" or "including" and similar terms mean that the elements or objects preceding "comprising" or "including" encompass the elements or objects listed following "comprising" or "including" and their equivalents, and do not exclude other elements or objects. "Above," "below," "left," "right," etc., are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0045] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A high-precision double-seam welding robot, comprising a robot body, a control system, a rotary base located at the end of the robot body, and a welding torch, a vision module, a laser positioning module, and a welding fume extraction module mounted on the rotary base, characterized in that, The rotary seat is elongated and the welding torch is mounted at one end of the rotary seat away from the robot body. The rotation axis of the rotary seat is configured to be orthogonal to the welding surface of the workpiece. When the rotary table is in the 0-degree position, the welding torch is tilted outwards and points to the first weld seam on one side of the workpiece; when the rotary table is rotated 180 degrees to the reverse position, the welding torch is tilted outwards and points to the second weld seam on the opposite side of the workpiece. The control system is configured to: use the laser positioning module to obtain the spatial positions of the first weld and the second weld, construct the geometric center lines of the two welds as virtual midline, and control the robot body to drive the rotation axis of the rotary seat to intersect with the virtual midline and the length midline to be parallel to the virtual midline.

2. The high-precision double-seam welding robot according to claim 1, characterized in that, The control system is configured as follows: Before welding begins, the laser positioning module is controlled to sequentially scan multiple feature points P of the first weld seam. 11 ...P 1n Multiple feature points P of the second weld 21 ...P 2n And calculate multiple line segments P 11 P 21 ...P 1n P 2n midpoint P m1 ...P mn With multiple midpoints P m1 ...P mn Construct the virtual median line; Throughout the welding process, the robot body is controlled to constrain the centerline of the flange mounted on the robot body to the virtual centerline in real time.

3. The high-precision double-seam welding robot according to claim 1, characterized in that, The vision module monitors the real-time gap width at the root of the second weld seam in front of the welding torch when the rotary table is in a 180-degree orientation for welding the second weld seam. The control system compares the real-time gap width with the initial gap width monitored by the vision module during the welding of the first weld, and calculates the thermal strain difference value ΔW. like The control system increases the wire feeding speed and oscillation amplitude of the welding torch to increase the weld metal deposition cross-sectional area. like The control system reduces the wire feeding speed of the welding torch to prevent the weld reinforcement from exceeding the standard.

4. A high-precision double-seam welding robot according to claim 3, characterized in that, If the thermal strain difference ΔW calculated by the control system exceeds a preset threshold, the control system controls the welding torch to stop feeding wire and controls the welding fume extraction module to increase the extraction power.

5. A high-precision double-seam welding robot according to claim 3, characterized in that, The control system is also configured to: At time t during the welding process of the second weld, the virtual median line is translated and rotated along the half-value vector direction of ΔW to generate the dynamic virtual median line at time t, and the robot body is controlled to correct the position of the rotary seat.

6. A high-precision double-seam welding robot according to claim 2, characterized in that, The welding robot also includes: The robot body moves on the guide rail. The base is used to install the guide rail and to set up tooling for fixing the workpiece to be welded. Two calibration blocks are provided and distributed at both ends of the guide rail. Each calibration block has two calibration points with known distances. The control system is also configured to: before each welding operation, control the rotary table to move so that the laser positioning module is aligned with two calibration points on the same calibration block, so as to correct and update the virtual median coordinate system.

7. A high-precision double-seam welding robot according to claim 6, characterized in that, The welding torch is located between the vision module and the welding fume extraction module. When the rotary seat is in the 0-degree and 180-degree positions, the vision module is located in front of the welding torch in the direction of travel during the welding process.

8. A high-precision double-seam welding robot according to claim 1, characterized in that, A three-axis adjustment mechanism is installed at the end of the rotary base away from the robot body, and an angle adjustment mechanism is installed at the output end of the three-axis adjustment mechanism. The welding torch is installed at the output end of the angle adjustment mechanism.

Citation Information

Patent Citations

  • Automatic welding device for large workpieces

    CN119077247A