Online planning method and related equipment for welding gun posture of multi-layer and multi-pass external welding of pipelines

Abstract

This application discloses a method and related equipment for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines. This method belongs to the technical field of intelligent welding and includes obtaining a laser line image and welding gun posture data of the pipeline welding groove; extracting characteristic pixel points from the laser line image; converting it into a three-dimensional point cloud by detecting a mathematical model and fitting it to obtain a three-dimensional welding groove diagram; using the three-dimensional welding groove diagram and welding gun posture data to calculate the welding groove size parameters, welding gun posture parameters, and the posture parameters of the workpiece to be welded, thereby determining the welding gun posture planning value; performing the current pass of filling welding along the welding groove according to the welding gun posture planning value, while simultaneously detecting and correcting the real-time welding gun posture value in real time; replanning the welding gun posture parameters during the return process; and after completing multiple passes of filling welding, controlling the welding gun to perform a cover weld on the welding groove to complete the pipeline external welding. It can be seen that this method can achieve better welding results.

Description

Technical Field

[0001] The present application belongs to the technical field of intelligent welding, and in particular relates to an online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines and related equipment. Background Art

[0002] When welding the outside of a pipeline, since the welding groove on the outside of the pipeline is circular, semi-circular or arc-shaped, the pipeline will go through a continuous welding process from flat welding to vertical welding and then to overhead welding. Therefore, during the continuous welding process, the welding gun posture needs to be changed in real time to adapt to the current welding groove. In addition, due to problems such as groove processing deviation, inconsistent gap between pipe nozzle groups, and welding thermal deformation, it is necessary to observe the welding status in real time and make fine adjustments to the welding gun posture to ensure the accuracy of the weld formation position and reduce the occurrence of welding defects.

[0003] However, in the current technology of multi-layer and multi-pass external welding of pipelines, servo-controlled welding is often implemented based on online sensing. That is, scanning the welding groove, planning the welding gun posture and welding are carried out simultaneously. However, when the welding gun posture parameters are calculated based on the real-time collected data and the welding gun posture servo control is performed based on the welding gun posture parameters, there will be relative deviations in the welding gun posture parameters and the servo control is not stable enough, resulting in poor welding results. Summary of the Invention

[0004] The embodiment of the present application provides an online planning method and related equipment for the welding gun posture for multi-layer and multi-pass external welding of pipelines. Before each welding pass, the welding gun posture planning value can be planned first, and during welding, the real-time value of the current welding gun posture parameter can be corrected to obtain a better welding effect.

[0005] In a first aspect, an embodiment of the present application provides a method for online planning of welding gun posture for multi-layer and multi-pass external welding of a pipeline, the method comprising:

[0006] For each filler weld during external pipe welding, the laser line image and welding gun posture data of the pipe welding groove are obtained;

[0007] Extracting laser line image feature pixel points from the laser line image;

[0008] The characteristic pixel points of the laser line image are converted into a 3D point cloud by detecting the mathematical model and fitting is performed to obtain a 3D welding groove diagram. The welding groove size parameters, welding gun posture parameters and workpiece posture parameters to be welded are calculated using the 3D welding groove diagram and welding gun posture data.

[0009] Determine the welding gun posture planning value for the current pass based on the welding groove size parameters, welding gun posture parameters and the posture parameters of the workpiece to be welded;

[0010] Control the welding gun to perform the current fill welding along the welding groove according to the welding gun posture planning value, detect the real-time value of the welding gun posture parameter in real time during the filling welding process, and correct the real-time value of the welding gun posture parameter according to the welding gun posture planning value;

[0011] During the return process after each fill welding pass, the return laser line image and return welding gun posture data of the welding groove containing the weld bead are obtained, and the welding gun posture planning value of the next pass is planned using the return laser line image and return welding gun posture data;

[0012] After completing multiple passes of filling welding, control the welding gun to perform cover welding on the welding groove to complete the multi-layer and multi-pass external welding of the pipeline.

[0013] Furthermore, the laser line image and welding gun posture data of the pipeline welding groove are obtained, including:

[0014] The welding equipment is controlled to move along the track set on the pipeline, and multiple laser lines are projected toward the welding groove during the movement. The composite sensor in the welding equipment is used to obtain the laser line image of the welding groove and the welding gun posture data.

[0015] Among them, the welding groove size parameters include groove width, groove depth, groove surface angle and misalignment; the posture parameters of the workpiece to be welded include the inclination angle and yaw angle of the groove end face in the world coordinates; the welding gun posture parameters include lateral deviation, relative height, first posture angle and second posture angle.

[0016] Furthermore, the three-dimensional welding groove diagram and welding gun posture data are used to calculate welding groove size parameters, welding gun posture parameters and workpiece posture parameters to be welded, including:

[0017] Calculate the groove width, groove depth, groove face angle and misalignment using the 3D welding groove diagram;

[0018] In the preset camera coordinate system, the groove width, groove depth, groove face angle, misalignment, groove center position and groove end position in the 3D welding groove image are used to calculate the lateral deviation of the welding gun from the groove center, the relative height of the welding gun from the groove end, and the first and second posture angles of the welding gun relative to the groove end;

[0019] The inclination angle and yaw angle of the groove end face in the world coordinate system are calculated using the three-dimensional welding groove diagram and welding gun posture data.

[0020] Among them, the planned value of the welding gun posture includes the planned lateral deviation of the welding gun from the center of the groove, the planned relative height of the welding gun from the end face of the groove, the planned first posture angle and the planned second posture angle of the welding gun relative to the end face of the groove; the real-time value of the welding gun posture parameter includes the real-time lateral deviation of the welding gun from the center of the groove, the real-time relative height of the welding gun from the end face of the groove, the real-time first posture angle and the real-time second posture angle of the welding gun relative to the end face of the groove.

[0021] Furthermore, the real-time value of the welding gun posture parameter is corrected according to the welding gun posture planning value, including:

[0022] When the difference between the real-time lateral deviation and the planned lateral deviation is greater than a preset first limit, the real-time lateral deviation is corrected according to the planned lateral deviation;

[0023] When the difference between the real-time relative altitude and the planned relative altitude is greater than a preset second limit, the real-time relative altitude is corrected according to the planned relative altitude;

[0024] When the difference between the real-time first attitude angle and the planned first attitude angle is greater than a preset third limit, correcting the real-time first attitude angle according to the planned first attitude angle;

[0025] When the difference between the real-time second posture angle and the planned second posture angle is greater than a preset fourth limit, the real-time second posture angle is corrected according to the planned second posture angle.

[0026] Furthermore, in the return process after each fill welding pass, the return laser line image and return welding gun posture data of the welding groove containing the weld bead are obtained, and the return laser line image and return welding gun posture data are used to plan the welding gun posture planning value of the next pass, including:

[0027] Extracting characteristic pixel points of the return laser line image from the return laser line image;

[0028] The characteristic pixel points of the laser line image of the return journey are converted into a three-dimensional point cloud by detecting a mathematical model and fitting is performed to obtain the three-dimensional welding groove diagram of the return journey. The three-dimensional welding groove diagram of the return journey and the welding gun posture data of the return journey are used to calculate the welding groove size parameters, welding gun posture parameters and posture parameters of the workpiece to be welded of the return journey;

[0029] The welding gun posture planning value of the next pass is determined based on the welding groove size parameters, welding gun posture parameters and posture parameters of the workpiece to be welded in the return process to replace the welding gun posture planning value of the current pass.

[0030] Furthermore, the welding gun is controlled to perform cap welding on the welding groove to complete the multi-layer and multi-pass external welding of the pipeline, including:

[0031] According to the welding gun posture planning value during the last filling weld, the welding groove is capped and welded to complete the multi-layer and multi-pass external welding of the pipeline.

[0032] Furthermore, extracting characteristic pixel points of the laser line image from the laser line image includes:

[0033] Dividing the laser line image into multiple sub-images according to different pixel grayscale distribution characteristics in the laser line image;

[0034] Each sub-image is input into a pre-trained neural network, and the pre-trained neural network is used to extract the characteristic pixel points of the laser line image in each sub-image;

[0035] The laser line image feature pixel points of multiple sub-images are fused to obtain the laser line image feature pixel points.

