Intelligent welding method and system for small group assembly components in a ship

Abstract

This invention provides an intelligent welding method and system for small assembly components in ships. The method includes the following steps: scanning the workpiece to be welded to obtain target point cloud data; matching a target workpiece model corresponding to the workpiece to be welded from a preset template workpiece library based on the target point cloud data; obtaining weld model coordinate information corresponding to the workpiece to be welded based on a workpiece weld information table and the target workpiece model; generating a rotation transformation matrix based on the target point cloud data and the template point cloud file corresponding to the target workpiece model; converting the digital model coordinate information into real coordinate information associated with the weld based on the rotation transformation matrix; and controlling a welding robot in a welding device to weld the workpiece to be welded based on the real coordinate information. This invention helps improve the welding accuracy of the welding robot, thereby improving the welding quality of small assembly components in ships.

Description

Technical Field

[0001] This invention relates to the field of shipbuilding technology, and more specifically, to an intelligent welding method and system for small assembly components in ships. Background Technology

[0002] Welding is a cornerstone of manufacturing and plays a crucial role in the development of the industrial economy. Ship welding, as a key technology in shipbuilding, directly affects the quality and efficiency of ship construction. Currently, most shipbuilding companies use traditional, primarily manual welding methods, which suffer from high labor costs, long working hours, inconsistent quality, and poor production continuity, failing to meet the requirements of modern shipbuilding enterprises. With the rapid advancement of robotic welding technology, its application in shipbuilding has received widespread attention.

[0003] The mid-level assembly components in shipbuilding are structurally simple and numerous. As an important intermediate product in shipbuilding, they are suitable for manufacturing on robotic assembly lines. Although intelligent production lines for mid-level assembly have been established in China, the extraction of weld information is usually done by scanning the workpiece. This can lead to situations where factors such as obstruction can prevent accurate extraction of weld lines, resulting in poor accuracy in robotic welding.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] In view of this, the present invention provides an intelligent welding method and system for small assembly components in ships, which helps to improve the welding accuracy of welding robots and thus improve their welding quality.

[0006] According to one aspect of the present invention, a smart welding method for small assembly components in a ship is provided, comprising the following steps:

[0007] S110, scans the workpiece to be welded to obtain target point cloud data;

[0008] S120, based on the target point cloud data, a target workpiece model corresponding to the workpiece to be welded is obtained from the preset template workpiece library; the preset template workpiece library stores a workpiece weld information table and a template point cloud file corresponding to each workpiece model.

[0009] S130, Based on the workpiece weld information table and the target workpiece model, obtain the weld digital model coordinate information corresponding to the workpiece to be welded;

[0010] S140, Generate a rotation transformation matrix based on the target point cloud data and the template point cloud file corresponding to the target workpiece model;

[0011] S150, based on the rotation transformation matrix, convert the digital model coordinate information into real coordinate information associated with the weld; and

[0012] S160, Based on the actual coordinate information, control the welding robot in the welding device to weld the workpiece to be welded.

[0013] Optionally, step S160 includes:

[0014] Based on a preset path planning algorithm, with robot obstacle avoidance and multi-robot coordination as constraints, and with the shortest welding time and / or the minimum welding deformation as optimization objectives, the robot obstacle avoidance welding path is calculated.

[0015] Based on the actual coordinate information and the robot's obstacle avoidance welding path, the welding robot in the welding device is controlled to weld the workpiece to be welded.

[0016] Optionally, step S120 includes:

[0017] Based on the target point cloud data, calculate the corresponding first Hu moment data, which is used as the first vector;

[0018] Calculate the second Hu moment data corresponding to each workpiece model in the preset template workpiece library, use it as the second vector, and combine all the second vectors to obtain the second vector group;

[0019] Calculate the distance between each second vector in the second vector group and the first vector, and take the second vector with the smallest distance as the target second vector;

[0020] The workpiece model corresponding to the target second vector is taken as the target workpiece model.

[0021] Optionally, step S120 includes:

[0022] The first boundary feature information is extracted from the target point cloud data;

[0023] Based on the first boundary feature information, a target workpiece model corresponding to the workpiece to be welded is obtained from a preset template workpiece library.

[0024] Optionally, step S140 includes:

[0025] Based on the template point cloud file, obtain the second boundary feature information corresponding to the workpiece to be welded;

[0026] A rotation transformation matrix is ​​generated based on the first boundary feature information and the second boundary feature information.

[0027] Optionally, generating the rotation transformation matrix based on the first boundary feature information and the second boundary feature information includes:

[0028] Based on a preset point cloud library, the second boundary feature information is rotated to be consistent with the first boundary feature information, and a rotation transformation matrix is ​​generated.

[0029] Optionally, before step S110, the method further includes:

[0030] Import the models corresponding to each of the multiple workpieces to create a preset template workpiece library;

[0031] Based on the model corresponding to each workpiece, the weld information corresponding to each workpiece model is extracted, and a workpiece weld information table is generated. The workpiece weld information table is stored in the preset template workpiece library.

[0032] Optionally, step S160 includes:

[0033] Before welding begins, the welding robot is controlled to perform a second positioning of the welding start and end points, and the coordinate information of the welding start and end points is corrected based on the second positioning results.

