Saddle-shaped welding seam track welding posture automatic planning method and storage medium

Abstract

The invention discloses a saddle-shaped welding seam track welding posture automatic planning method and a storage medium, the method comprises the steps that standard welding posture points are stored in a robot in advance, and the standard welding posture points comprise spatial positions and posture angles; scanning the weldment through a visual system, recognizing an actual welding point, carrying out feature matching on the actual welding point and the standard welding posture point, calculating a compensation amount, correcting a standard track formed based on the standard welding posture point through the compensation amount, and obtaining a corrected actual point position; welding parameters are automatically adjusted based on the rule base; and a welding track adapting to the actual working condition is generated based on the corrected actual point position, the welding track and the welding parameters are output to the robot, and the robot executes welding operation. The method can adapt to manufacturing errors and installation deviations of different weldments, it is ensured that the welding gun is always in the optimal welding posture, welding parameters can be adjusted in a self-adaptive mode, actual welding seam conditions can be matched in real time, and the occurrence rate of welding defects is greatly reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of pipeline weld processing, in particular to a saddle-shaped weld trajectory welding posture automatic planning method and a storage medium. BACKGROUND

[0002] Pipeline saddle-shaped weld is widely used in the fields of oil and gas engineering, chemical equipment manufacturing, pressure vessel and boiler manufacturing, marine engineering, shipbuilding and nuclear power equipment, etc. It mainly appears at the intersection position of the main pipeline and the branch pipeline. Due to the intersection of two cylindrical pipelines in space forming a complex three-dimensional curve, the weld at the junction presents a special geometric feature similar to a saddle-shaped weld (Saddle Weld). In various industrial production systems, the connection of the main pipe and the branch pipe is a common and key structure form. For example, the branch connection of the oil and gas long-distance pipeline, the process branch of the chemical plant pipe network, the branch of the nuclear power plant cooling water system, the deck drainage and ventilation pipeline system of the ship, etc. all use this interface structure. The existence of the saddle-shaped weld makes the pipeline system more flexible and compact, while meeting the requirements of fluid transportation, structural support and process.

[0003] In the traditional manufacturing process, the preparation of the saddle-shaped weld depends on experienced technicians, who manually draw the weld contour line, cut the groove, and pair the welding. The process is tedious and prone to deviation. With the development of industrial automation and intelligent manufacturing technology, the existing technology directly calculates the welding posture in the robot coordinate system, where the welding posture refers to the spatial position and direction of the welding gun or tool in the welding process, including position (X, Y, Z) and angle (Rx, Ry, Rz). This method is more commonly used for simple welds, but less commonly used for complex welds. Because the robot position and angle in the coordinate system need to be finally used by the six joints of the robot to move, complex welds are prone to unsuccessful calculations or six-joint collisions. On the other hand, the existing technology can also use fixed welding postures, such as using a fixed set or several welding postures. This method can only be used for some specific welds and cannot be adapted to different specifications and models.

[0004] Based on the two cases of the existing technology, one uses the welding posture, which has a narrow application range and cannot automatically plan and adaptively adjust. The other automatically generates the welding posture, which can obtain results from mathematical logic, but has a large unattainable problem in robot operation, and has low practicality. Therefore, it is necessary to further research and improve the existing technology to solve the problems of the existing technology. SUMMARY

[0005] In order to solve the problems that the welding posture cannot be automatically planned and the welding parameters cannot be self-adaptively adjusted in the welding process, the present application provides a saddle-shaped welding seam track welding posture automatic planning method and a storage medium.

[0006] The first embodiment of the present application provides a saddle-shaped welding seam track welding posture automatic planning method, which comprises the following steps: pre-storing standard welding posture points in a robot, wherein the standard welding posture points comprise spatial positions and posture angles; the robot runs the standard welding posture points and synchronously collects actual running data in the running process, pre-processes the collected actual running data, and stores the processed actual running data into a database or a local storage as subsequent standard welding posture points; scanning a welding part through a visual system, identifying actual welding points of the welding part, performing feature matching on the actual welding points and the standard welding posture points, calculating a compensation amount, correcting a standard track formed based on the standard welding posture points through the compensation amount, and obtaining corrected actual point positions; automatically adjusting welding parameters based on a rule base; generating welding tracks adaptive to actual working conditions based on the corrected actual point positions, and outputting the generated welding tracks and welding parameters to the robot, so that the robot performs a welding operation according to the received welding tracks and welding parameters.