[0036] In a second aspect, an embodiment of the present application provides an online planning device for welding gun posture for multi-layer and multi-pass external welding of a pipeline, the device comprising:

[0037] An acquisition module is used to acquire the laser line image and welding gun posture data of the pipeline welding groove for each filling weld during pipeline external welding;

[0038] An extraction module, used for extracting characteristic pixel points of the laser line image from the laser line image;

[0039] A calculation module is used to convert the characteristic pixel points of the laser line image into a three-dimensional point cloud by detecting a mathematical model and perform fitting to obtain a three-dimensional welding groove diagram. The three-dimensional welding groove diagram and welding gun posture data are used to calculate the welding groove size parameters, welding gun posture parameters, and posture parameters of the workpiece to be welded;

[0040] A planning module is used to determine the welding gun posture planning value of the current pass based on the welding groove size parameters, welding gun posture parameters and the posture parameters of the workpiece to be welded;

[0041] The correction module is used to control the welding gun to perform the current fill welding along the welding groove according to the welding gun posture planning value, detect the real-time value of the welding gun posture parameter in real time during the filling welding process, and correct the real-time value of the welding gun posture parameter according to the welding gun posture planning value;

[0042] The return module is used to obtain the return laser line image and return welding gun posture data of the welding groove containing the weld bead during the return process after each filling welding pass, and use the return laser line image and return welding gun posture data to plan the welding gun posture planning value for the next pass;

[0043] The completion module is used to control the welding gun to perform cover welding on the welding groove after completing multiple passes of filling welding, thereby completing the multi-layer and multi-pass external welding of the pipeline.

[0044] In a third aspect, an embodiment of the present application provides an electronic device, the device comprising:

[0045] a processor and a memory storing computer program instructions;

[0046] When the processor executes the computer program instructions, it implements the welding gun posture online planning method for multi-layer and multi-pass external welding of pipelines as described in any of the above items.

[0047] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, an online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines as described in any of the above items is implemented.

[0048] In a fifth aspect, an embodiment of the present application provides a method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines as described in any of the above items, when instructions in a computer program product are executed by a processor of an electronic device.

[0049] In the sixth aspect, an embodiment of the present application also provides a welding device, which includes a composite sensor, a welding robot and an industrial computer, wherein the composite sensor is used to collect laser line images and welding gun posture data, the welding robot is used to adjust the welding gun posture, and the industrial computer is used to execute any one of the above-mentioned welding gun posture online planning methods for multi-layer and multi-pass external welding of pipelines.

[0050] The welding gun posture online planning method and related equipment for multi-layer and multi-pass external welding of pipelines in the embodiment of the present application can extract characteristic pixel points of the laser line image based on the laser line image obtained at the welding groove during each filling weld.

[0051] Based on this, the characteristic pixels of the laser line image are converted into a three-dimensional point cloud by detecting the mathematical model, so that the three-dimensional welding groove diagram can be fitted and obtained. By using the three-dimensional welding groove diagram and the welding gun posture data, the welding groove size parameters, the posture of the workpiece to be welded and the welding gun posture parameters are calculated, so that the welding gun posture planning value can be obtained by using the above parameters.

[0052] Furthermore, during welding, the welding gun posture planning value can be directly used to control the welding gun for fill welding, and when the welding gun performs fill welding, the real-time value of the welding gun posture parameter can be detected in real time, and then the current real-time value of the welding gun posture parameter can be corrected according to the welding gun posture planning value, ensuring that the welding gun can perform welding operations in real time with the posture for the best welding effect during the welding process.

[0053] Furthermore, in the return process after each filling weld, for the welding groove containing the weld bead, by obtaining its return laser line image and return welding gun posture data, the welding gun posture planning value can be replanned according to the obtained return laser line image and return welding gun posture data, thereby replacing the welding gun posture planning value of the previous pass.

[0054] Furthermore, after completing the multi-pass filling welding, the multi-layer and multi-pass external welding of the pipeline can be completed by performing cap welding on the welding groove. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0056] Figure 1 This is a first schematic diagram of a welding device provided in an embodiment of the present application;

[0057] Figure 2 is a second schematic diagram of a welding device provided in an embodiment of the present application;

[0058] Figure 3 This is a first flow chart of a method for online planning of welding gun posture for multi-layer and multi-pass external welding of a pipeline provided in an embodiment of the present application;

[0059] Figure 4 Schematic diagram of a laser line image of an online planning method for welding gun posture for multi-layer and multi-pass external welding of a pipeline provided in an embodiment of the present application;

[0060] Figure 5 Schematic diagram of multi-pass welding of a welding groove according to an online planning method for welding gun posture for multi-layer and multi-pass external welding of a pipeline provided in an embodiment of the present application;

[0061] Figure 6 This is a second flow chart of a method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines provided in an embodiment of the present application;

[0062] Figure 7 This is a structural schematic diagram of a fitting welding groove in an online planning method for welding gun posture for multi-layer and multi-pass external welding of a pipeline provided in an embodiment of the present application;

[0063] Figure 8 This is a structural schematic diagram of an online planning device for welding gun posture for multi-layer and multi-pass external welding of pipelines provided in an embodiment of the present application;

[0064] Figure 9 This is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0065] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is merely to provide a better understanding of the present application by illustrating the examples of the present application.

[0066] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, the elements defined by the phrase "comprising..." do not exclude the presence of other identical elements in the process, method, article, or device comprising the elements.

[0067] As described in the background technology section, the related online planning technology for welding gun posture for multi-layer and multi-pass external welding of pipelines is still difficult to meet the needs of actual work.

[0068] In the process of welding multiple cylindrical pipes into one cylindrical pipe, the circular pipe openings of the two pipes need to be butted together, and the butt joint of the circular pipe openings needs to be welded externally.

[0069] Wherein, the butt joint of the externally welded circular pipe mouth is a welding groove.

[0070] Since the welding groove is circular, semicircular or arc-shaped formed around the pipeline, when performing external welding on the welding groove of the pipeline, it often goes through a continuous welding process from flat welding to vertical welding and then to overhead welding. Furthermore, during the continuous welding process, the welding gun posture needs to be changed in real time so that the welding gun posture can maintain an appropriate relative position and angle with the welding groove in real time, thereby ensuring the welding effect of the overall weld formation.

[0071] In addition, due to problems such as groove processing deviation, inconsistent gaps at the joints of circular pipe mouths, and welding thermal deformation, it is necessary to observe the welding status in real time and make fine adjustments to the welding gun posture to ensure the accuracy of the weld formation position and reduce the occurrence of welding defects.

[0072] However, in the process of implementing this application, it was found that in the current online planning technology of welding gun posture for multi-layer and multi-pass external welding of pipelines, servo-controlled welding is often implemented based on online sensing. That is to say, when performing external welding of pipelines at the butt joint of circular pipe mouths, the operations of scanning the welding groove and controlling the welding gun posture are performed simultaneously with the operation of external welding of the pipeline. However, the method of performing parameter calculation and welding gun posture servo control based on real-time collected data will lead to poor welding results. For example, when the welding gun posture calculated in real time is not good, it is difficult to correct it. For example, due to thermal deformation of welding, there is an error between the actual welding gun posture and the planned welding gun posture.

[0073] In order to solve the problems of the existing technology, the embodiment of the present application provides an online planning method and related equipment for the welding gun posture for multi-layer and multi-pass external welding of pipelines. The welding gun posture can be planned before welding and corrected during the welding process, so as to obtain a better welding gun posture servo control effect and thus obtain a high-quality weld formation effect.

[0074] The following is a detailed description of the online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines provided in an embodiment of the present application in conjunction with the accompanying drawings.

[0075] In one embodiment of the present application, welding equipment for performing external welding of a pipeline includes a composite sensor, a welding robot, a welding gun, and an industrial computer.

[0076] Among them, such as Figure 1 As shown, the composite sensor includes a visual sensor and a dual-axis tilt sensor. The visual sensor is fixedly connected to the dual-axis tilt sensor. The composite sensor is used to collect relevant data of the welding groove and the welding gun, for example, collecting laser line images and welding gun posture data.

[0077] The dual-axis tilt sensor may be, for example, a gravity sensor.

[0078] The gravity sensor is fixedly connected to the welding gun through a connector and is used to detect the two angles between the welding gun and the direction of gravity, thereby obtaining the absolute posture of the welding gun and the welding robot as a whole.

[0079] Welding robots can be Figure 1 The welding carriage shown in FIG is fixedly connected to the welding gun.

[0080] The welding robot can include 5 degrees of freedom of motion, for example, the walking degree of freedom D v , horizontal lateral freedom D e , Height adjustment freedom D h , inclination degree of freedom D α and angular pendulum degrees of freedom D βBy adjusting the above five degrees of freedom of motion, the welding robot can be used to adjust the position of the welding gun, so that the welding gun can be placed in any position that meets the welding process requirements during the full-position welding process of the pipeline.

