Intelligent robot full penetration welding method for steel bridge

Abstract

The present application belongs to the field of automatic welding of steel structure bridge, and particularly relates to a full penetration welding method of steel bridge intelligent robot, which comprises the following steps: 1) placing a line laser sensor in front of the welding torch tip of the welding robot; 2) completing the teaching of the first workpiece; 3) scanning the whole workpiece by the laser sensor to obtain the welding track and adjust the welding posture of the starting point; 4) starting the welding, during which the robot control system tracks the position of the welding torch in real time; 5) determining the end point according to the set welding length, and ending the welding operation when the welding torch moves to the end point or there is no detection signal of the laser sensor; 6) moving the welding torch to the other side of the workpiece by the welding robot, and repeating steps 3) to 5); and completing the welding of one workpiece to be welded. The present application has the beneficial effect that, since the workpiece is a corrugated plate, the prior art needs to teach multiple points to complete the welding of the weld, and the present application only needs to teach two points after laser tracking.

Description

Technical Field

[0001] This invention belongs to the field of automated welding of steel structure bridges, specifically a full penetration welding method for steel bridges using intelligent robots. Background Technology

[0002] The advent of the 21st century has spurred the construction and improvement of highway, railway, and urban transportation systems, driving the rapid development and progress of bridge construction. Corrugated steel web girder bridges, using corrugated steel instead of ordinary concrete for the web, effectively reduce bridge weight and improve span capacity, and have received widespread attention and in-depth research in recent decades. Currently, corrugated plate welding is mainly done manually, with mechanical contouring and semi-automatic welding primarily used in the container industry. Currently, semi-automatic welding production lines all have high requirements for waveform processing accuracy; different waveforms and lengths of corrugated plates even require program readjustment, and full weld penetration cannot be guaranteed. Some industrial robot welding systems rely solely on laser tracking to complete corrugated plate welding; the brightness of the workpiece during welding significantly affects the laser data acquisition, and an error of 1mm can lead to weld defects. This has hindered the advancement of manufacturing technology and the transformation of the industry's growth model in my country's construction industry. Summary of the Invention

[0003] This invention primarily addresses the problem of full penetration welding between the corrugated steel web and top plate of steel structure bridges. It employs a welding method that eliminates the need for a teach pen, utilizes real-time welding torch tracking of the corrugated plate, and eliminates the need for root cleaning, achieving single-sided welding with double-sided forming after welding. With arbitrary changes in length and waveform, only the starting point position variation range of 20mm and the workpiece length need to be manually input to automatically complete the welding of the members. During the welding process, the robot automatically plans the welding trajectory and adjusts its welding posture to ensure the welding angle and forward tilt angle are maintained.

[0004] The technical solution adopted by the present invention to achieve the above objectives is: a full penetration welding method for steel bridges using an intelligent robot, comprising the following steps:

[0005] 1) Position the line laser sensor at the tip of the welding torch of the welding robot so that the line laser sensor can collect weld position data on the workpiece.

[0006] 2) The laser sensor is used to find the welding start point of the workpiece, and the welding robot is taught on the first workpiece. The trajectory of the welding robot to the welding start point after teaching and the welding posture of the welding robot after teaching are obtained.

[0007] 3) The laser sensor scans the entire workpiece and finds the welding starting point. The welding robot moves towards the welding starting point of the taught trajectory and sends the weld position data set collected in real time by the laser sensor to the robot control system. The robot control system optimizes the weld position data set to obtain the welding trajectory and adjusts the welding posture of the welding robot at the welding starting point.

[0008] 4) The welding torch starts welding from the welding starting point along the welding trajectory. During the welding process, the robot control system tracks the position of the welding torch in real time.

[0009] 5) Determine the welding endpoint based on the set welding length. When the welding torch reaches the welding endpoint or there is no laser sensor detection signal, end the welding operation.