[0034] During the welding process, the welding robot is controlled to track and detect changes in the weld seam, and the welding trajectory is corrected based on the tracking results;

[0035] During the welding process, the actual welding formation data is collected in real time, and the welding process parameter database referenced by the welding robot during the welding process is corrected based on the actual welding formation data.

[0036] Optionally, the welding device further includes a torch cleaning and wire cutting mechanism for cleaning spatter generated by the welding robot during the welding process and clogging the gas protective sleeve of the welding torch, and ensuring that the extension length of the welding wire remains consistent.

[0037] Optionally, the welding device further includes a welding gantry, a welding torch, a line laser scanner, a laser tracking sensor, a camera module, an industrial control computer, and a host computer; the host computer is integrated into the industrial control computer; the welding gantry includes a guide rail, a gantry body, and a rotating platform, the guide rail and the rotating platform being connected to the gantry body respectively; the line laser scanner and the camera module are mounted on the gantry body; the gantry body can slide along the X direction on the guide rail; the rotating platform can slide along the Y direction on the gantry body; the welding torch is connected to the rotating platform, and the laser tracking sensor is mounted on the welding torch.

[0038] Optionally, the welding device further includes a wire spool and a wire feeding mechanism. The wire spool and the wire feeding mechanism are respectively connected to the rotary platform. The wire spool is located above the wire feeding mechanism. The wire spool rotates synchronously with the rotation axis of the rotary platform. The wire output port of the wire spool is aligned with the wire input port of the wire feeding mechanism.

[0039] Optionally, both the rotary platform and the guide rail are driven by servo motors, the rotary platform uses an RV reducer for deceleration, and the guide rail uses a planetary reducer for deceleration.

[0040] According to another aspect of the present invention, a smart welding system for small assembly components in a ship is provided, for implementing any of the above-described smart welding methods, comprising:

[0041] The workpiece scanning module scans the workpiece to be welded to obtain target point cloud data;

[0042] The template matching module matches the target workpiece model corresponding to the workpiece to be welded from a preset template workpiece library based on the target point cloud data; the preset template workpiece library stores a workpiece weld information table and template point cloud files corresponding to each workpiece model;

[0043] The weld seam digital model coordinate acquisition module acquires the weld seam digital model coordinate information corresponding to the workpiece to be welded based on the workpiece weld seam information table and the target workpiece model.

[0044] The rotation transformation matrix generation module generates a rotation transformation matrix based on the target point cloud data and the template point cloud file corresponding to the target workpiece model.

[0045] The weld seam real coordinate transformation module converts the digital model coordinate information into real coordinate information associated with the weld seam based on the rotation transformation matrix; and

[0046] The welding execution module controls the welding robot in the welding device to weld the workpiece based on the actual coordinate information.

[0047] Optionally, the welding execution module includes:

[0048] The obstacle avoidance welding path planning unit, based on a preset path planning algorithm, uses robot obstacle avoidance and multi-robot coordination as constraints, and takes the shortest welding time and / or the minimum welding deformation as optimization objectives to calculate the robot obstacle avoidance welding path.

[0049] The obstacle avoidance welding control unit controls the welding robot in the welding device to weld the workpiece to be welded, based on the real coordinate information and the robot's obstacle avoidance welding path.

[0050] The advantages of this invention compared to the prior art are as follows:

[0051] The intelligent welding method and system for small assembly components in ships provided by this invention realizes fully automated welding of small assembly component production lines in ships. By pre-analyzing the digital models of various workpieces, identifying weld seams, and establishing a workpiece weld seam information table corresponding to each workpiece, the weld seam coordinate information corresponding to the workpiece being welded can be extracted from the workpiece weld seam information table during welding. It is not necessary to obtain the weld seam by fitting point cloud data from on-site scanning, which can avoid the situation where the welding line cannot be accurately extracted due to factors such as occlusion during on-site scanning. Then, the weld seam digital model coordinates are transformed to the real coordinate system to obtain the actual weld seam position for welding, which helps to improve the welding accuracy of the welding robot and thus improve the welding quality.

[0052] On the other hand, since the workpiece model contains coordinate and type information of curves such as straight lines and arcs, it is also possible to extract arc welds. Attached Figure Description

[0053] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0054] Figure 1 This is a schematic diagram of the structure of a welding apparatus disclosed in an embodiment of the present invention;

[0055] Figure 2 This is another structural schematic diagram of the welding apparatus disclosed in one embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram of the connection structure between the gantry body and the rotating platform in a welding apparatus disclosed in an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the structure of the rotary platform in a welding apparatus disclosed in an embodiment of the present invention;

[0058] Figure 5 This is another structural schematic diagram of the rotary platform in the welding apparatus disclosed in an embodiment of the present invention;

[0059] Figure 6 This is a schematic flowchart of an intelligent welding method for small assembly components in a ship, as disclosed in an embodiment of the present invention.

[0060] Figure 7 This is a schematic diagram of the structure of a workpiece to be welded according to an embodiment of the present invention;

[0061] Figure 8 This is a flowchart illustrating step S120 of an intelligent welding method for small assembly components in a ship, as disclosed in another embodiment of the present invention.

[0062] Figure 9 This is a flowchart illustrating an intelligent welding method for small assembly components in a ship, as disclosed in another embodiment of the present invention.