[0081] like Figure 1 As shown, the welding equipment may further include a track for use with the welding carriage, the track being arranged around the pipe to be welded, and the welding equipment may further include a wire feeding reel and the like.

[0082] In this application, an industrial computer is used to control the movement of the welding robot on the track, and use the relevant data collected by the composite sensor to calculate the required parameters, and control the movement of the welding robot and the welding gun posture according to the parameters, thereby completing the online planning method of the welding gun posture for multi-layer and multi-pass external welding of pipelines in any of the following embodiments.

[0083] like Figure 2 As shown, the visual sensor in the composite sensor is a dual-line structured light visual sensor, which includes a camera with a lens and two straight-line laser emitters. The straight-line laser emitters are fixedly connected to the gravity sensor through a laser emitter fixing member.

[0084] Among them, the straight line laser transmitter is used to project laser lines onto the pipes to be welded and the welding grooves.

[0085] Specifically, the straight-line laser emitter adopts the oblique-direct reception mode, and the angle between the straight-line laser emitter and the camera optical axis is 30°. The camera in the visual sensor is used to capture the deformed laser line projected onto the surface of the pipe to be welded and the groove during the welding process to obtain a laser line image.

[0086] The visual sensor is placed on the side of the welding gun in the forward direction, and the central axis of the welding gun, the optical axis of the camera and the central axis of the two straight line laser emitters are coplanar. The bottom surface of the gravity sensor is installed perpendicular to the optical axis of the camera, and its width direction, that is, along the X D The direction of the axis is parallel to the width of the camera target surface.

[0087] Figure 3 A schematic diagram of a scenario of an online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines provided by an embodiment of the present application is shown.

[0088] refer to Figure 3 In one embodiment of the present application, a method for online planning of welding gun posture for multi-layer and multi-pass external welding of a pipeline is applied to an industrial computer in welding equipment and specifically includes the following steps:

[0089] Step S301: For each filling weld during external welding of a pipeline, a laser line image and welding gun posture data of the welding groove of the pipeline are obtained.

[0090] Among them, the laser line image is an image of the welding groove taken when a straight line laser emitter projects a laser line onto the welding groove; the welding gun posture data can be, for example, two angles formed by the welding gun and the direction of gravity in three-dimensional space.

[0091] In this embodiment, when externally welding the weld groove at the butt joint of the pipe ends, multiple welding layers can be formed by performing multiple passes of filling welding and at least one pass of cap welding.

[0092] When performing each pass of filling welding on the welding groove, the composite sensor in the aforementioned embodiment can collect the laser line image and welding gun posture data of the welding groove position.

[0093] in, Figure 4 A captured laser line image is shown.

[0094] Figure 4 Specifically, it shows that when the line laser emitter 1 and the line laser emitter 2 are projected onto the welding groove, a deformed laser line is formed.

[0095] In one example, since the dual-axis inclination sensor, that is, the gravity sensor, is fixedly connected to the welding gun, and the welding gun is also fixedly connected to the welding robot, the dual-axis inclination sensor can be used to detect the posture data of the welding gun and the welding robot.

[0096] Furthermore, the industrial computer can obtain the laser line image and welding gun posture data collected by the composite sensor.

[0097] Step S302: extracting laser line image feature pixel points from the laser line image.

[0098] In this embodiment, the industrial computer can extract characteristic pixel points of the laser line image suitable for the following steps based on the laser line image acquired in the previous step S301.

[0099] Step S303: Convert the characteristic pixel points of the laser line image into a three-dimensional point cloud by detecting a mathematical model and perform fitting to obtain a three-dimensional welding groove diagram; use the three-dimensional welding groove diagram and the welding gun posture data to calculate the welding groove size parameters, welding gun posture parameters, and posture parameters of the workpiece to be welded.

[0100] The welding groove includes a groove bottom surface, multiple groove end surfaces and multiple groove surfaces, and two groove surfaces are connected at the bottom intersection line.

[0101] Based on this, the welding groove size parameters specifically include groove width, groove depth, groove surface angle and misalignment, as well as the various positions of the welding groove, such as the size data of the groove bottom surface, multiple groove end faces and multiple groove faces.

[0102] The posture parameters of the workpiece to be welded refer to the inclination angle and yaw angle data of the groove end face of the workpiece in the world coordinate system;

[0103] The welding gun posture parameters may be data of the position and posture of the welding gun relative to the welding groove in the camera coordinate system, as well as data of the posture in the world coordinate system.

[0104] In this embodiment, based on the acquired characteristic pixel points of the laser line image, the characteristic pixel points of the laser line image are converted into a three-dimensional point cloud by detecting a mathematical model, and then fitting is performed to obtain a three-dimensional welding groove diagram.

[0105] Furthermore, the three-dimensional welding groove diagram can be used to determine the welding groove size parameters.

[0106] Since the welding groove size parameters are obtained by using the laser line image captured by the camera, the welding groove size parameters are data in the camera coordinate system.

[0107] In this embodiment, since the acquired welding gun posture data is obtained by using a dual-axis tilt sensor, the acquired welding gun posture data is the absolute posture of the welding gun in the world coordinate system.

[0108] Based on this, the welding gun posture data in the world coordinate system can be converted into the camera coordinate system, and the welding groove size parameters, position and posture data can be used to calculate the welding gun posture parameters.

[0109] Step S304: determining a welding gun posture planning value for the current pass based on the welding groove size parameters, welding gun posture parameters, and the posture parameters of the workpiece to be welded.

[0110] The welding gun posture planning value is the posture of the welding gun at various positions in the welding groove when performing the filling welding along the welding groove before each filling welding pass.

[0111] In this embodiment, before using a welding gun to weld the welding groove, the industrial computer can use the welding groove size parameters, the workpiece posture parameters to be welded and the current welding gun posture parameters determined in the previous steps to predict the welding gun posture planning value for the current fill welding.

[0112] Step S305: Control the welding gun to perform the current pass of filling welding along the welding groove according to the welding gun posture planning value, detect the real-time value of the welding gun posture parameter in real time during the filling welding process, and correct the real-time value of the welding gun posture parameter according to the welding gun posture planning value.

[0113] In this embodiment, based on the welding gun posture planning value predicted in the aforementioned steps, the current fill welding pass can be started for the welding groove.

[0114] Furthermore, after starting the current fill welding, during the current fill welding, the industrial computer can control the welding gun and the welding robot to move along the welding groove and perform fill welding on the welding groove.

[0115] Among them, during the process of filling welding along the welding groove, the industrial computer can use the predicted welding gun posture planning value to control the welding gun posture in real time.

[0116] Furthermore, during the process of filling welding along the welding groove, the welding groove size parameters can be obtained in real time, and the welding gun posture parameters can be detected in real time and used as the real-time values ​​of the welding gun posture parameters.

[0117] Furthermore, based on the currently executed welding gun posture planning value and the real-time value of the welding gun posture parameter calculated by detection, the welding gun posture planning value can be used to correct the current real-time value of the welding gun posture parameter, so that during the filling welding process of the current pass, the welding gun posture can be maintained at the ideal welding gun posture planning value in real time.

[0118] In one example, the welding gun posture parameters can be obtained in the aforementioned steps. During the filling welding of the current pass, a straight-line laser emitter can be used to project a laser line to the welding groove in real time, and a camera can be used to capture the laser line image of the welding groove. This can construct a real-time three-dimensional welding groove diagram, and a dual-axis inclination sensor can be used to obtain real-time welding gun posture data.

[0119] Based on this, the real-time three-dimensional welding groove diagram and the real-time welding gun posture data can be used to calculate the current posture parameters of the welding gun, and use them as the real-time values ​​of the welding gun posture parameters.

[0120] Furthermore, the welding gun posture planning value determined above can be used to correct the real-time value of the welding gun posture parameter, so that the welding gun can maintain an appropriate real-time value of the welding gun posture parameter.

[0121] Step S306: In the return process after each fill welding pass, a return laser line image and return welding gun posture data of the welding groove containing the weld bead are obtained, and the return laser line image and the return welding gun posture data are used to plan the welding gun posture planning value of the next pass.

[0122] Among them, after completing each pass of filling welding, the welding gun needs to return to the starting point of the filling welding, and the process of returning to the starting point of the filling welding is regarded as a return stroke.

[0123] In this embodiment, during each return stroke of the welding gun, the composite sensor can be used again to obtain the laser line image of the welding groove and the welding gun posture data during the return stroke, and use them as the return laser line image and the return welding gun posture data respectively.

[0124] Furthermore, characteristic pixel points of the return laser line image may be extracted from the return laser line image.