[0010] 6) The welding robot moves the welding torch to the other side of the workpiece and repeats steps 3) to 5); the welding operation on the other side of the workpiece is completed, and the robot control system records the weld position data on the other side of the workpiece in real time as the root pass trajectory record; the welding of one workpiece to be welded is completed.

[0011] It also includes: when a workpiece requires multi-layer and multi-pass welding, calling the root pass trajectory record, covering the weld pass for offset welding, and completing multi-layer and multi-pass welding of the workpiece.

[0012] The process of calling the root trace record, covering the weld bead for offset welding, and completing multi-layer, multi-pass welding of the workpiece specifically includes the following steps:

[0013] (1) During the root pass welding process of the robot, the robot control system records the weld position data of the first layer welding on the other side of the workpiece in real time.

[0014] (2) Perform multi-layer and multi-pass welding: Before welding, determine the welding parameters based on the welding procedure qualification data;

[0015] (3) Based on the weld position data on the other side of the workpiece in step (1), establish a root track coordinate system, where the X-axis is the welding direction, the direction perpendicular to the welding direction X-axis is the Y-axis, and the vertical direction upward from the X-axis and Y-axis is the Z-axis.

[0016] (4) According to step (2), offset in the Y-axis and Z-axis directions according to the welding parameters to complete the multi-layer and multi-pass welding of the weld.

[0017] The welding parameters refer to the offset of other layers relative to the first weld pass in multi-layer, multi-pass welding.

[0018] The laser sensor is a line laser sensor, and the distance between the laser sensor and the welding torch tip is 35-40mm.

[0019] The welding posture includes the welding angle and the forward tilt angle of the welding torch during the welding process.

[0020] Step 2) yields the welding posture of the taught welding robot, specifically as follows:

[0021] The first piece is taught to start and end points of welding, and a teaching coordinate system is established, with X representing the welding direction and Z representing vertical upward. At this time, the welding robot's welding gun will form an angle in the teaching coordinate system, thus obtaining the welding posture after teaching.

[0022] In step 3), the robot control system optimizes the weld position data set, specifically as follows:

[0023] The robot control system receives the weld position data collected by the laser, arranges the weld position data in an orderly manner, and filters out values ​​that exceed the average error threshold.

[0024] Step 3), adjusting the welding posture of the welding robot at the welding starting point, specifically includes the following steps:

[0025] (1) The laser sensor continuously collects weld position data and establishes a coordinate system at the welding start point, with the welding direction as the X direction and the vertical upward direction as the Z direction;

[0026] (2) Based on the comparison between the welding coordinate system and the teaching coordinate system, adjust the welding angle between the welding torch and the workpiece, as well as the forward tilt angle of the welding torch during the welding process, so that they are the same as the welding angle between the welding torch and the workpiece during the teaching process and the forward tilt angle of the welding torch during the welding process.

[0027] The workpiece is a corrugated steel web or a corrugated top web.

[0028] The present invention has the following beneficial effects and advantages:

[0029] 1. This invention can complete the full fusion weld of corrugated steel web / top web.

[0030] 2. This invention doubles production efficiency and achieves a 96% first-pass yield rate in non-destructive testing; while existing automated welding technologies can only complete fillet welds, and manual welding is still required for full fusion welding. This invention can completely replace manual welding.

[0031] 3. This invention is applicable to any corrugated board, as long as the thickness remains unchanged, and can be completed without adjusting the program; while existing technologies require adjustments to the tooling structure.

[0032] 4. In this invention, only the starting point is taught; subsequent parts do not require teaching. When assembly errors are large, and wavelength inconsistencies occur on the same member, welding remains unaffected. Existing technologies, however, cannot adjust welding posture in real time.