[0007] Optionally, the actual running data comprises actual positions, posture changes, and rotation angles of a positioner.

[0008] Optionally, the rule base at least comprises relationships between posture changes, welding part deviations, thickness changes, and welding parameters.

[0009] Optionally, the welding parameters at least comprise one or more of current, wire feeding speed, welding gun angle, and swing mode.

[0010] Optionally, the welding tracks at least comprise one or more of positions, postures, welding parameters, motion speeds, and welding processes of each path.

[0011] Optionally, the method of scanning the welding part through the visual system and identifying the actual welding points comprises the following steps: photographing the welding part at multiple angles through the visual system to obtain surface point cloud data; performing point cloud fusion on the point cloud data obtained through the multiple-angle photographing through spatial registration to form three-dimensional point cloud data, processing the three-dimensional point cloud data to identify geometric features of the welding part; extracting welding seam edge data and identifying a welding seam contour; performing secondary fitting and optimization on the extracted welding seam edge data, eliminating abnormal points, improving continuity and smoothness of the welding seam edge curve, and refining accurate positions of the welding seam points through a calibration method; and processing the welding seam edge curve to extract actual welding points.

[0012] Optionally, the visual system comprises a 3D camera or a laser scanner.

[0013] Optionally, the compensation amount includes a position compensation and a posture compensation.

[0014] A second embodiment of the present application provides a computer readable storage medium, wherein program instructions are stored in the computer readable storage medium, and the program instructions are used to execute the saddle-shaped weld seam trajectory welding posture automatic planning method according to any one of the preceding embodiments when running.

[0015] In the technical scheme provided by the embodiment of the present application, the standard welding posture points are stored in the robot in advance, then the welding part is scanned by the vision system, the actual welding points are identified, the actual welding points are matched with the standard welding posture points, the compensation amount is calculated, the standard trajectory formed based on the standard welding posture points is corrected by the compensation amount, the corrected actual point position is obtained, the welding parameters are automatically adjusted based on the rule base, the welding trajectory adaptive to the actual working condition is generated based on the corrected actual point position, the generated welding trajectory and welding parameters are output to the robot, and the robot performs the welding operation according to the received welding trajectory and welding parameters. The present application can realize the highly automated adaptive welding of the saddle-shaped welding part of the pipeline. And the present application has good adaptive capacity for different specifications and models. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 FIG. 1 is a flowchart of an embodiment of the saddle-shaped weld seam trajectory welding posture automatic planning method of the present application.

[0017] Figure 2 FIG. 2 is a flowchart of another embodiment of the saddle-shaped weld seam trajectory welding posture automatic planning method of the present application. DETAILED DESCRIPTION

[0018] The technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0019] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "connected", "connected" should be understood broadly, for example, it can be directly connected, or indirectly connected through an intermediate medium, or the connection between two elements inside. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances. In addition, the technical features involved in the different embodiments of the present application described below can be combined with each other as long as they do not conflict with each other.

[0020] The present application provides a saddle-shaped weld seam trajectory welding posture automatic planning method, please refer to Figure 1As shown, the method comprises:

[0021] Step S10, a standard welding pose point is pre-stored in the robot.

[0022] A complete set of standard welding pose points is pre-recorded in the system, the standard welding pose points include spatial position and pose angle, each pose point includes position (X, Y, Z), pose (Rx, Ry, Rz), and roll angle. The standard welding pose points recorded by the present application accurately cover the key sections of the saddle-shaped weld. The standard welding pose points can be recorded by manual teaching or digital modeling, or can be realized by automatic trajectory scanning modeling or directly importing CAD models.

[0023] Step S20, the robot runs the standard welding pose point and synchronously collects actual running data during running, pre-processes the collected actual running data, and stores the processed actual running data into a database or local storage as subsequent standard welding pose points.

[0024] The robot runs according to the standard welding pose point, and synchronously collects actual running data of each point during running, including actual position, pose change, and roll angle of the roll machine. The roll machine is a device used to rotate or adjust the pose of the welding part during the welding process, so that the welding part reaches a suitable pose for welding.

[0025] The collected actual running data is pre-processed by filtering, registration, and abnormality rejection. The pre-processed actual running data is stored in a database or local storage as subsequent standard welding pose points.