[0125] Furthermore, by using the aforementioned detection mathematical model, the characteristic pixel points of the return laser line image can be converted into a three-dimensional point cloud, and the three-dimensional welding groove diagram in the return process can be obtained by fitting.

[0126] Furthermore, based on the three-dimensional welding groove diagram and the return welding gun posture data in the return process, the return welding groove size parameters, the return welding gun posture parameters and the return workpiece posture parameters to be welded in the return process are calculated.

[0127] Based on this, the return welding groove size parameters, return welding gun posture parameters and return workpiece posture parameters can be used to calculate the welding gun posture planning value for the next filling welding pass.

[0128] Step S307: After completing multiple passes of filling welding, control the welding gun to perform cap welding on the welding groove to complete the multi-layer and multi-pass external welding of the pipeline.

[0129] The welding of the weld groove of the weld specifically includes multiple passes of filling welding and one or more passes of cap welding.

[0130] In this embodiment, after one pass of filling welding is completed on the welding groove, a layer of filling weld is formed in the welding groove. After multiple passes of filling welding are completed on the welding groove, multiple layers of filling welds are formed in the welding groove.

[0131] After completing multiple passes of filling welding, when the weld groove is filled with multiple layers of filling welds, one or more passes of cap welding can be performed on the position of the last layer of filling weld in the weld, thereby completing the multi-layer and multi-pass external welding of the pipeline.

[0132] exist Figure 5 In the example shown, the hot welding is the first pass of the filling weld, and thereafter 5 passes of the filling weld are performed, thereby forming 6 layers of filling welds by performing 6 passes of the filling weld operation, and after forming the 6th layer of the filling weld, by performing 2 passes of the capping weld, that is, Figure 5 The seventh cover weld C1 and the eighth cover weld C2 form the 7th layer of cover weld.

[0133] Among them, the multi-layer filling weld formed can be divided into deep layer F1, middle layer F i and shallow F n , among which, the last completed filling weld layer is the shallow layer, and the first completed filling weld layer is the deep layer or the middle layer.

[0134] Based on this, in this embodiment, when performing each filling weld, based on the laser line image at the obtained welding groove, the characteristic pixel points of the laser line image can be extracted therefrom, so that after the characteristic pixel points are converted into a three-dimensional point cloud and fitted by detecting the mathematical model, a complete three-dimensional welding groove map can be obtained, and based on the welding gun posture data obtained from the gravity sensor, it is possible to calculate the welding groove size parameters, the posture parameters of the workpiece to be welded and the welding gun posture parameters before starting the filling weld, so that the welding groove size parameters, the posture parameters of the workpiece to be welded and the welding gun posture parameters can be used to obtain the welding gun posture planning value.

[0135] Furthermore, when performing each fill welding pass, the welding gun posture planning value can be directly used to control the welding gun to perform the fill welding of that pass, and when the welding gun performs the fill welding, the real-time value of the welding gun posture parameter can be calculated through real-time detection, and then the current real-time value of the welding gun posture parameter can be corrected according to the welding gun posture planning value, ensuring that the welding gun can perform the welding operation in real time with the posture for the best welding effect during the welding process.

[0136] Furthermore, in the return process after each filling weld, for the welding groove containing the weld bead, by obtaining its return laser line image and return welding gun posture data, the welding gun posture planning value can be replanned according to the obtained return laser line image and return welding gun posture data, thereby replacing the welding gun posture planning value of the previous pass.

[0137] Accordingly, after completing the multi-pass filling welding, the multi-layer and multi-pass external welding of the pipeline can be completed by performing cover welding on the welding groove.

[0138] In another embodiment of the present application, in the process of obtaining the laser line image and welding gun posture data of the welding groove of the pipeline, the welding equipment can be controlled to move along a track set on the pipeline, and during the movement, one or more laser lines can be projected toward the welding groove in the weld bead, so that the welding groove with the laser line can be photographed by the visual sensor in the composite sensor to obtain the laser line image, and the welding gun posture data can be obtained by using the dual-axis inclination sensor in the composite sensor.

[0139] In this embodiment, Figure 6 An example of an online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines is shown.

[0140] in, Figure 6 The circle on the left side of the middle represents the cross-sectional annular surface at the circular pipe mouth of the pipe. Figure 1 As shown, the welding groove is along Figure 6The cross-sectional annular surface is surrounded by a track surrounding the cross-sectional annular surface at a preset distance from the welding groove, so that the welding robot can move around the cross-sectional annular surface along the welding groove.

[0141] In this embodiment, the cross-sectional annular surface can be divided into two parts: clockwise from point A to point B and counterclockwise from point A to point B, and two welding devices are used to complete the multi-layer and multi-pass pipeline external welding clockwise from point A to point B and the multi-layer and multi-pass pipeline external welding counterclockwise from point A to point B respectively.

[0142] Among them, take the multi-layer and multi-pass pipeline external welding from point A to point B clockwise as an example. Figure 6 As shown, during the pre-detection process, the industrial computer can control the welding robot to move clockwise from point A to point B along the welding groove around the cross-sectional annulus.

[0143] When the industrial computer controls the welding robot to move clockwise from point A to point B, the industrial computer can control two straight-line laser emitters to project two laser lines toward the welding groove.

[0144] Based on this, in the process of moving clockwise from point A to point B, the industrial computer can use the visual sensor to capture images of multiple welding grooves and use them as multiple laser line images.

[0145] At the same time, the industrial computer can also use the gravity sensor to obtain the two angles between the welding gun and the direction of gravity, and use them as welding gun posture data.

[0146] Based on this, in this embodiment, the industrial computer controls the movement of the welding equipment on the track of the pipeline, thereby realizing movement along the welding groove in the direction surrounding the pipeline, and then continuously projecting multiple laser lines to the welding groove. Accordingly, by continuously photographing the welding groove, multiple laser line images of the welding groove can be obtained, and at the same time, the gravity sensor can be used to obtain the welding gun posture data.

[0147] In another embodiment of the present application, Figure 6 In the pre-detection process in which the welding robot moves clockwise from point A to point B, the industrial computer can extract characteristic pixel points of the laser line image from each acquired laser line image.

[0148] In this embodiment, for each acquired laser line image, the laser line image can be divided into multiple regions of interest (ROIs) according to different pixel grayscale distribution characteristics in the laser line image, where each ROI is used as a sub-image, and the laser line feature pixel points in each sub-image are extracted.

[0149] Specifically, Figure 4 As an example, based on the laser line image Figure 4 The vertical grayscale projection value and horizontal grayscale projection value of the laser line image shown can be Figure 4 The laser line in the laser line image shown is divided into three ROIs.

[0150] ROI1 includes the laser line at the end face of the groove, ROI2 includes the laser line at the groove surface at the upper part of the welding groove after the laser line is deformed, and ROI3 includes the laser line on the groove surface at the bottom of the welding groove.

[0151] Furthermore, feature extraction can be performed on the three ROIs respectively, so as to obtain the laser line feature pixel points corresponding to each ROI.

[0152] Based on this, the extracted laser line feature pixel points of each sub-image can be fused to obtain the laser line image feature pixel points of the complete laser line image.

[0153] In this embodiment, when the bottom of the welding groove has been filled with a filling weld layer, the imaging of the laser line projected in ROI3 is approximately deformed into a quadratic curve, so that there are obvious differences in the image features of the three ROIs. Therefore, in this embodiment, by dividing the laser line image into multiple ROIs and performing independent feature extraction on each ROI, the real-time and accuracy of the overall image processing can be improved.

[0154] In another embodiment of the present application, in the process of extracting the laser line feature pixel points of each sub-image and fusing the laser line feature pixel points, a pre-trained neural network can be used to extract and fuse the feature pixel points to obtain the laser line image feature pixel points of the complete laser line image.

[0155] The pre-trained neural network may specifically be, for example, an image segmentation network (U-NET network). The U-NET network includes an encoder and a decoder. The encoder is used for downsampling, and the decoder is used for upsampling.

[0156] In this embodiment, each sub-image can be input into the U-NET network separately, and the encoder extracts feature pixel points for each sub-image separately, and the feature pixel points extracted by the encoder are input into the decoder, and the decoder interpolates and reconstructs each feature pixel point and outputs it, thereby obtaining the laser line feature pixel points of the sub-image.

[0157] Furthermore, according to the positional relationship of each sub-image in the complete laser line image, the laser line feature pixel points of each sub-image are fused to obtain the laser line image feature pixel points of the complete laser line image.

[0158] The characteristic pixel points of the laser line image can specifically represent the shape of the welding groove.