[0033] 5. Since the workpiece is a corrugated plate, existing technologies require teaching multiple points to complete the weld. This invention, using laser tracking, only requires teaching two points. Once the laser sensor collects forward distance data, the welding robot's torch tip has reached the weld start point, and the attitude adjustment is complete. Since the workpiece is a corrugated plate, the welding attitude needs to be adjusted. The welding torch attitude is quickly adjusted after each data segment is collected. Attached Figure Description

[0034] Figure 1 A flowchart illustrating the method of this invention. Detailed Implementation

[0035] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0036] This invention includes: a three-axis semi-gantry system, a semi-gantry slide table on one side of the jig, and a semi-gantry installed on the slide table;

[0037] The Siasun SR10C welding robot is mounted upside down on a semi-gantry system and has three external axes in the X, Y, and Z directions, for a total of nine axes.

[0038] The gas shielded welding power supply, welding torch, and laser sensor adopt a line laser sensor, specifically: META line laser weld seam tracking sensor.

[0039] like Figure 1 The diagram shown is a flowchart of the method of the present invention, which can perform full fusion weld on the corrugated steel web and the corrugated top web.

[0040] This embodiment uses a corrugated steel web plate with a thickness of 16mm as an example. The operation method is as follows:

[0041] (1) Place the line laser sensor in front of the welding robot's welding torch target point (tip), with the welding torch distance being 35-40mm. This will allow for good acquisition of weld data on straight sections, as well as on concave and convex arc sections.

[0042] (2) The laser sensor and the welding gun target point (tip) of the welding robot are calibrated according to the calibration algorithm of the laser sensor. The laser sensor is placed in front of the welding gun tip of the welding robot. The laser sensor and the welding gun tip of the welding robot are calibrated to obtain the positional relationship between the laser sensor and the welding gun tip of the welding robot. This is so that the position of the welding gun of the welding robot on the weld can be adjusted through the weld position data to realize the welding of the weld.

[0043] (3) Find the welding start point of the workpiece through the laser sensor, complete the teaching of the welding robot only on the first workpiece, obtain the trajectory of the welding robot to the welding start point after teaching and the welding posture of the welding robot after teaching.

[0044] The welding posture of the welding robot after receiving the instruction is as follows:

[0045] The first piece is taught to start and end points of welding, and a teaching coordinate system is established, with X representing the welding direction and Z representing vertical upward. At this time, the welding robot's welding gun will form an angle in the teaching coordinate system, thus obtaining the welding posture after teaching.

[0046] (4) Start the system, the line laser finds the welding starting point, the welding torch TCP moves towards the welding starting point, and during the movement of the welding robot, the line laser collects the weld position data. The line laser sends the collected data to the robot control system. After optimization processing by the control system, the welding posture of the welding robot's welding torch at the welding starting point (the welding angle and forward tilt angle formed by the welding robot's welding torch and the horizontal plane of the workpiece) is adjusted.

[0047] The specific steps involved in adjusting the welding robot's welding posture at the welding starting point are as follows:

[0048] A. The laser sensor continuously collects weld position data and establishes a coordinate system at the welding start point, with the welding direction as the X direction and the vertically upward Z direction;

[0049] B. Compare the welding coordinate system with the teaching coordinate system, and adjust the welding angle between the welding torch and the workpiece, as well as the forward tilt angle of the welding torch during the welding process, to be the same as the welding angle between the welding torch and the workpiece during the teaching process, and the forward tilt angle of the welding torch during the welding process.

[0050] (5) After the welding torch reaches the starting point, it is ready to start the arc welding. The line laser is placed in front of the welding torch target point (tip) of the welding robot, and sends the accumulated array of weld position data to the robot control system. After optimization and processing by the robot control system, the welding trajectory is planned and sent to the welding robot to drive the welding torch to move along the planned welding trajectory;

[0051] The aforementioned optimization of the weld position data set by the robot control system is as follows:

[0052] The robot control system receives the weld position data collected by the laser, arranges the weld position data in an orderly manner, and filters out values ​​that exceed the average error threshold.