[0063] Figure 10 This is a flowchart illustrating an intelligent welding method for small assembly components in a ship, as disclosed in another embodiment of the present invention.

[0064] Figure 11 This is a schematic diagram of the structure of an intelligent welding system for small assembly components in a ship, as disclosed in an embodiment of the present invention.

[0065] Figure label:

[0066] 11. Gantry body; 12. Rotary platform; 13. Guide rail; 14. Welding torch; 15. Line laser scanner; 16. Laser tracking sensor; 17. Camera module; 18. Welding wire spool; 19. Wire feeding mechanism; 21. Welding robot; 22. Welding machine; 23. Rectangular rail; 24. RV reducer; 25. Planetary reducer; 26. Wire shearing machine; 27. Control cabinet; 71. Workpiece to be welded; Detailed Implementation

[0067] Example embodiments will now be described more fully with reference to the accompanying drawings. However, these example embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided so that the invention will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced without one or more of the specific details described, or other methods, materials, apparatus, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring aspects of this disclosure. The same reference numerals in the figures denote the same or similar structures, and therefore their detailed descriptions are omitted.

[0068] The terms “a,” “one,” “the,” “the,” and “at least one” are used to indicate the presence of one or more elements / components / etc.; the terms “including,” “having,” and “have” are used to indicate an open-ended inclusion meaning and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0069] One embodiment of the present invention discloses an intelligent welding method for small assembly components in ships. This welding method controls a welding device to weld small assembly components in ships, specifically through automated welding performed by a welding robot within the welding device. The present invention does not limit the number of welding robots in the welding device; those skilled in the art can configure the number as needed.

[0070] like Figures 1 to 5 As shown, the welding apparatus used for welding in this embodiment includes a welding gantry, a welding torch 14, a line laser scanner 15, a laser tracking sensor 16, a camera module 17, a wire spool 18, a wire feeding mechanism 19, a welding robot 21, an industrial computer, and a host computer. The host computer is a software control system deployed within the industrial computer, meaning it uses the industrial computer as a platform to implement the functions. The industrial computer establishes communication connections with the line laser scanner 15, the laser tracking sensor 16, and the camera module 17, respectively, and the host computer controls the welding robot 21 to perform welding operations.

[0071] refer to Figure 1 and Figure 2 The aforementioned welding gantry includes a guide rail 13, a gantry body 11, and a rotating platform 12. The guide rail 13 and the rotating platform 12 are respectively connected to the gantry body 11. The gantry body 11 is transversely connected to the guide rail 13. This welding gantry is a three-axis gantry: an X-axis, a Y-axis, and a rotation axis on the rotating platform 12 (i.e.,... Figure 2 The rotation axis is equivalent to rotating around the Z-axis. The X-axis, Y-axis, and Z-axis are mutually perpendicular. The X-axis extends in the same direction as the guide rail 13. The Y-axis extends in the same direction as the gantry body 11. The welding robot 21 is mounted on the rotary platform 12. It should be noted that two welding robots 21 are shown in the accompanying drawings, but this application does not impose a specific limit on the number of welding robots 21. The two welding robots 21 mounted on the rotary platform 12 can rotate ±185°, making the welding operation more flexible.

[0072] refer to Figure 3 The line laser scanner 15 and camera module 17 are mounted on the gantry body 11. The gantry body 11 is slidably connected to the guide rail 13 and can slide along the X-direction (i.e., the extension direction of the X-axis) on the guide rail 13. The rotary platform 12 can slide along the Y-direction (i.e., the extension direction of the Y-axis) on the gantry body 11. The welding torch 14 is connected to the rotary platform 12. The laser tracking sensor 16 is mounted on the welding torch 14.

[0073] refer to Figure 4 and Figure 5The aforementioned wire spool 18 and wire feeding mechanism 19 are respectively connected to the rotary platform 12. The wire spool 18 is located above the wire feeding mechanism 19. The wire spool 18 rotates synchronously with the rotation axis of the rotary platform 12; this helps to eliminate the influence of the rotational damping of the wire spool 18 on the wire feeding stability of the wire feeding mechanism 19, thereby ensuring wire feeding stability. The wire output port of the wire spool 18 is aligned with the wire input port of the wire feeding mechanism 19; this also helps to ensure wire feeding stability.

[0074] refer to Figure 5 In this embodiment, the rotary platform 12 is mounted on the gantry body 11 and moves together with the gantry body 11. Both the rotary platform 12 and the guide rail 13 are driven by servo motors. The rotary platform 12 uses an RV reducer 24 for speed reduction and achieves high-precision rotary motion through gear and toothed slewing bearing transmission; it also features a gear meshing adjustment mechanism to eliminate gear and rack transmission backlash and improve transmission accuracy. The guide rail 13 uses a planetary reducer 25 for speed reduction and achieves high-precision rotary motion through gear and rack transmission. It also features a gear and rack meshing adjustment mechanism to eliminate gear and rack transmission backlash and improve transmission accuracy. Furthermore, in this embodiment, the guide rail 13 has a structure with a linear guide rail 13 on one side and a rectangular rail 23 on the other side, which ensures the smooth operation of the guide rail 13 and avoids low-speed crawling.