[0159] Based on this, in this embodiment, by utilizing the pre-trained U-NET network, the laser line feature pixel points of each sub-image can be accurately extracted, and the laser line feature pixel points can be fused through the positional relationship between the sub-images to obtain the laser line image feature pixel points representing the shape of the welding groove.

[0160] In another embodiment of the present application, before using the pre-trained U-NET network to extract and fuse feature pixels, the U-NET network to be trained can be trained to obtain a trained U-NET network.

[0161] In this embodiment, a training image of the pipeline laser line may be obtained first, where the training image includes laser markings.

[0162] For example, 1,000 laser line images collected during the welding process can be used, and each laser line image can be divided into multiple sub-images. The laser lines in each sub-image are marked, and the images are normalized, so that the normalized sub-images obtained are used as training images, wherein each training image can be, for example, 100*2000 pixels.

[0163] Furthermore, the training image is input into the U-NET network to be trained, the encoder in the neural network to be trained is used to extract the feature pixels of the training image, and the decoder in the U-NET network to be trained is used to output, for example, multiple images of size 112*2000 pixels, and each image is interpolated and reconstructed into 100*2000 pixels, thereby obtaining the training result.

[0164] Furthermore, the training results and laser markings are input into a preset cross entropy loss function for calculation to obtain the loss value.

[0165] In this embodiment, a loss threshold may be preset, for example, 0.05.

[0166] Based on this, when the loss value is greater than the preset loss threshold, the parameters of the U-NET network to be trained are adjusted, and each training image is input into the adjusted neural network for training until the loss value of the adjusted neural network is less than or equal to the loss threshold, and a U-NET network that has completed training is obtained.

[0167] Based on this, in this embodiment, through the pre-set training images and the laser mark set for each training image, the U-NET network to be trained can be used to extract features of each training image, and after the cross entropy calculation of the training results and the laser mark, it is determined whether the U-NET network to be trained has completed the training. In this way, if the training is not completed, the parameters of the U-NET network to be trained can be adjusted to achieve re-training of the adjusted U-NET network, and a U-NET network that has completed the training can be obtained.

[0168] In another embodiment of the present application, Figure 6 In the pre-detection process of the welding robot moving from point A to point B shown in the figure, for each extracted laser line image feature pixel point, the industrial computer can use the detection mathematical model to convert it into a three-dimensional point cloud, and fit the three-dimensional point cloud corresponding to the laser line image feature pixel point to obtain a three-dimensional welding groove diagram.

[0169] In this embodiment, based on the acquired characteristic pixel points of the laser line image, the laser line image may be segmented first.

[0170] Specifically, the two-dimensional laser line image feature may be divided into multiple data segments according to the inflection points of the laser line.

[0171] Furthermore, each data segment is mapped into a three-dimensional point cloud.

[0172] Furthermore, the 3D point clouds of each data segment are reconstructed as follows Figure 7 The 3D weld groove diagram corresponding to the laser line image is shown.

[0173] Among them, such as Figure 7 As shown, the obtained three-dimensional welding groove image specifically shows the two groove end faces of the welding groove corresponding to the laser image: the fitting groove end face S1 and the fitting groove end face S6; the four groove surfaces: the fitting groove surface S2, the fitting groove surface S3, the fitting groove surface S4 and the fitting groove surface S5; and the parallel bottom surface S7.

[0174] like Figure 7 As shown, the fitting groove surface S2 and the fitting groove surface S5 are the side surfaces of the upper portion of the welding groove, and the fitting groove surface S3 and the fitting groove surface S4 are at the bottom of the welding groove.

[0175] Based on this, in this embodiment, the shape of the two-dimensional welding groove laser line image can be fitted into a three-dimensional image through the three-dimensional point cloud, that is, a three-dimensional welding groove diagram, so that the specific situation of the welding groove can be displayed more accurately.

[0176] In another embodiment of the present application, Figure 6In the pre-detection process in which the welding robot moves clockwise from point A to point B, the welding groove size parameters, the workpiece posture parameters to be welded, and the welding gun posture parameters can be calculated based on the fitted three-dimensional welding groove diagram and the acquired welding gun posture data.

[0177] Among them, the welding groove size parameters include groove width, groove depth, groove face angle and misalignment, as well as the various positions of the welding groove, such as the size data of the groove bottom surface, multiple groove end surfaces and multiple groove surfaces; the posture parameters of the workpiece to be welded include the inclination angle α of the groove end surface of the workpiece in the world coordinate system. W and yaw angle β W ; The welding gun posture parameters include the lateral deviation of the welding gun from the center of the groove, the relative height of the welding gun from the end face of the groove; the first posture angle α and the second posture angle β of the welding gun relative to the end face of the groove.

[0178] In this embodiment, based on the three-dimensional welding groove diagram determined in the above embodiment, the groove width, groove depth, groove face angle and misalignment, as well as the position of the groove center and the position of the groove end face can be calculated.

[0179] Specifically, the coordinates of each position of the welding groove can be calculated according to the following formula (1), so that the groove width, groove depth, groove face angle and misalignment, as well as the position of the groove center and the position of the groove end face can be obtained:

[0180] (1)

[0181] Among them, (X C , Y C , Z C ) is the three-dimensional coordinate of the laser projection point on the surface of the workpiece to be welded in the camera coordinate system, that is, the three-dimensional coordinate of each point in the welding groove in the three-dimensional welding groove image, (u, v) is the coordinate of each pixel in the three-dimensional welding groove image; f x , f y , u0 and v0 are both camera internal parameters, A i , B i , C i and D i They are all equation parameters of the laser plane projected by laser i in the camera coordinate system. When i=1, it represents a line laser emitter 1, and when i=2, it represents a line laser emitter 2. The camera intrinsic parameters and equation parameters together constitute the intrinsic parameters of the visual sensor.

[0182] Based on this, the coordinates of each point in the three-dimensional welding groove diagram can be determined using formula (1), thereby determining the groove width, groove depth, groove surface angle and misalignment in the camera coordinate system, as well as the position of the groove center and the position of the groove end face.

[0183] Furthermore, based on the calculated position of the groove center, the lateral deviation e of the welding gun from the groove center can be calculated in the camera coordinate system; based on the calculated position of the groove end face, the height H of the welding gun from the groove end face can be calculated in the camera coordinate system; and based on the calculated groove width, groove depth, groove face angle and misalignment, the first attitude angle α and the second attitude angle β of the welding gun relative to the groove end face can be calculated in the camera coordinate system.

[0184] Specifically, the lateral deviation e of the welding gun from the groove center in the camera coordinate system can be calculated using the following formula (2):

[0185] (2)

[0186] Among them, M represents Figure 7 Any point on the bottom intersection line l1, its coordinates are (X CM , Y CM , Z CM The intersection of the camera optical axis and the parallel bottom surface S7 is Q, and the coordinates of Q are (0, 0, ); MQ represents the connecting vector between two coordinate points.

[0187] Furthermore, the relative height H between the welding gun and the groove end face in the camera coordinate system can be calculated using the following formula (3):

[0188] (3)

[0189] Among them, the relative height H represents the distance from the camera focus along the optical axis to the groove surface, C p and D p are the coefficients in the plane equation of the fitting groove plane S1 and the plane equation of the fitting groove plane S6.

[0190] Furthermore, the first posture angle α of the welding gun relative to the groove end face in the camera coordinate system can be calculated using the following formula (4):

[0191] (4)

[0192] Among them, (X v , Y v , Z v ) represents the normal vector n of the groove end face fitted in the camera coordinate system v When the α value is positive, it means that the welding gun is tilted forward relative to the visual sensor. When the α value is negative, it means that the welding gun is tilted forward or backward relative to the visual sensor.

[0193] Furthermore, the second posture angle β of the welding gun relative to the groove end face in the camera coordinate system can be calculated using the following formula (5):

[0194] (5)

[0195] Among them, when the β value is positive, it means that the welding gun swings to the left relative to the visual sensor, and when the β value is negative, it means that the welding gun swings to the right relative to the visual sensor.

[0196] Furthermore, the coordinates in the camera coordinate system can be converted to the world coordinate system using the following formula (6):

[0197] (6)

[0198] Among them, θ represents the visual sensor relative to the world coordinate system Y measured by the dual-axis tilt sensor, that is, the gravity sensor w The rotation angle of the axis, ω represents the visual sensor relative to the world coordinate system X measured by the dual-axis inclination sensor, that is, the gravity sensor w Yaw angle of the axis; n w Represents the normal vector n of the fitted groove end face v The expression in the world coordinate system (X W , Y W , Z W ).