[0026] Step S30, the welding part is scanned by a vision system to identify the actual welding points of the welding part, the actual welding points are matched with the standard welding pose points, the compensation amount is calculated, the standard trajectory formed based on the standard welding pose points is corrected by the compensation amount, and the corrected actual point position is obtained. The vision system is a camera and image processing algorithm used to scan and identify the feature points on the surface of the welding part to obtain the actual data of the welding; the feature matching is to compare the standard data with the actually collected data to identify the similarity between them, so as to compensate or adjust.

[0027] Step S40, automatically adjust the welding parameters based on the rule base.

[0028] The rule base is used to guide the logical rule set of the welding parameter adjustment, which can adjust the welding parameters based on experience or model reasoning, and the rule base at least includes the relationship between the attitude change, the welding part deviation, the thickness change and the welding parameters. The welding parameters at least include one or more of the current, the wire feeding speed, the welding gun angle and the swing mode. According to the actual welding point (such as the attitude change, the welding part deviation, etc.), the welding parameters are dynamically adjusted, so that the welding process can adapt to various changes.

[0029] In step S50, the welding trajectory adapted to the actual working condition is generated based on the corrected actual point position.

[0030] The welding trajectory at least includes one or more of the position, the attitude, the welding parameter, the motion speed and the welding process of each path. The welding process includes arc extinguishing, arc collecting, arc striking transition section processing, etc. The generation of the welding trajectory of the present application is based on the actual shape of the welding part and the welding requirement, and a suitable welding path is calculated and executed in combination with the welding parameters.

[0031] In step S60, the generated welding trajectory and welding parameters are output to the robot, and the robot executes the welding operation according to the received welding trajectory and welding parameters.

[0032] In one embodiment of the present application, referring to Figure 2 As shown in the figure, step S30 specifically includes the following steps:

[0033] In step S31, the surface point cloud data of the welding part is obtained by multi-angle shooting through the vision system.

[0034] In one embodiment of the present application, the vision system includes a 3D camera. A high-precision 3D camera can shoot and generate high-resolution three-dimensional point cloud data camera equipment for accurate modeling and measurement. The point cloud is a data set composed of a large number of three-dimensional coordinate points, which is used to describe the spatial form of the object surface.

[0035] When the vision system is a 3D camera, the high-precision 3D camera is used to shoot the welding part to be welded from multiple angles to obtain comprehensive surface point cloud data. Reasonable exposure and collection parameter settings are used to ensure that the weld area contour is clear and the data loss caused by reflection or shadow is reduced.

[0036] The present application can also use a laser scanner instead of a 3D camera to collect point cloud data and obtain three-dimensional topographic data of the weld area for subsequent weld extraction and trajectory generation. The laser scanner has higher precision but slower speed, which is suitable for higher requirement scenarios.

[0037] In step S32, the point cloud data shot from multiple angles is fused through space registration to form three-dimensional point cloud data, and the three-dimensional point cloud data is processed to identify the geometric features of the welding part.

[0038] Point cloud fusion is the process of aligning and merging point cloud data from multiple sources to generate a complete three-dimensional model. Several features include planes, cylinders, spheres, and other features.

[0039] The present application fuses point cloud data obtained by shooting at different angles through spatial registration (such as ICP algorithm) to form unified and complete three-dimensional scene point cloud data. Through filtering, denoising, resampling and other preprocessing, the overall quality and continuity of the point cloud are further improved, laying a foundation for subsequent recognition and extraction.

[0040] The ICP algorithm (Iterative Closest Point Algorithm) is a commonly used point cloud registration method, which is used for spatial registration (alignment) of two point clouds based on iterative nearest point matching. The goal of the ICP algorithm is to minimize the distance between two sets of point clouds to accurately adjust the alignment of image data and point cloud data so that they coincide as accurately as possible in space. The ICP algorithm is an algorithm that optimizes the alignment of point cloud data by minimizing the distance between corresponding points in the point cloud, and finds the best transformation parameters (including rotation and translation) through iteration to align the point cloud data and target data as accurately as possible.

[0041] In one embodiment of the present application, the step of processing the three-dimensional point cloud data to identify the geometric features of the weld includes:

[0042] The three-dimensional point cloud data is processed by point cloud segmentation, and the geometric features of the segmented point cloud data are extracted to identify the geometric features of the weld.

[0043] Point cloud segmentation is the process of dividing point cloud into several regions or objects with specific meaning for further processing. Geometric feature extraction is the technology of extracting geometric properties (such as planes, cylinders, spheres, etc.) from point cloud data.