[0075] In this embodiment, a movable slide is provided on the gantry body 11, which moves together with the gantry body 11. The movable slide facilitates the sliding of the rotary platform 12 on the gantry body 11. The gantry body 11 is mounted on the guide rail 13, and is also equipped with a servo motor, reducer, and gears, which are transmitted through a rack and pinion to achieve high-speed, high-precision linear motion. Heavy-duty rectangular precision guides are installed on both sides of the guide rail 13 to provide reliable load-bearing and rolling guidance for the gantry's movement. A camera module 17 is mounted on the gantry body 11 for monitoring the welding process.

[0076] In this embodiment, the welding device further includes a welding machine 22, a wire cutter 26, and a control cabinet 27. The welding machine 22, the wire cutter 26, and the control cabinet 27 are all installed on the gantry body 11 and move together with the gantry body 11.

[0077] In this embodiment, the welding apparatus also includes a torch cleaning and wire cutting machine 26. The torch cleaning and wire cutting machine 26 is used to clean the spatter generated by the welding robot 21 during the welding process and stuck in the gas protective sleeve of the welding torch 14, and to ensure that the extension length of the welding wire remains consistent; thereby helping to ensure the welding quality of the sub-assembly components in the ship.

[0078] like Figure 6 As shown, an embodiment of the present invention discloses an intelligent welding method for small assembly components in ships. The welding method includes the following steps:

[0079] S110 scans the workpiece to be welded to obtain target point cloud data. Figure 7 This is a schematic diagram of the structure of a workpiece 71 to be welded, as exemplarily shown in this invention. In specific implementation, step S110 may include: scanning the workpiece to be welded using a line laser scanner in the welding device to obtain raw data, and then converting the raw data into initial point cloud data that the system can process. Then, the initial point cloud data is filtered to remove erroneous and redundant data, and the initial point cloud data is segmented to obtain target point cloud data.

[0080] S120, based on the aforementioned target point cloud data, a target workpiece model corresponding to the workpiece to be welded is obtained from the preset template workpiece library. That is, the identification information of the target workpiece model, such as its model number, can be obtained, which in turn allows the identification information of the workpiece to be welded to be obtained. The preset template workpiece library contains multiple workpiece models, each with unique identification information.

[0081] The preset template workpiece library stores workpiece weld information tables and corresponding template point cloud files for each workpiece model. During the creation of the preset template workpiece library and the import of various workpiece models, the weld information for each workpiece model can be parsed and identified, and template point cloud files containing boundary feature information of each workpiece model can be generated. The weld information of each workpiece model can then be used to form a workpiece weld information table. This table records the digital model coordinates of the welds corresponding to each workpiece model, such as a one-to-one correspondence between workpiece model numbers and weld digital model coordinates.

[0082] like Figure 8 As shown, in this embodiment, step S120 includes:

[0083] S121, the first boundary feature information is extracted from the above target point cloud data.

[0084] S122, based on the aforementioned first boundary feature information, calculate the first Hu moment data corresponding to the target point cloud data, as the first vector. Here, the Hu moment of an image is an image feature with translation, rotation, and scale invariance.

[0085] S123, calculate the second Hu moment data corresponding to each workpiece model in the preset template workpiece library, use it as the second vector, and combine all the second vectors to obtain the second vector group.

[0086] S124, calculate the distance between each second vector in the second vector group and the first vector, and take the second vector with the smallest distance as the target second vector.

[0087] S125, take the workpiece model corresponding to the above-mentioned target second vector as the target workpiece model; and obtain the identification information corresponding to the above-mentioned workpiece to be welded from the above-mentioned target workpiece model.

[0088] This embodiment utilizes the first boundary feature information to calculate the Hu moment data, which can reduce the amount of related calculations and improve computational efficiency; thereby helping to improve the welding efficiency of the welding robot.

[0089] S130, based on the aforementioned workpiece weld information table and the aforementioned target workpiece model, obtain the weld digital model coordinate information corresponding to the workpiece to be welded. Since the model number of the target workpiece model, i.e., the model number of the workpiece to be welded, can be obtained from the workpiece weld information table, the weld digital model coordinate information of that workpiece can be obtained accordingly.

[0090] The aforementioned target workpiece model is essentially a model of the workpiece. Analyzing the target workpiece model to identify the weld seam corresponding to the workpiece to be welded has two advantages: 1. Extracting the weld seam by analyzing the digital model avoids situations where factors such as occlusion during scanning or flanges on the upper edge of the stiffener prevent accurate extraction of the weld line. 2. In existing technologies, Hough transform is required to reconstruct the model from the scanned point cloud data when extracting the weld seam, resulting in the ability to process only straight stiffeners and not workpieces with arcs. Since the digital model contains the coordinates and type information of curves such as straight lines and arcs, this embodiment can extract arc weld seams without having to fit the weld seam to the point cloud data scanned on-site. This improves the accuracy of weld seam extraction, thereby improving welding accuracy and ultimately enhancing the welding quality of small assembly components in ships.

[0091] S140, based on the aforementioned target point cloud data and the template point cloud file corresponding to the aforementioned target workpiece model, generate a rotation transformation matrix. Specifically, as follows: Figure 9 As shown, in this embodiment, step S140 includes:

[0092] S141, Based on the above template point cloud file, obtain the second boundary feature information corresponding to the above workpiece to be welded.