[0199] Furthermore, the inclination angle α of the groove end face of the workpiece to be welded in the world coordinate system can be calculated using the following formula (7): W :

[0200] (7)

[0201] The following formula (8) can be used to calculate the yaw angle β of the groove end face of the workpiece to be welded in the world coordinate system: W :

[0202] (8)

[0203] Based on this, in this embodiment, the welding gun posture parameters including lateral deviation, relative height, first posture angle and second posture angle can be calculated based on the three-dimensional welding groove diagram, thereby completing the pre-detection operation.

[0204] In another embodiment of the present application, after completing the pre-detection, the industrial computer can control the welding robot to Figure 7 Point B in the figure moves back to point A, that is, Figure 7In the return 0, the industrial computer can plan the welding gun posture planning value for the filling welding based on the determined welding groove size parameters, the workpiece posture parameters to be welded and the welding gun posture parameters before performing the filling welding of the current pass.

[0205] In this embodiment, in return 0, the industrial computer can use another pre-trained neural network to predict the welding gun posture planning value, wherein the other pre-trained neural network can be, for example, a recurrent neural network (RNN).

[0206] Specifically, the welding gun posture parameters and the welding groove size parameters in each laser line image determined in the aforementioned embodiment can be input into the RNN network, and the RNN network can be used to predict the welding groove size parameters, the posture parameters of the workpiece to be welded and the welding gun posture parameters in each laser line image, so as to obtain the change in the welding gun posture when the welding gun moves along the welding groove, that is, the welding gun posture planning value.

[0207] In this embodiment, the architecture of the RNN network may adopt, for example, a long short-term memory network architecture (Long Short-Term Memory architecture, LSTM architecture).

[0208] Among them, the predicted welding gun posture planning value is specifically used to control the welding gun during the filling welding process of the corresponding pass, so that it can maintain a predetermined angle with the weld with the required welding process, thereby obtaining an ideal welding effect.

[0209] Based on this, in this embodiment, after completing the pre-detection, by inputting the determined welding gun posture parameters and welding groove size parameters into the pre-trained RNN network during the return process 0, it is possible to pre-plan the welding gun posture planning values ​​during the filling welding process of each pass before performing the filling welding of the pass.

[0210] In another embodiment of the present application, before using the pre-trained RNN network to predict the welding gun posture planning value, the RNN network to be trained can be trained to obtain a fully trained RNN network.

[0211] In this embodiment, multiple sets of training data of welding gun posture parameters and welding groove size parameters can be obtained first, and corresponding posture labels can be set for each set of training data.

[0212] Furthermore, each set of training data is input into the RNN network to be trained, and the RNN network to be trained is used to predict the corresponding pose prediction result for each set of training data in each round of training.

[0213] In each round of training, the time step can be set to 1.

[0214] Furthermore, the loss value of the loss function is calculated based on the pose prediction results and the corresponding pose labels.

[0215] The loss function may specifically be, for example, mean square error.

[0216] Furthermore, if the loss value is greater than a preset loss threshold, the parameters of the RNN network to be trained are adjusted, and the adjusted RNN network is used to perform the next round of training again; if the loss value is less than or equal to the loss threshold, the trained RNN network is obtained.

[0217] In another embodiment of the present application, during the filling welding process, based on the real-time values ​​of the welding gun posture parameters calculated in real time, when any one or more real-time values ​​are greater than the corresponding preset limit values, the real-time values ​​of the welding gun posture parameters are corrected according to the welding gun posture planning values.

[0218] Among them, the planned value of the welding gun posture can specifically include the planned lateral deviation of the welding gun from the center of the groove, the planned relative height of the welding gun from the end face of the groove, the planned first posture angle and the planned second posture angle of the welding gun relative to the end face of the groove; the real-time value of the welding gun posture parameter includes the real-time lateral deviation of the welding gun from the center of the groove, the real-time relative height of the welding gun from the end face of the groove, the real-time first posture angle and the real-time second posture angle of the welding gun relative to the end face of the groove.

[0219] In this embodiment, if Figure 6 As shown, when the welding robot completes the return stroke 0 and returns to point A, it can start from point A and perform fill welding along the welding groove, that is, Figure 6 Heat welding in.

[0220] Among them, during the hot welding process of the welding robot, the welding gun posture parameters can be calculated in the aforementioned steps. By shooting the real-time laser line image, the real-time laser line image feature pixel points can be extracted, and the laser line image feature pixel points can be converted into a three-dimensional point cloud by detecting the mathematical model, and then fitted to obtain a three-dimensional welding groove diagram. At the same time, the gravity sensor is used to obtain the real-time posture data of the welding gun.

[0221] Furthermore, the real-time three-dimensional welding groove diagram and the real-time welding gun posture data are used to calculate the real-time welding gun posture parameters, that is, the real-time values ​​of the welding gun posture parameters including the real-time lateral deviation of the welding gun from the center of the groove, the real-time relative height of the welding gun from the end face of the groove, the real-time first posture angle and the real-time second posture angle of the welding gun relative to the end face of the groove.

[0222] Based on this, the industrial computer can compare the gap between the planned values ​​of each welding gun posture and the real-time values ​​of the welding gun posture parameters in real time.

[0223] Specifically, compare whether the difference between the real-time lateral deviation and the planned lateral deviation is greater than the preset first limit, compare whether the difference between the real-time relative height and the planned relative height is greater than the preset second limit, compare whether the difference between the real-time first attitude angle and the planned first attitude angle is greater than the preset third limit, and compare whether the difference between the real-time second attitude angle and the planned second attitude angle is greater than the preset fourth limit.

[0224] The first limit value may be, for example, 0.4 mm, the second limit value may be, for example, 1 mm, the third limit value may be, for example, 1°, and the fourth limit value may be, for example, 1°.

[0225] Furthermore, if the difference between the real-time lateral deviation and the planned lateral deviation is greater than the first limit, the difference between the real-time relative height and the planned relative height is greater than the preset second limit, and / or the difference between the real-time first posture angle and the planned first posture angle is greater than the preset third limit, and / or the difference between the real-time second posture angle and the planned second posture angle is greater than the preset fourth limit, the real-time value of the welding gun posture parameter is corrected according to the planned value of the welding gun posture.

[0226] Based on this, in this embodiment, based on the calculation of the real-time values ​​of the welding gun posture parameters, the real-time values ​​of the welding gun posture parameters can be corrected by comparing whether the difference between the real-time lateral deviation and the planned lateral deviation is greater than the preset first limit, comparing whether the difference between the real-time relative height and the planned relative height is greater than the preset second limit, comparing whether the difference between the real-time first posture angle and the planned first posture angle is greater than the preset third limit, and comparing whether the difference between the real-time second posture angle and the planned second posture angle is greater than the preset fourth limit, thereby ensuring that the welding gun can complete welding in the optimal posture during the filling welding process.

[0227] In another embodiment of the present application, when using a welding gun to perform a cover weld on a weld groove that has completed a fill weld, the industrial computer can use the welding gun posture planning value planned for the last fill weld to control the welding gun to perform the cover weld.

[0228] In this embodiment, since the tracking deviation requirement for the welding gun during cap welding is relatively low, for example, the accuracy can be <1mm, and the laser line deformation information collected by the visual sensor during the cap welding process is almost invisible, it is no longer necessary to use a composite sensor to collect data during cap welding. Figure 5 As shown, the industrial computer can use the welding gun posture planning value planned for the last fill weld, that is, the welding gun posture planning value planned for the 6th layer of fill weld, and control the welding gun to perform two cover welds of the 7th layer according to the pipeline position of the welding gun and the welding robot.

[0229] Based on this, in this embodiment, based on the process requirements of the cap welding, the cap welding is performed by adopting the welding gun posture planning value of the last fill welding pass, thereby reducing the redundancy of data processing during the cap welding.

[0230] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, an embodiment of the present application also provides an online planning device for welding gun posture for multi-layer and multi-pass external welding of pipelines.