[0044] The present application identifies the geometric shape of the weld based on point cloud segmentation and geometric feature extraction technology (such as RANSAC algorithm). The present application randomly selects some points from a large number of matching points to determine whether the matching result meets the matching of most points. If some matching points are incorrectly paired, the RANSAC algorithm will automatically remove them to ensure that the remaining matching points can provide the most accurate alignment result.

[0045] Step S33, extract the weld edge data to identify the weld contour.

[0046] Specifically, the theoretical intersection line is calculated in combination with the fitting model of the welding part and compared with the actual point cloud, the real welding seam edge data is corrected and extracted. The boundary extraction algorithm is a process of identifying and extracting the boundary contour of an object in point cloud or image data.

[0047] The present application can also extract the welding seam contour by visual recognition, that is, using 2D image edge detection and depth compensation to extract the welding seam contour as the basis of the trajectory.

[0048] In step S34, the extracted welding seam edge data is subjected to secondary fitting and optimization, abnormal points are removed, the continuity and smoothness of the welding seam edge curve are improved, and the accurate position of the welding seam point is refined by the calibration method.

[0049] The present application performs secondary fitting and optimization on the preliminarily extracted welding seam edge, removes abnormal points, improves the continuity and smoothness of the welding seam curve. The correction method is introduced to refine the accurate position of the welding seam point, and ensure the accuracy of the trajectory generation.

[0050] In step S35, the welding seam edge curve is processed to extract the actual welding points.

[0051] According to the processed welding seam edge curve, the present application extracts welding point positions at equal intervals or at a certain density. Each point position includes spatial three-dimensional coordinates (X, Y, Z), providing basic data for subsequent pose matching and trajectory planning.

[0052] Other embodiments of the present application can also achieve pose matching through deep learning or feature curve fitting based algorithms.

[0053] The present application pre-stores standard welding pose points, then scans the welding part through a vision system, identifies the actual welding points, performs feature matching between the actual welding points and the standard welding pose points, calculates the compensation amount, corrects the standard trajectory formed based on the standard welding pose points through the compensation amount, obtains the corrected actual point positions, automatically adjusts the welding parameters based on the rule base, and generates a welding trajectory that adapts to the actual working condition based on the corrected actual point positions. The present application can realize highly automated adaptive welding of a saddle-shaped pipe welding part. And has good adaptive ability to different specifications and models.

[0054] One embodiment of the present application also provides a computer readable storage medium, the computer readable storage medium stores program instructions, the program instructions run to execute a saddle-shaped welding seam trajectory welding pose automatic planning method, the method comprises:

[0055] In step S10, standard welding pose points are pre-stored in a robot.

[0056] Step S20, the robot runs the standard welding posture point and synchronously collects actual operation data during running, pre-processes the collected actual operation data, and stores the processed actual operation data into a database or a local storage as the subsequent standard welding posture point.

[0057] Step S30, the welding part is scanned by a vision system, actual welding points of the welding part are recognized, feature matching is performed between the actual welding points and the standard welding posture points, a compensation amount is calculated, the standard track formed based on the standard welding posture points is corrected through the compensation amount, and the corrected actual point position is obtained.

[0058] Step S40, welding parameters are automatically adjusted based on a rule base.

[0059] Step S50, welding tracks adaptive to actual working conditions are generated based on the corrected actual point position.

[0060] Step S60, the generated welding tracks and welding parameters are output to the robot, and the robot performs welding operation according to the received welding tracks and welding parameters.

[0061] In one of the embodiments of the present application, referring to FIG. 1, step S30 specifically includes the following steps. Figure 2

[0062] Step S31, the welding part is photographed at multiple angles by a vision system, and surface point cloud data thereof is obtained.

[0063] Step S32, the point cloud data photographed at multiple angles is fused through space registration to form three-dimensional point cloud data, and the three-dimensional point cloud data is processed to recognize geometric features of the welding part.

[0064] Step S33, welding seam edge data is extracted, and a welding seam contour is recognized.

[0065] Step S34, the extracted welding seam edge data is subjected to secondary fitting and optimization, abnormal points are eliminated, continuity and smoothness of the welding seam edge curve are improved, and the accurate position of the welding seam point is refined through calibration.

[0066] Step S35, the welding seam edge curve is processed to extract actual welding points.