[0093] S142, Generate a rotation transformation matrix based on the first boundary feature information and the second boundary feature information.

[0094] Specifically, based on a preset point cloud library, the second boundary feature information is rotated to match the first boundary feature information, generating a rotation transformation matrix. The preset point cloud library can be a PCL (Point Cloud Library) program library; that is, the process can be implemented based on a PCL program library. This embodiment will not elaborate on the specific implementation process. This embodiment rotates based on the first boundary feature information and the second boundary feature information to obtain the rotation transformation matrix, which reduces the related computational load and improves computational efficiency; thus, it helps to improve the welding efficiency of the welding robot.

[0095] S150, based on the aforementioned rotation transformation matrix, the aforementioned digital model coordinate information is converted into the actual coordinate information associated with the aforementioned weld. Specifically, multiplying the digital model coordinate information and the rotation transformation matrix yields the actual coordinate information associated with the weld.

[0096] And S160, based on the aforementioned real coordinate information, controls the welding robot in the welding device to weld the aforementioned workpiece. In specific implementation, such as... Figure 10 As shown, step S160 may include:

[0097] S161, based on a preset path planning algorithm, with robot obstacle avoidance and multi-robot coordination as constraints, and with the shortest welding time and / or the minimum welding deformation as optimization objectives, calculates the robot obstacle avoidance welding path.

[0098] S162, based on the above-mentioned real coordinate information and the above-mentioned robot obstacle avoidance welding path, control the welding robot in the welding device to weld the above-mentioned workpiece to be welded.

[0099] Specifically, during the welding operation, the welding robot may collide with the workpiece itself, other workpieces, or other non-workpiece obstacles, affecting the quality and efficiency of the welding operation, and thus affecting the welding quality of the small assembly components in the ship.

[0100] To address this issue, this embodiment incorporates obstacle avoidance during welding path planning, generating a robot obstacle-avoidance welding path to prevent interference and collisions during the welding process. This embodiment utilizes a repulsive field strategy path planning method based on the PRM (Probabilistic Roadmaps) algorithm framework. With optimization objectives such as minimizing welding path or time, minimizing welding deformation, and / or energy saving, it sequentially plans jump point information for each workpiece in the robot's motion space. The robot's motion posture is solved based on its forward and inverse kinematics, considering constraints such as obstacle avoidance and multi-robot coordination, to achieve multi-objective path optimization and effective obstacle avoidance for multiple welding robots.

[0101] The basic idea of ​​the repulsive field strategy is to abstract the robot's motion in the environment as the motion of a point particle in an artificial gravitational field. Obstacles exert a repulsive force on the robot, while the target point exerts an attractive force. Under the combined effect of these forces, the robot avoids obstacles and reaches the target point. The closer the robot is to the repulsive field, the greater the repulsive force it experiences, and the higher the risk of collision with the surface. Obstacle avoidance processing for the welding path can provide a high-quality reference trajectory for subsequent trajectory tracking, exhibiting good robustness and real-time computational efficiency.

[0102] During the preparation process before welding, the actual coordinates of the weld seam, the robot's obstacle-avoidance welding path, the welding torch posture, the weld seam shape, and the welding cycle time can be combined to automatically retrieve welding process parameters, generate a multi-robot collaborative operation program file, and download it. The robot's obstacle-avoidance welding path can be imported into the simulation module to verify the robot's welding trajectory and confirm whether there are any problems. If problems are found, the trajectory can be manually corrected.

[0103] In an optional embodiment, step S160 may further include:

[0104] Before welding begins, the welding robot performs secondary positioning of the welding start and end points, and corrects the coordinate information of the welding start and end points based on the secondary positioning results. This helps improve the welding quality of small assembly components in ships. The specific process of secondary positioning can be implemented with reference to line laser positioning technology, and will not be described in detail in this embodiment.

[0105] During the welding process, the welding robot is controlled to track and detect changes in the weld seam, and the welding trajectory is corrected based on the tracking results. Due to errors in processing and assembly, conditions such as weld seam position and gap width may change. This module is used to track and detect these changes during the welding process and feed the tracking results back to the robot welding system to correct the coordinate information of the welding start and end points; thereby improving the welding quality of small assembly components in ships.

[0106] Furthermore, during the welding process, the actual welding formation data is collected in real time, and the welding process parameter database referenced by the welding robot during the welding process is corrected based on the actual welding formation data, thereby improving the welding quality of subsequent welding operations.

[0107] In another optional embodiment, step S160 may further include: real-time monitoring of the welding status, online control of welding parameters, welding process management, and data statistical analysis during the welding process to achieve traceability of the welding process; thereby helping to ensure the welding quality of the welding robot. Specifically, the data acquisition module can collect information such as the robot welding system's operating current, voltage, gas flow rate, wire feeding speed, ambient temperature and humidity, and alarms, and can also bind workpiece information to achieve traceability of welding quality. The process monitoring module parses, stores, and monitors the data obtained by the data acquisition module. The statistical analysis module statistically analyzes the robot welding system's welding efficiency, actual welding time, preparation time, completed welding tasks, workpiece welding data, and welding quality within a certain period to evaluate the system's operating status and improve production efficiency. The process management module is used to set and manage welds, products, and processes. Based on the weld information obtained in step S130, the corresponding process parameter information is retrieved and distributed to the welding robot.

