Single-side welding and double-side forming process for K-shaped welding seam of block node

Abstract

The invention discloses a single-side welding and double-side forming process for a K-shaped welding seam of a block node, and belongs to the technical field of welding processes, and the single-side welding and double-side forming process comprises a welding process based on a model and a welding process based on reversion. According to the method, firstly, a welding test of the block nodes is conducted, a test result is applied to robot automatic welding work of the block nodes, 3D scanning is conducted on the surface of a workpiece through a large-view-field laser sensor for a plane workpiece welding scene without a CAD model through the intelligent welding seam reverse extraction technology, and the intelligent welding seam reverse extraction technology is adopted; the position information of the welding seam can be automatically extracted by means of a laser scanning device, a workpiece is scanned along with the movement of the x axis and the y axis of a gantry on a tool installed at the tail end of a robot flange plate through large-view-field laser, noise reduction and optimization processing are conducted on a point cloud model, ambient light interference is removed, and splicing errors are reduced; and meanwhile, based on an AI algorithm, the point cloud data is processed and analyzed, the contour is extracted, and the position data of the welding seam is calculated.

Description

Technical Field

[0001] The invention belongs to the technical field of welding technology, and in particular relates to a single-sided welding and double-sided forming process for a K-type weld of a block node. Background Art

[0002] Building a strong maritime nation is the only way to enhance comprehensive national strength. As the foundation of the deep-sea strategy, marine engineering equipment, especially the upper modules of marine oil and gas production platforms, is a complex of high-tech technologies such as advanced manufacturing technology, information technology and new materials technology. It has the characteristics of high technology, high investment, high output, high added value and high risk. It has a strong industrial radiation capacity and a great driving effect on the national economy.

[0003] With the rapid development of offshore oil industry, the connection technology of offshore oil engineering block nodes has become one of the key technologies in this field. Among them, K-type welds are the main connection method of block nodes. Their quality and efficiency directly affect the safety and economy of the entire offshore oil project. The traditional K-type weld welding process usually adopts double-sided welding. Although this method can ensure the quality of the weld, it is complicated to operate, inefficient, and requires a very high technical level of the welder. Therefore, the development of a single-sided welding and double-sided forming process that can not only ensure the quality of the weld, but also improve the welding efficiency and lower the technical threshold has become an urgent problem to be solved in the offshore oil engineering field. In the past few decades, domestic and foreign researchers have conducted a lot of research and exploration on the single-sided welding and double-sided forming process. For example, some researchers have Researchers have tried to improve the forming effect of welds by changing welding parameters, optimizing welding materials and other methods. However, these methods can only improve the quality of welds to a certain extent, and it is difficult to fully meet the high requirements of offshore oil projects for weld quality. In addition, some researchers have proposed the use of special welding equipment and processes to achieve single-sided welding and double-sided forming, but these methods are often costly and difficult to be widely used in actual projects. Therefore, the research on the single-sided welding and double-sided forming process of K-type welds at the block nodes of offshore oil projects still has important practical significance and broad application prospects. This technology aims to develop an efficient, economical and practical single-sided welding and double-sided forming process through in-depth research and practice, so as to provide strong technical support for the safe, efficient and economic construction of offshore oil projects.

[0004] Therefore, it is urgent to design a single-sided welding and double-sided forming process for the K-type weld of the block node to solve the above-mentioned problems. Summary of the invention

[0005] The purpose of the present invention is to provide a single-sided welding and double-sided forming process for a block node K-type weld, which has the advantages of high efficiency, economy and practicality, provides strong technical support for the safe, efficient and economic construction of offshore oil projects, and solves the problems mentioned in the background technology.

[0006] To achieve the above-mentioned purpose, the specific technical scheme of a single-sided welding and double-sided forming process of a K-type weld of a block node of the present invention is as follows: A single-sided welding and double-sided forming process for a K-type weld of a block node includes a model-based welding process and a reverse-based welding process. The model-based welding process includes the following steps: S10, model import; S11, generating a robot path program; S12, the workpiece is clamped in place; S13, move the gantry to above the workpiece; S14, visual positioning; S15, weld scanning measurement; S16, layer planning and process matching; S17, generating a welding program; S18, perform welding seam tracking and parameter monitoring; S19, welding is completed; The reverse-based welding process includes the following steps: S20, the workpiece is placed in place; S21, move the gantry to above the workpiece; S22, laser scanning reverse modeling; S23, automatically extracting welds and manually confirming welds; S24, generating a robot path program; S25, weld scanning measurement; S26, layer planning and process matching; S27, generating a welding program; S28, performing welding seam tracking and parameter monitoring; S29, welding is completed.