[0067] The present application has at least one of the following beneficial technical effects:

[0068] ​Significantly improve the stability of welding quality: The application can adapt to the manufacturing errors and installation deviations of different welding parts by standard welding posture point input and automatic matching (automatic identification of the difference between the actual form of the welding part and the standard data, and calculation of the required compensation), ensuring that the welding gun is always in the optimal welding posture. By self-adaptive adjustment of welding parameters, the actual welding seam conditions can be matched in real time, greatly reducing the occurrence rate of welding defects (such as incomplete fusion, undercut, porosity, etc.).

[0069] Improve the degree of welding automation: The application realizes automatic identification, automatic planning, automatic issuance and automatic execution, reduces manual demonstration and intervention, greatly improves the automation level, is suitable for complex saddle-shaped welding seam surfaces, does not depend on combing welders, and realizes self-adaptive welding of the equipment.

[0070] Shorten the welding preparation and debugging time: The traditional method needs manual demonstration or repeated debugging of the trajectory, and the application greatly shortens the trajectory writing and parameter adjustment time through standard posture welding points and automatic identification, improves the utilization rate of the equipment, and speeds up the process deployment.

[0071] Enhance the fault tolerance to welding part deformation and manufacturing errors: Through visual recognition and posture adaptive matching, even if the welding part has machining errors or thermal deformation, the welding path and parameters can be dynamically corrected to ensure the welding quality.

[0072] Realize standardized and data-based management of the welding process: Collect, process and store all welding-related data to facilitate subsequent welding process tracing, quality analysis and process optimization.

[0073] Improve the overall welding efficiency: The optimal adjustment of the welding trajectory and parameters improves the overall welding speed, reduces the rework rate, and thus improves the overall production efficiency. Compared with the traditional method, the beat is shorter and the overall cost is lower.

[0074] Have good expansibility and adaptability: It is suitable for saddle-shaped welds of different pipe diameters, different intersection angles and different materials, and can also be extended to other complex spatial welding application scenarios.

[0075] The above embodiments are only used to illustrate the technical solutions of the application, but not to limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method for automatically planning the welding posture of a saddle-shaped weld trajectory, characterized in that, The method includes: Standard welding posture points are pre-stored in the robot, and the standard welding posture points include spatial position and posture angle; The robot operates at standard welding posture points and simultaneously collects actual operating data during operation. The collected actual operating data is preprocessed and stored in a database or local storage as subsequent standard welding posture points. The weldment is scanned by a vision system to identify the actual welding point. The actual welding point is then matched with the standard welding posture point to calculate the compensation amount. The compensation amount is used to correct the standard trajectory formed based on the standard welding posture point to obtain the corrected actual position. Welding parameters are automatically adjusted based on a rule base. Welding trajectories adapted to actual working conditions are generated based on the corrected actual locations. The generated welding track and welding parameters are output to the robot, which then performs the welding operation based on the received welding track and welding parameters.

2. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 1, characterized in that, The actual operating data includes actual position, attitude changes, and positioner rotation angle.

3. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 1, characterized in that, The rule base includes at least the relationship between posture changes, weldment deviations, thickness changes, and welding parameters.

4. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 1, characterized in that, The welding parameters include at least one or more of the following: current, wire feed speed, welding torch angle, and oscillation mode.

5. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 1, characterized in that, The welding trajectory includes at least one or more of the following: position, posture, welding parameters, movement speed, and welding process for each segment of the path.

6. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 1, characterized in that, The method for scanning the weldment using a vision system to identify the actual welding points includes: The weldment is photographed from multiple angles using a vision system to obtain its surface point cloud data; Point cloud data captured from multiple angles is fused through spatial registration to form three-dimensional point cloud data. The three-dimensional point cloud data is then processed to identify the geometric features of the weldment. Extract weld edge data and identify weld contours; The extracted weld edge data is subjected to secondary fitting and optimization to remove outliers and improve the continuity and smoothness of the weld edge curve. The precise location of the weld points is refined through calibration. The actual welding points are extracted by processing the edge curve of the weld.

7. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 6, characterized in that, The vision system includes a 3D camera or a laser scanner.

8. The automatic planning method for welding posture of saddle-shaped weld seam trajectory according to claim 1, characterized in that, The compensation amount includes position compensation and attitude compensation.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program instructions, which, when executed, are used to perform the automatic planning method for welding posture of saddle-shaped weld seam trajectory as described in any one of claims 1-8.

Images