[0108] In another alternative embodiment, prior to step S110, the method further includes:

[0109] S101, import the models corresponding to multiple workpieces respectively, and establish a preset template workpiece library.

[0110] And S102, based on the model corresponding to each workpiece, extract the weld information corresponding to each workpiece model, generate a workpiece weld information table, and store the workpiece weld information table in the above-mentioned preset template workpiece library.

[0111] All of the above models are three-dimensional digital models. In specific implementation, the preset template workpiece library contains multiple workpiece models. After importing the workpiece models, the weld information corresponding to each workpiece model can be identified. For example, a preset algorithm is used to find the intersecting lines between stiffeners in the three-dimensional digital model and determine them as welds. It is also necessary to extract the boundary feature information of the corresponding workpiece from the workpiece model, generate a template point cloud file, and store the template point cloud file in the preset template workpiece library. For example, the above preset algorithm can be implemented based on the open-source Open CASCADE platform, which will not be described in detail in this embodiment. This embodiment can identify various types of welds, including flat fillet welds, vertical fillet welds, and butt welds, and segment the welds according to the relative positional relationship of each weld.

[0112] This embodiment pre-analyzes the digital models of various workpieces to identify weld seams and establish a corresponding workpiece weld seam information table for each workpiece. During welding, the weld seam coordinate information corresponding to the workpiece being welded can be extracted from the workpiece weld seam information table. It is not necessary to fit the weld seam through point cloud data scanned on-site, which can avoid the situation where the welding line cannot be accurately extracted due to factors such as occlusion during on-site workpiece scanning. Then, the weld seam coordinates are transformed into the user coordinate system to obtain the actual weld seam position for welding, which helps to improve the welding accuracy of the welding robot and thus improve the welding quality.

[0113] An embodiment of the present invention also discloses an operation process for the assembly of small components in a ship, the operation process including the following steps:

[0114] S211, import the models corresponding to multiple workpieces respectively, and establish a preset template workpiece library.

[0115] S212, based on the model corresponding to each workpiece, extract the weld information corresponding to each workpiece model, generate a workpiece weld information table, and store the workpiece weld information table in the aforementioned preset template workpiece library. Specifically, calculate all welds (information) of various types, including flat fillet welds and vertical fillet welds, on each workpiece, and segment the welds according to the relative positional relationship of each weld to generate weld information tables corresponding to each workpiece.

[0116] S213: Input the scanning range of the current welding task into the host computer. The line laser scanner in the welding device scans the workpiece to be welded. The gantry body moves to the scanning start position, and after the line laser scanner is turned on, the gantry body moves at a constant speed to scan the workpiece. The scanned data is sent to the field industrial control computer in real time for processing by the host computer. After the gantry body moves to the end position, the scanning is completed, and the line laser scanner is turned off. After scanning, the host computer obtains the raw data and the quantity and position information of the workpieces to be welded.

[0117] S214, the host computer converts the raw data into initial point cloud data, and filters and segments the initial point cloud data to remove erroneous and redundant data, obtaining target point cloud data. Based on the target point cloud data, a target workpiece model corresponding to the workpiece to be welded is matched from a preset template workpiece library. Second boundary feature information is extracted from the target workpiece model. Identification information corresponding to the workpiece to be welded can be obtained from the target workpiece model. A rotation transformation matrix is ​​generated based on the first and second boundary feature information. Based on the rotation transformation matrix, the digital model coordinate information is converted into real coordinate information associated with the weld. The identification information can be the model number of the workpiece to be welded.

[0118] S215, based on the quantity, location, model, and actual coordinates of the weld seams obtained in the previous steps, the host computer plans and generates a robot obstacle avoidance welding path using the repulsive field strategy path planning method within the PRM (Probabilistic Roadmaps) algorithm framework. Then, combining the welding torch posture, weld seam morphology, and welding cycle time, it automatically retrieves welding process parameters and outputs a multi-robot collaborative welding program. The robot obstacle avoidance welding path is then imported into the simulation module to verify the robot welding trajectory and confirm whether there are any issues. If there are no issues, the welding program is confirmed and sent to each robot. If there are issues, the trajectory is manually corrected before confirmation and sending.

[0119] S216, Welding program issued. After welding is started, the intelligent welding host computer software receives the robot welding program file and issues it to the robot controller.

[0120] S217: Simultaneously with the start of welding, the adaptive control module for the welding process is activated. The start / end point positioning module drives the tracking and positioning sensors in the gantry welding system to perform secondary positioning of the welding start and end points, and feeds the positioning results back to the robotic welding system to correct the coordinate information of the welding start and end points. During welding, the weld seam tracking module tracks and detects changes in the weld seam, and feeds the tracking results back to the robotic welding system to correct errors caused by processing and assembly.

[0121] S218, Data Acquisition and Monitoring. The data acquisition module collects information such as the working current, voltage, gas flow rate, wire feed speed, ambient temperature and humidity, and alarms of the robotic welding system during the welding process. It can also bind workpiece information to achieve traceability of welding quality. The process monitoring module parses, stores, and monitors the data acquired by the data acquisition module.