[0053] During the welding process, the program tracks the laser in real time to correct laser tracking errors, ensuring welding quality and achieving single-sided welding with double-sided forming after welding.

[0054] (6) The welding torch TCP moves along the trajectory sent by the robot control system. After encountering an inflection point, it adjusts the welding posture evenly to ensure that the welding angle and forward tilt angle do not change.

[0055] (7) Repeat this process until the laser detection signal is no longer present near the set distance. Once the welding endpoint is reached, the welding is stopped and the torch is lifted.

[0056] (8) The robot moves to the other side and performs the same starting point search and trajectory planning. Due to the assembly gap and the influence of the blunt edge, the uniformity of the single-sided welding double-sided forming back-through is inconsistent. Corrugated plate intelligent tracking welding software is required to track in real time to complete the root pass welding. Root pass trajectory records are formed on both sides of the weld.

[0057] (9) When the workpiece needs to be welded in multiple layers and multiple passes, call the root pass trajectory record, cover the weld pass and perform offset welding to complete the multi-layer and multi-pass welding of the workpiece.

[0058] The process of calling the root trace record, covering the weld bead for offset welding, and completing multi-layer, multi-pass welding of the workpiece includes the following steps:

[0059] 1) During the root pass welding process of the robot, the robot control system records the weld position data of the first layer welding on the other side of the workpiece in real time;

[0060] 2) Perform multi-layer, multi-pass welding: Determine welding parameters based on welding procedure qualification data before welding;

[0061] The welding parameters are the offsets of other layers relative to the first weld pass in multi-layer, multi-pass welding.

[0062] 3) Based on the weld position data on the other side of the workpiece in step 1), establish a root pass trajectory coordinate system, where the X-axis is the welding direction, the direction perpendicular to the X-axis is the Y-axis, and the vertical direction upward from the X-axis and Y-axis is the Z-axis.

[0063] 4) According to step 2), offset in the Y-axis and Z-axis directions according to the welding parameters to complete multi-layer and multi-pass welding of the weld.

[0064] The working principle of this invention is as follows:

[0065] The workpiece to be welded is fixed and clamped on the work platform, with the starting position fixed and the welding length set. The start button is pressed, and the industrial robot begins to move, using a line laser to locate the welding starting point. After scanning the starting point, the robot moves towards it, optimizing and adjusting the welding posture based on continuously scanned data. The welding wire (TCP) reaches the welding starting point, initiating arc welding. The laser tracker, positioned ahead, continuously scans the welding trajectory and adjusts the welding posture. The robot plans the welding trajectory based on the data provided by the line laser. During welding, the program tracks and corrects the welding position in real time, achieving single-sided welding with double-sided forming. The line laser first reaches the inflection point, and the robot optimizes, plans the welding trajectory, and adjusts the welding posture based on the data collected by the line laser. When the robot moves to near the set length and the laser no longer provides feedback, it is considered the welding endpoint, completing the root pass welding. The robot uses the root pass recorded data to offset and fill the bevel, completing the corresponding cover pass welding.

[0066] The key point of this invention is that it eliminates the need to teach all workpieces; only the first workpiece needs to be taught to achieve batch full penetration welding operations on the production line. Laser scanning is only used for trajectory planning and welding posture adjustment. Through the adoption of the method of this invention in the embodiments, the production efficiency of this invention is increased by 100%, and the first-pass yield of non-destructive testing is 96%. However, for full penetration welding operations, the existing technology can only complete fillet weld welding, which cannot achieve the final effect of this application.