[0231] refer to Figure 8 The welding gun posture online planning device for multi-layer and multi-pass external welding of pipelines includes:

[0232] An acquisition module 801 is used to acquire a laser line image and welding gun posture data of the welding groove of the pipeline for each filling weld during the external welding of the pipeline;

[0233] An extraction module 802 is used to extract characteristic pixel points of the laser line image from the laser line image;

[0234] The calculation module 803 is used to convert the characteristic pixels of the laser line image into a three-dimensional point cloud by detecting a mathematical model and perform fitting to obtain a three-dimensional welding groove diagram. The three-dimensional welding groove diagram and welding gun posture data are used to calculate welding groove size parameters, welding gun posture parameters, and workpiece posture parameters to be welded;

[0235] Planning module 804, for determining a welding gun posture planning value for a current pass based on welding groove size parameters, welding gun posture parameters, and posture parameters of a workpiece to be welded;

[0236] Correction module 805, used to control the welding gun to perform the current pass of filling welding along the welding groove according to the welding gun posture planning value, detect the real-time value of the welding gun posture parameter in real time during the filling welding process, and correct the real-time value of the welding gun posture parameter according to the welding gun posture planning value;

[0237] The return module 806 is used to obtain the return laser line image and return welding gun posture data of the welding groove containing the weld bead during the return process after each filling welding pass, and use the return laser line image and return welding gun posture data to plan the welding gun posture planning value for the next pass;

[0238] The completion module 807 is used to control the welding gun to perform cover welding on the welding groove after completing multiple passes of filling welding, thereby completing the multi-layer and multi-pass external welding of the pipeline.

[0239] In one embodiment, the acquisition module 801 is specifically configured to:

[0240] The welding equipment is controlled to move along the track set on the pipeline, and multiple laser lines are projected toward the welding groove during the movement. The composite sensor in the welding equipment is used to obtain the laser line image of the welding groove and the welding gun posture data.

[0241] In another embodiment, the extraction module 802 is specifically configured to:

[0242] Dividing the laser line image into multiple sub-images according to different pixel grayscale distribution characteristics in the laser line image;

[0243] Each sub-image is input into a pre-trained neural network, and the pre-trained neural network is used to extract the characteristic pixel points of the laser line image in each sub-image;

[0244] The laser line image feature pixel points of multiple sub-images are fused to obtain the laser line image feature pixel points.

[0245] Among them, in another embodiment, the welding groove size parameters include groove width, groove depth, groove surface angle and misalignment; the posture parameters of the workpiece to be welded include the inclination angle and yaw angle of the groove end face in the world coordinates; the welding gun posture parameters include lateral deviation, relative height, first posture angle, and second posture angle.

[0246] The calculation module 803 is specifically configured to:

[0247] Calculate the groove width, groove depth, groove face angle and misalignment using the 3D welding groove diagram;

[0248] In the preset camera coordinate system, the groove width, groove depth, groove face angle, misalignment, groove center position and groove end position in the 3D welding groove image are used to calculate the lateral deviation of the welding gun from the groove center, the relative height of the welding gun from the groove end, and the first and second posture angles of the welding gun relative to the groove end;

[0249] The inclination angle and yaw angle of the groove end face in the world coordinate system are calculated using the three-dimensional welding groove diagram and welding gun posture data.

[0250] In another embodiment, the planned welding gun posture value includes the planned lateral deviation of the welding gun from the center of the groove, the planned relative height of the welding gun from the end face of the groove, the planned first posture angle and the planned second posture angle of the welding gun relative to the end face of the groove; the real-time value of the welding gun posture parameter includes the real-time lateral deviation of the welding gun from the center of the groove, the real-time relative height of the welding gun from the end face of the groove, the real-time first posture angle and the real-time second posture angle of the welding gun relative to the end face of the groove.

[0251] The correction module 805 is specifically used for:

[0252] When the difference between the real-time lateral deviation and the planned lateral deviation is greater than a preset first limit, the real-time lateral deviation is corrected according to the planned lateral deviation;

[0253] When the difference between the real-time relative altitude and the planned relative altitude is greater than a preset second limit, the real-time relative altitude is corrected according to the planned relative altitude;

[0254] When the difference between the real-time first attitude angle and the planned first attitude angle is greater than a preset third limit, correcting the real-time first attitude angle according to the planned first attitude angle;

[0255] When the difference between the real-time second posture angle and the planned second posture angle is greater than a preset fourth limit, the real-time second posture angle is corrected according to the planned second posture angle.

[0256] In another embodiment, the backhaul module 806 is specifically configured to:

[0257] Extracting characteristic pixel points of the return laser line image from the return laser line image;

[0258] The characteristic pixel points of the laser line image of the return journey are converted into a three-dimensional point cloud by detecting a mathematical model and fitting is performed to obtain the three-dimensional welding groove diagram of the return journey. The three-dimensional welding groove diagram of the return journey and the welding gun posture data of the return journey are used to calculate the welding groove size parameters, welding gun posture parameters and posture parameters of the workpiece to be welded of the return journey;

[0259] The welding gun posture planning value of the next pass is determined based on the welding groove size parameters, welding gun posture parameters and posture parameters of the workpiece to be welded in the return process to replace the welding gun posture planning value of the current pass.

[0260] In another embodiment, the completion module 807 is specifically configured to:

[0261] According to the welding gun posture planning value during the last filling weld, the welding groove is capped and welded to complete the multi-layer and multi-pass external welding of the pipeline.

[0262] For the convenience of description, the above devices are described as being divided into various modules according to their functions. Of course, when implementing the embodiments of the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0263] The device of the above embodiment is used to implement the corresponding online planning method of welding gun posture for multi-layer and multi-pass external welding of pipelines in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0264] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, it implements the online planning method of the welding gun posture for multi-layer and multi-pass external welding of pipelines as in any of the above embodiments.

[0265] FIG9 shows a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application.

[0266] The electronic device may include a processor 901 and a memory 902 storing computer program instructions.

[0267] Specifically, the processor 901 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0268] The memory 902 may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory 902 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 902 may include removable or non-removable (or fixed) media. Where appropriate, the memory 902 may be internal or external to the electronic device. In certain embodiments, the memory 902 is a non-volatile solid-state memory.

[0269] The memory 902 may include read-only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physical / tangible memory storage devices. Thus, generally, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.

[0270] The processor 901 reads and executes computer program instructions stored in the memory 902 to implement any one of the above-mentioned methods for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines.

[0271] In one example, the electronic device may further include a communication interface 903 and a bus 910. Figure 9 As shown, the processor 901 , the memory 902 , and the communication interface 903 are connected via a bus 910 and communicate with each other.

[0272] The communication interface 903 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0273] The bus 910 includes hardware, software, or both, and couples the components of the electronic device to each other. By way of example, and not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, the bus 910 may include one or more buses. Although the embodiments of the present application describe and illustrate specific buses, the present application contemplates any suitable bus or interconnect.

[0274] The electronic device can execute the welding gun posture online planning method for multi-layer and multi-pass external welding of pipelines in the embodiment of the present application based on the pre-planned welding posture planning value, thereby realizing the combination of Figure 3 and Figure 6 The paper describes an online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines.

[0275] In addition, in conjunction with the above-described embodiments of the online planning method for welding gun posture for multi-layer, multi-pass external pipe welding, embodiments of the present application may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any of the above-described embodiments of the online planning method for welding gun posture for multi-layer, multi-pass external pipe welding is implemented.

[0276] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements any one of the above-mentioned embodiments of the welding gun posture online planning method for multi-layer and multi-pass external welding of pipelines.

[0277] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0278] The functional blocks shown in the block diagrams described above can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they may be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, and the like. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments may be stored in a machine-readable medium or transmitted via a data signal carried in a carrier wave over a transmission medium or communication link. "Machine-readable medium" may include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROMs, flash memory, erasable ROMs (EROMs), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, and the like. Code segments may be downloaded via a computer network such as the Internet or an intranet.

[0279] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0280] Aspects of the present disclosure have been described above with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each block in the flowcharts and / or block diagrams, as well as combinations of blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine such that execution of these instructions by the processor of the computer or other programmable data processing device enables the implementation of the functions / actions specified in one or more blocks in the flowcharts and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It should also be understood that each block in the block diagrams and / or flowcharts, as well as combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by dedicated hardware that performs the specified functions or actions, or by a combination of dedicated hardware and computer instructions.

[0281] The above description is only a specific embodiment of the present application. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the scope of protection of the present application.