[0122] S219, determine if there are any unwelded workpieces. If so, jump to S216 to continue searching for welding paths for the unwelded workpieces; if not, end the welding process.

[0123] An embodiment of the present invention also provides an intelligent welding system for small assembly components in ships, used to implement the intelligent welding method disclosed in any of the above embodiments. Detailed processes and advantages of the intelligent welding method can be found in the descriptions of the above embodiments, and will not be repeated here. Figure 11 As shown, in this embodiment, the intelligent welding system 8 includes:

[0124] The workpiece scanning module 81 scans the workpiece to be welded to obtain target point cloud data.

[0125] The template matching module 82, based on the aforementioned target point cloud data, matches the target workpiece model corresponding to the aforementioned workpiece to be welded from the preset template workpiece library.

[0126] The weld seam digital model coordinate acquisition module 83 acquires the weld seam digital model coordinate information corresponding to the workpiece to be welded based on the preset template workpiece library.

[0127] The rotation transformation matrix generation module 84 generates a rotation transformation matrix based on the target point cloud data and the target workpiece model.

[0128] The weld seam real coordinate transformation module 85, based on the aforementioned rotation transformation matrix, converts the aforementioned digital model coordinate information into real coordinate information associated with the aforementioned weld seam.

[0129] The welding execution module 86 controls the welding robot in the welding device to weld the workpiece to be welded based on the above-mentioned real coordinate information.

[0130] In one optional embodiment, the intelligent welding system comprises a control layer, a field layer, and an equipment layer. The control layer includes an intelligent welding control module and a server-side database, deployed as a server. The field layer includes an industrial control computer, a local database, a digital model preprocessing module, an identification and positioning module, a system control module, and an autonomous programming module. The equipment layer comprises a welding robot, a welding power source, a welding torch, a wire spool, a positioning and tracking module, a data acquisition unit, a line laser scanner, a wire feeding mechanism, and a camera module.

[0131] In an optional embodiment, the welding execution module includes:

[0132] The obstacle avoidance welding path planning unit, based on a preset path planning algorithm, uses robot obstacle avoidance and multi-robot coordination as constraints, and takes the shortest welding time and / or the minimum welding deformation as optimization objectives to calculate the robot obstacle avoidance welding path.

[0133] The obstacle avoidance welding control unit controls the welding robot in the welding device to weld the workpiece to be welded, based on the above-mentioned real coordinate information and the above-mentioned robot obstacle avoidance welding path.

[0134] In an optional embodiment, the template matching module includes:

[0135] The first boundary feature information extraction unit extracts the first boundary feature information from the aforementioned target point cloud data.

[0136] The first vector generation unit calculates the first Hu moment data corresponding to the target point cloud data based on the aforementioned first boundary feature information, and uses it as the first vector.

[0137] The second vector group generation unit calculates the second Hu moment data corresponding to each workpiece model in the preset template workpiece library, uses it as the second vector, and combines all the second vectors to obtain the second vector group.

[0138] The target second vector determination unit calculates the distance between each second vector in the second vector group and the first vector, and takes the second vector with the smallest distance as the target second vector.

[0139] The target workpiece model determination unit takes the workpiece model corresponding to the second target vector as the target workpiece model and obtains the identification information corresponding to the workpiece to be welded from the target workpiece model.

[0140] In summary, the intelligent welding method and system for small assembly components in ships provided by this invention have at least the following advantages:

[0141] The intelligent welding method and system for small assembly components in ships disclosed in the above embodiments of the present invention realize fully automated welding of small assembly component production lines in ships. By pre-analyzing the digital models of various workpieces, identifying weld seams, and establishing a workpiece weld seam information table corresponding to each workpiece, the weld seam digital model coordinate information corresponding to the workpiece being welded can be extracted from the workpiece weld seam information table during welding. It is not necessary to obtain the weld seam by fitting point cloud data from on-site scanning, which can avoid the situation where the welding line cannot be accurately extracted due to factors such as occlusion during on-site scanning. Then, the weld seam digital model coordinates are transformed to the real coordinate system to obtain the actual weld seam position for welding, which helps to improve the welding accuracy of the welding robot and thus improve the welding quality.

[0142] On the other hand, since the workpiece model contains coordinate and type information of curves such as straight lines and arcs, it is also possible to extract arc welds.

[0143] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A smart welding method for small assembly components in ships, characterized in that, Includes the following steps: S110, scans the workpiece to be welded to obtain target point cloud data; S120, based on the target point cloud data, extract the first boundary feature information, and based on the first boundary feature information, match the target workpiece model corresponding to the workpiece to be welded from the preset template workpiece library; the preset template workpiece library stores a workpiece weld information table and template point cloud files corresponding to each workpiece model. S130, Based on the workpiece weld information table and the target workpiece model, obtain the weld digital model coordinate information corresponding to the workpiece to be welded; S140, based on the template point cloud file corresponding to the target workpiece model, obtain the second boundary feature information corresponding to the workpiece to be welded, and generate a rotation transformation matrix based on the first boundary feature information and the second boundary feature information; S150, according to the rotation transformation matrix, the digital model coordinate information is converted into real coordinate information associated with the weld; as well as S160, Based on the actual coordinate information, control the welding robot in the welding device to weld the workpiece to be welded.