[0067] The above description is merely an embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, extensions, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A steel bridge intelligent robot full penetration welding method, characterized in that, Includes the following steps: 1) Position the laser sensor at the tip of the welding torch on the welding robot so that the laser sensor can collect weld position data on the workpiece; 2) The laser sensor is used to find the welding start point of the workpiece, and the welding robot is taught on the first workpiece. The trajectory of the welding robot to the welding start point after teaching and the welding posture of the welding robot after teaching are obtained. 3) The laser sensor scans the entire workpiece and finds the welding starting point. The welding robot moves towards the welding starting point of the taught trajectory and sends the weld position data set collected in real time by the laser sensor to the robot control system. The robot control system optimizes the weld position data set to obtain the welding trajectory and adjusts the welding posture of the welding robot at the welding starting point. The welding posture includes: the welding angle and the forward tilt angle of the welding torch during the welding process; 4) The welding torch starts welding from the welding starting point along the welding trajectory. During the welding process, the robot control system tracks the position of the welding torch in real time. 5) Determine the welding endpoint based on the set welding length. When the welding torch reaches the welding endpoint or there is no laser sensor detection signal, end the welding operation. 6) The welding robot moves the welding torch to the other side of the workpiece and repeats steps 3) to 5); the welding operation on the other side of the workpiece is completed. The robot control system records the weld position data on the other side of the workpiece in real time as the root pass trajectory record; the welding of one workpiece to be welded is completed. When a workpiece requires multi-layer and multi-pass welding, the root pass trajectory record is called to cover the weld pass and perform offset welding to complete the multi-layer and multi-pass welding of the workpiece. The process of calling the root trace record, covering the weld bead for offset welding, and completing multi-layer, multi-pass welding of the workpiece specifically includes the following steps: (1) During the root pass welding process of the robot, the robot control system records the weld position data of the first layer welding on the other side of the workpiece in real time; (2) Perform multi-layer and multi-pass welding: Before welding, determine the welding parameters based on the welding procedure qualification data; (3) Based on the weld position data on the other side of the workpiece in step (1), establish the root track coordinate system, where the X-axis is the welding direction, the direction perpendicular to the welding direction X-axis is the Y-axis, and the vertical direction upward from the X-axis and Y-axis is the Z-axis. (4) According to step (2), offset in the Y-axis and Z-axis directions according to the welding parameters to complete the multi-layer and multi-pass welding of the weld.

2. A steel bridge intelligent robot full penetration welding method according to claim 1, characterized in that, The welding parameters refer to the offset of other layers relative to the first weld pass in multi-layer, multi-pass welding.

3. The intelligent robot full penetration welding method for steel bridges according to claim 1, characterized in that, The laser sensor is a line laser sensor, and the distance between the laser sensor and the tip of the welding torch is 35-40mm.

4. The steel bridge intelligent robot full-penetration welding method of claim 1, wherein, Step 2), obtaining the welding posture of the taught welding robot, specifically: The first piece is taught to start and end points of welding, and a teaching coordinate system is established, with X representing the welding direction and Z representing vertical upward. At this time, the welding robot's welding gun will form an angle in the teaching coordinate system, thus obtaining the welding posture after teaching.

5. A steel bridge intelligent robot full penetration welding method according to claim 1, characterized in that, In step 3), the robot control system optimizes the weld position data set, specifically as follows: The robot control system receives the weld position data collected by the laser, arranges the weld position data in an orderly manner, and filters out values ​​that exceed the average error threshold.

6. A steel bridge intelligent robot full-penetration welding method according to claim 1, characterized in that, Step 3), adjusting the welding posture of the welding robot at the welding starting point, specifically includes the following steps: (1) The laser sensor continuously collects weld position data and establishes a coordinate system at the welding start point, with the welding direction as the X direction and the vertical upward direction as the Z direction; (2) Based on the comparison between the welding coordinate system and the teaching coordinate system, adjust the welding angle between the welding torch and the workpiece, as well as the forward tilt angle of the welding torch during the welding process, so that they are the same as the welding angle between the welding torch and the workpiece during the teaching process and the forward tilt angle of the welding torch during the welding process.

7. The intelligent robot full penetration welding method for steel bridges according to claim 1, characterized in that, The workpiece is a corrugated steel web or a corrugated top web.

Images