Claims

1. An online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines, characterized in that: include: For each filler weld during external welding of the pipeline, a laser line image and welding gun posture data of the welding groove of the pipeline are obtained; extracting laser line image feature pixel points from the laser line image; The characteristic pixel points of the laser line image are converted into a three-dimensional point cloud by detecting a mathematical model and fitted to obtain a three-dimensional welding groove diagram, and the three-dimensional welding groove diagram and the welding gun posture data are used to calculate welding groove size parameters, welding gun posture parameters and workpiece posture parameters to be welded, wherein the welding groove size parameters include groove width, groove depth, groove face angle and misalignment; the workpiece posture parameters to be welded include the inclination angle and yaw angle of the groove end face in world coordinates; the welding gun posture parameters include lateral deviation, relative height, first posture angle and second posture angle; Determining a welding gun posture planning value for a current pass based on the welding groove size parameters, welding gun posture parameters, and posture parameters of a workpiece to be welded, wherein the welding gun posture planning value includes the posture of the welding gun at various positions in the welding groove when performing the current pass of filling welding along the welding groove, and represents changes in the welding gun posture when the welding gun moves along the welding groove; Controlling the welding gun to perform the current pass of filling welding along the welding groove according to the welding gun posture planning value, detecting the real-time value of the welding gun posture parameter in real time during the filling welding process, and correcting the real-time value of the welding gun posture parameter according to the welding gun posture planning value; In the return process after each fill welding pass, a return laser line image and return welding gun posture data of the welding groove containing the weld bead are obtained, and the welding gun posture planning value of the next pass is planned using the return laser line image and the return welding gun posture data; After completing multiple passes of filling welding, controlling the welding gun to perform cap welding on the welding groove to complete the multi-layer and multi-pass external welding of the pipeline; The welding gun posture planning value includes a planned lateral deviation of the welding gun from the groove center, a planned relative height of the welding gun from the groove end face, a planned first posture angle of the welding gun relative to the groove end face, and a planned second posture angle; the welding gun posture parameter real-time value includes a real-time lateral deviation of the welding gun from the groove center, a real-time relative height of the welding gun from the groove end face, a real-time first posture angle of the welding gun relative to the groove end face, and a real-time second posture angle; and the correcting of the welding gun posture parameter real-time value according to the welding gun posture planning value includes: When the difference between the real-time lateral deviation and the planned lateral deviation is greater than a preset first limit, correcting the real-time lateral deviation according to the planned lateral deviation; When the difference between the real-time relative altitude and the planned relative altitude is greater than a preset second limit, correcting the real-time relative altitude according to the planned relative altitude; When the difference between the real-time first attitude angle and the planned first attitude angle is greater than a preset third limit, correcting the real-time first attitude angle according to the planned first attitude angle; When the difference between the real-time second attitude angle and the planned second attitude angle is greater than a preset fourth limit, the real-time second attitude angle is corrected according to the planned second attitude angle.

2. The method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines according to claim 1 is characterized in that: The step of obtaining the laser line image and welding gun posture data of the welding groove of the pipeline includes: The welding equipment is controlled to move along a track set on the pipeline, and multiple laser lines are projected toward the welding groove during the movement. The composite sensor in the welding equipment is used to obtain the laser line image of the welding groove and the welding gun posture data.

3. The method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines according to claim 1 is characterized in that: The method of calculating welding groove size parameters, welding gun posture parameters, and workpiece posture parameters to be welded by using the three-dimensional welding groove diagram and the welding gun posture data includes: Calculating the groove width, the groove depth, the groove surface angle, and the misalignment amount using the three-dimensional welding groove diagram; In a preset camera coordinate system, using the groove width, the groove depth, the groove face angle, the misalignment, the position of the groove center, and the position of the groove end face in the three-dimensional welding groove image, calculate the lateral deviation of the welding gun from the groove center, the relative height of the welding gun from the groove end face, and the first and second posture angles of the welding gun relative to the groove end face; The three-dimensional welding groove diagram and the welding gun posture data are used to calculate the inclination angle and the yaw angle of the groove end face in the world coordinate system.

4. The method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines according to claim 1 is characterized in that: In the return process after each fill welding pass, a return laser line image of the welding groove containing the weld bead and return welding gun posture data are obtained, and a welding gun posture planning value for the next pass is planned using the return laser line image and the return welding gun posture data, including: Extracting characteristic pixel points of the return laser line image from the return laser line image; The characteristic pixel points of the laser line image of the return journey are converted into a three-dimensional point cloud by the detection mathematical model and fitted to obtain a three-dimensional welding groove diagram of the return journey, and the welding groove size parameters, welding gun posture parameters and workpiece posture parameters of the return journey are calculated using the three-dimensional welding groove diagram of the return journey and the welding gun posture data of the return journey; The welding gun posture planning value of the next pass is determined based on the welding groove size parameters, welding gun posture parameters and posture parameters of the workpiece to be welded in the return process to replace the welding gun posture planning value of the current pass.

5. The method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines according to claim 1 is characterized in that: The controlling of the welding gun to perform cap welding on the welding groove to complete multi-layer and multi-pass external welding of the pipeline includes: According to the welding gun posture planning value during the last filling welding, the welding groove is capped and welded to complete the multi-layer and multi-pass external welding of the pipeline.

6. The method for online planning of welding gun posture for multi-layer and multi-pass external welding of pipelines according to claim 1 is characterized in that: The extracting characteristic pixel points of the laser line image from the laser line image includes: dividing the laser line image into a plurality of sub-images according to different pixel grayscale distribution characteristics in the laser line image; Each sub-image is input into a pre-trained neural network, and the pre-trained neural network is used to extract the characteristic pixel points of the laser line image in each sub-image; The laser line image feature pixel points of multiple sub-images are fused to obtain the laser line image feature pixel points.

7. An online planning device for welding gun posture for multi-layer and multi-pass external welding of pipelines, characterized in that: The device comprises: An acquisition module, for acquiring a laser line image and welding gun posture data of the welding groove of the pipeline for each filling weld during the external welding of the pipeline; An extraction module, configured to extract characteristic pixel points of the laser line image from the laser line image; a calculation module, configured to convert the characteristic pixel points of the laser line image into a three-dimensional point cloud by detecting a mathematical model and perform fitting to obtain a three-dimensional welding groove diagram, and calculate welding groove dimension parameters, welding gun posture parameters, and posture parameters of a workpiece to be welded using the three-dimensional welding groove diagram and the welding gun posture data, wherein the welding groove dimension parameters include groove width, groove depth, groove face angle, and misalignment; the posture parameters of the workpiece to be welded include the inclination angle and yaw angle of the groove end face in world coordinates; and the welding gun posture parameters include lateral deviation, relative height, first posture angle, and second posture angle; a planning module, which determines a welding gun posture planning value for a current pass based on the welding groove size parameters, welding gun posture parameters, and posture parameters of the workpiece to be welded, wherein the welding gun posture planning value includes the posture of the welding gun at various positions in the welding groove when performing the current fill welding pass along the welding groove, and the welding gun posture planning value represents a change in the welding gun posture when the welding gun moves along the welding groove; a correction module, configured to control the welding gun to perform the current pass of filling welding along the welding groove according to the welding gun posture planning value, detect the real-time value of the welding gun posture parameter in real time during the filling welding process, and correct the real-time value of the welding gun posture parameter according to the welding gun posture planning value; A return module is used to obtain a return laser line image and return welding gun posture data of the welding groove containing the weld bead during the return process after each filling welding pass, and plan the welding gun posture planning value of the next pass using the return laser line image and the return welding gun posture data; A completion module, configured to control the welding gun to perform cap welding on the welding groove after completing multiple passes of filling welding, thereby completing multi-layer and multi-pass external welding of the pipeline; Wherein, the welding gun posture planning value includes the planned lateral deviation of the welding gun from the center of the groove, the planned relative height of the welding gun from the end face of the groove, the planned first posture angle and the planned second posture angle of the welding gun relative to the end face of the groove; the real-time value of the welding gun posture parameter includes the real-time lateral deviation of the welding gun from the center of the groove, the real-time relative height of the welding gun from the end face of the groove, the real-time first posture angle and the real-time second posture angle of the welding gun relative to the end face of the groove; the real-time value of the welding gun posture parameter is corrected according to the welding gun posture planning value, including, When the difference between the real-time lateral deviation and the planned lateral deviation is greater than a preset first limit, correcting the real-time lateral deviation according to the planned lateral deviation; When the difference between the real-time relative altitude and the planned relative altitude is greater than a preset second limit, correcting the real-time relative altitude according to the planned relative altitude; When the difference between the real-time first attitude angle and the planned first attitude angle is greater than a preset third limit, correcting the real-time first attitude angle according to the planned first attitude angle; When the difference between the real-time second attitude angle and the planned second attitude angle is greater than a preset fourth limit, the real-time second attitude angle is corrected according to the planned second attitude angle.

8. A welding device, characterized in that: The method comprises a composite sensor, a welding robot and an industrial computer, wherein the composite sensor is used to collect laser line images and welding gun posture data, the welding robot is used to adjust the welding gun posture, and the industrial computer is used to execute the welding gun posture online planning method for multi-layer and multi-pass external welding of pipelines as described in any one of claims 1 to 6.

9. An electronic device, characterized in that: The device includes: a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, it implements the online planning method for welding gun posture for multi-layer and multi-pass external welding of pipelines as described in any one of claims 1 to 6.

Images