[0007] Furthermore, both the model-based welding process and the reverse-based welding process include a multi-layer and multi-pass process, which is between performing welding seam tracking, parameter monitoring and generating welding programs.

[0008] Furthermore, the block nodes in the model-based welding process and the reverse-based welding process are composed of three groups of I-beams.

[0009] Further, the block node welding method in the model-based welding process and the reverse-based welding process includes the following steps: S30, butt weld is full penetration weld; S31. The groove processing method is flame or plasma cutting, the groove angle error is between 50°±5°, and varies randomly with the weld, the root gap range is between 3-5mm, and varies randomly with the weld, and single-sided welding and double-sided forming and no backing are required, the inner / outer skin misalignment range is between 0-2mm, and varies randomly with the weld, the blunt edge thickness range is between 0-2mm, and varies randomly with the weld, and the weld position deviation is ≥3mm; S32, select K-type groove to form symmetrical welding, so that the joint forms a symmetrical stress field, the welding position is horizontal butt, the preheating requirements must reach the minimum preheating temperature and interlayer temperature specified in WPS, arrange the welding sequence to control deformation, the welding method is robot fully automatic welding, and the welding is carried out by melting extremely rich argon gas shielded arc welding; S33. Rust, moisture, oil and other debris should be removed from the groove and within 20mm on both sides before welding; S34, welding is performed by workers, the positioning welding length is 40-50mm, the spacing is 350-400mm, and the height of the butt weld layer is 4-5mm; S35, T-type butt joint, arc-starting plate and arc-extinguishing plate are welded at both ends of the weld. After welding, the arc-starting plate and arc-extinguishing plate are removed by gas cutting, and then the cut is polished smooth; S36, the welding gun points to the midpoint of the gap, and the welding gun is tilted forward 70°, with an angle of 30° with the bottom plate to complete the base welding; S37. The filling cap welding gun swings in a triangular shape, with the counterpoint pointing to the midpoint of the base weld, and the welding gun tilts forward 70° to complete the second weld. The welding is along the intersection of the second weld and the base plate, and the welding gun tilts forward 70°.

[0010] Furthermore, it also includes a block node robot intelligent welding system, which includes intelligent weld reverse extraction technology, welding real-time tracking technology, intelligent weld measurement technology, intelligent welding process development and intelligent welding decision-making technology.

[0011] Furthermore, for flat workpiece welding scenarios without CAD models, the intelligent weld reverse extraction technology uses a large-field-of-view laser sensor to perform 3D scanning on the workpiece surface, and uses intelligent weld reverse extraction technology to automatically extract the weld position information; the point cloud model is denoised and optimized to remove ambient light interference and reduce stitching errors; at the same time, based on the Al algorithm, the point cloud data is processed and analyzed to extract the contour and calculate the weld position data.

[0012] Furthermore, the real-time welding seam tracking technology uses visual real-time tracking, optimizes the optical path of the laser sensor, and filters external light interference in the process. At the same time, based on the Al algorithm, deep learning of the image is performed to significantly filter out external light interference and completely extract the contour curve of the groove under multi-layer and multi-pass conditions.

[0013] Furthermore, the intelligent weld contour reverse modeling technology accurately scans the weld contour, reconstructs the 3D data of the weld contour, measures the size of the groove, and performs weld layer planning and process matching. The scanning speed is more than 20 times the welding speed, and a 3D contour calculation algorithm is used. At the same time, the equal area method and equal height method are used to plan the layer of the groove measurement results.

[0014] Furthermore, the development route of intelligent welding process development in the development process is: material-based process database + industry scenario-based matching model, that is, the process development is divided into a basic process database and an inference process database, experiments are carried out based on fixed-precision materials, and a basic process database is formed through unified processing. By measuring the weld shape, the corresponding welding parameters are obtained; on the basis of the process database, an intelligent reasoning model is combined.