2. The intelligent welding method as described in claim 1, characterized in that, Step S160 includes: Based on a preset path planning algorithm, with robot obstacle avoidance and multi-robot coordination as constraints, and with the shortest welding time and / or the minimum welding deformation as optimization objectives, the robot obstacle avoidance welding path is calculated. Based on the actual coordinate information and the robot's obstacle avoidance welding path, the welding robot in the welding device is controlled to weld the workpiece to be welded.

3. The intelligent welding method as described in claim 1, characterized in that, Step S120 includes: Based on the target point cloud data, calculate the corresponding first Hu moment data, which is used as the first vector; Calculate the second Hu moment data corresponding to each workpiece model in the preset template workpiece library, use it as the second vector, and combine all the second vectors to obtain the second vector group; Calculate the distance between each second vector in the second vector group and the first vector, and take the second vector with the smallest distance as the target second vector; The workpiece model corresponding to the target second vector is taken as the target workpiece model.

4. The intelligent welding method as described in claim 1, characterized in that, The step of generating a rotation transformation matrix based on the first boundary feature information and the second boundary feature information includes: Based on a preset point cloud library, the second boundary feature information is rotated to be consistent with the first boundary feature information, and a rotation transformation matrix is ​​generated.

5. The intelligent welding method as described in claim 1, characterized in that, Prior to step S110, the method further includes: Import the models corresponding to each of the multiple workpieces to create a preset template workpiece library; Based on the model corresponding to each workpiece, the weld information corresponding to each workpiece model is extracted, and a workpiece weld information table is generated. The workpiece weld information table is stored in the preset template workpiece library.

6. The intelligent welding method as described in claim 1, characterized in that, Step S160 includes: Before welding begins, the welding robot is controlled to perform a second positioning of the welding start and end points, and the coordinate information of the welding start and end points is corrected based on the second positioning results. During the welding process, the welding robot is controlled to track and detect changes in the weld seam, and the welding trajectory is corrected based on the tracking results; During the welding process, the actual welding formation data is collected in real time, and the welding process parameter database referenced by the welding robot during the welding process is corrected based on the actual welding formation data.

7. The intelligent welding method as described in claim 1, characterized in that, The welding device also includes a torch cleaning and wire cutting mechanism, which is used to clean the spatter generated by the welding robot during the welding process and stick to the gas protective sleeve of the welding torch, and to ensure that the extension length of the welding wire remains consistent.

8. The intelligent welding method as described in claim 1, characterized in that, The welding apparatus further includes a welding gantry, a welding torch, a line laser scanner, a laser tracking sensor, a camera module, an industrial control computer, and a host computer; the host computer is integrated into the industrial control computer; the welding gantry includes a guide rail, a gantry body, and a rotating platform, the guide rail and the rotating platform being connected to the gantry body respectively; the line laser scanner and the camera module are mounted on the gantry body; the gantry body can slide along the X direction on the guide rail; the rotating platform can slide along the Y direction on the gantry body; the welding torch is connected to the rotating platform, and the laser tracking sensor is mounted on the welding torch.

9. The intelligent welding method as described in claim 8, characterized in that, The welding device further includes a wire spool and a wire feeding mechanism. The wire spool and the wire feeding mechanism are respectively connected to the rotary platform. The wire spool is located above the wire feeding mechanism. The wire spool rotates synchronously with the rotation axis of the rotary platform. The wire output port of the wire spool is aligned with the wire input port of the wire feeding mechanism.

10. The intelligent welding method as described in claim 8, characterized in that, Both the rotary platform and the guide rail are driven by servo motors. The rotary platform uses an RV reducer for deceleration, and the guide rail uses a planetary reducer for deceleration.

11. An intelligent welding system for small assembly components in ships, used to implement the intelligent welding method as described in claim 1, characterized in that, include: The workpiece scanning module scans the workpiece to be welded to obtain target point cloud data; The template matching module extracts first boundary feature information from the target point cloud data, and matches the target workpiece model corresponding to the workpiece to be welded from a preset template workpiece library based on the first boundary feature information; the preset template workpiece library stores a workpiece weld information table and template point cloud files corresponding to each workpiece model; The weld seam digital model coordinate acquisition module acquires the weld seam digital model coordinate information corresponding to the workpiece to be welded based on the workpiece weld seam information table and the target workpiece model. The rotation transformation matrix generation module obtains the second boundary feature information corresponding to the workpiece to be welded based on the template point cloud file corresponding to the target workpiece model, and generates a rotation transformation matrix based on the first boundary feature information and the second boundary feature information. The weld seam real coordinate conversion module converts the digital model coordinate information into real coordinate information associated with the weld seam based on the rotation transformation matrix. as well as The welding execution module controls the welding robot in the welding device to weld the workpiece based on the actual coordinate information.

12. The intelligent welding system as described in claim 11, characterized in that, The welding execution module includes: The obstacle avoidance welding path planning unit, based on a preset path planning algorithm, uses robot obstacle avoidance and multi-robot coordination as constraints, and takes the shortest welding time and / or the minimum welding deformation as optimization objectives to calculate the robot obstacle avoidance welding path. The obstacle avoidance welding control unit controls the welding robot in the welding device to weld the workpiece to be welded, based on the real coordinate information and the robot's obstacle avoidance welding path.

Images