[0015] Furthermore, intelligent welding decision-making technology uses artificial intelligence technology to conduct big data model training and actual welding tests on welding processes, measurement results, image information, process models and other information to determine the optimal matching relationship. Finally, based on the inference process database, it combines the scene understanding of each industry, process quality requirements, and working conditions to form a matching model, and generates a matching model with real-time correction capabilities, thereby realizing intelligent decision-making in the welding process.

[0016] Furthermore, the model-based welding process, reverse-based welding process and block node robot intelligent welding system all transmit data at high speed through 5G communication, and use digital twin technology to dynamically match the production process screen with the actual production process screen in real time; and based on large model data, perform online fault diagnosis and preventive maintenance functions.

[0017] The present invention has the following advantages: (1) In the present invention, a welding test of a block node is first carried out, and the test results will be applied to the robot automatic welding work of the block node. For the planar workpiece welding scene without a CAD model, a large-field-of-view laser sensor is used to perform 3D scanning on the workpiece surface. The intelligent weld reverse extraction technology can be used to automatically extract the position information of the weld. The large-field-of-view laser is installed on the tooling at the end of the robot flange. The workpiece is scanned as the gantry moves along the x-axis and y-axis. The point cloud model is denoised and optimized to remove ambient light interference and reduce splicing errors. At the same time, based on the AI ​​algorithm, the point cloud data is processed and analyzed to extract the contour and calculate the position data of the weld.

[0018] (2) In the present invention, the real-time weld tracking technology is different from the traditional arc tracking in that it is restrictive to the scene. Visual real-time tracking can effectively improve the scope of application and accuracy of weld tracking. During the process, the optical path of the laser sensor is optimized, external light interference is filtered, and image quality is improved. Based on the AI ​​algorithm, deep learning of the image can be performed, which can significantly filter out external light interference and completely extract the contour curve of the groove under multi-layer and multi-channel conditions.

[0019] (3) In the present invention, the weld contour is accurately scanned by intelligent weld contour reverse modeling technology, and the 3D data of the weld contour is reconstructed, so as to measure the size of the groove, perform weld layer planning and process matching, and the scanning speed can reach more than 20 times the welding speed. In addition, the 3D contour calculation algorithm is used in the process. Compared with the traditional contour line algorithm, it can eliminate the error caused by the non-perpendicularity between the laser and the groove and improve the measurement accuracy. The equal area method and the equal height method are used to plan the layer of the groove measurement results, realizing an efficient, economical and practical single-sided welding and double-sided forming process, providing strong technical support for the safe, efficient and economic construction of offshore oil projects. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a schematic diagram of the process flow of the single-sided welding and double-sided forming process of the present invention; Figure 2 It is a schematic diagram of the block node robot intelligent welding system of the present invention; Figure 3 It is a front view structural schematic diagram of a block node in the single-sided welding and double-sided forming process of the present invention; Figure 4 It is a schematic diagram of the side view structure of the assembly node in the single-sided welding and double-sided forming process of the present invention; Figure 5 It is a schematic diagram of the top view structure of the assembly node in the single-sided welding and double-sided forming process of the present invention; Figure 6 It is a schematic diagram of the assembly requirements of the block nodes in the single-sided welding and double-sided forming process of the present invention; Figure 7 It is a schematic diagram of the bottom welding posture and welding position of the assembly node of the present invention; Figure 8 A schematic diagram of the second bottom welding posture and welding position of the assembly node of the present invention; Fig. 9 It is a schematic diagram of the third bottom welding posture and welding position of the assembly node of the present invention; Fig.10 It is a schematic diagram of the fourth bottom welding posture and welding position of the assembly node in the present invention; Fig.11 It is a schematic diagram of the fifth bottom welding posture and welding position of the assembly node of the present invention; Fig.12 It is a schematic diagram of the sixth bottom welding posture and welding position of the assembly node of the present invention; Fig.13 It is a schematic diagram of the seventh bottom welding posture and welding position of the assembly node of the present invention; Fig.14 It is a schematic diagram of the eighth bottom welding posture and welding position of the block node of the present invention; Fig.15 It is a schematic diagram of the ninth bottom welding posture and welding position of the block node of the present invention; Fig.16 It is a schematic diagram of the tenth bottom welding posture and welding position of the assembly node of the present invention; Fig.17 It is a schematic diagram of the eleventh bottom welding posture and welding position of the assembly node of the present invention; Fig.18 It is a schematic diagram of the twelfth bottom welding posture and welding position of the assembly node of the present invention; Fig.19 This is a schematic diagram of the thirteenth bottom welding posture and welding position of the assembly node of the present invention; Fig. 20 It is a schematic diagram of the weld distribution structure of the block node of the present invention; DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] Those skilled in the art will appreciate that, although some embodiments herein include certain features included in other embodiments but not other features, the combination of features of different embodiments is meant to be within the scope of the present invention and form different embodiments. For example, in the claims, any one of the claimed embodiments may be used in any combination.

[0023] Please refer to the attached Figure 1 To Attachment Fig. 20 The invention describes a single-sided welding and double-sided forming process for a K-type weld of a block node.

[0024] At present, researchers at home and abroad have conducted a lot of research and exploration on the single-sided welding and double-sided forming process. Some researchers have tried to improve the forming effect of the weld by changing welding parameters, optimizing welding materials and other methods. However, these methods can only improve the quality of the weld to a certain extent, and it is difficult to fully meet the high requirements of offshore oil engineering for weld quality. In addition, some researchers have proposed the use of special welding equipment and processes to achieve single-sided welding and double-sided forming, but these methods are often costly and difficult to be widely used in actual engineering.

[0025] Therefore, the single-sided welding and double-sided forming process of the K-type weld of the block node includes a model-based welding process and a reverse-based welding process. The model-based welding process includes the following steps: S10, model import; S11, generating a robot path program; S12, the workpiece is clamped in place; S13, move the gantry to above the workpiece; S14, visual positioning; S15, weld scanning measurement; S16, layer planning and process matching; S17, generating a welding program; S18, perform welding seam tracking and parameter monitoring; S19, welding is completed; The reverse-based welding process includes the following steps: S20, the workpiece is placed in place; S21, move the gantry to above the workpiece; S22, laser scanning reverse modeling; S23, automatically extracting welds and manually confirming welds; S24, generating a robot path program; S25, weld scanning measurement; S26, layer planning and process matching; S27, generating a welding program; S28, performing welding seam tracking and parameter monitoring; S29, welding is completed.

[0026] Furthermore, both the model-based welding process and the reverse-based welding process include a multi-layer and multi-pass process, which is between performing welding seam tracking, parameter monitoring and generating welding programs.

[0027] Furthermore, the block nodes in the model-based welding process and the reverse-based welding process are composed of three groups of I-beams.

[0028] Further, the block node welding method in the model-based welding process and the reverse-based welding process includes the following steps: S30, butt weld is full penetration weld; Specifically, the impact energy of the welded joint is ≥34J (-20℃ / -40℃), and the materials are carbon steel DH36 / EH36 (PG1200 / PG900), SM490YB (H700 / H588), and EH36-Z35 / Z36 (reinforced flange plate). At the same time, the node weight is not less than 3 tons, and the overall dimensions are not less than 5000×2000×1500mm; S31. The groove processing method is flame or plasma cutting, manual grinding and the surface is free of oil, rust and coating. The groove angle error is between 50°±5° and varies randomly with the weld. The root gap range is between 3-5mm and varies randomly with the weld. Single-sided welding and double-sided forming are performed and no gasket is required. The inner / outer skin misalignment range is between 0-2mm and varies randomly with the weld. The blunt edge thickness range is between 0-2mm and varies randomly with the weld. The weld position deviation is ≥3mm. S32, select K-type groove, form symmetrical welding, so that the joint forms a symmetrical stress field, the welding position is horizontal butt (full penetration, single-sided welding and double-sided forming, no root cleaning on the back), the preheating requirements must reach the minimum preheating temperature and interlayer temperature specified in WPS, arrange the welding sequence to control deformation, the welding method is robot fully automatic welding, and the welding is carried out by melting extremely rich argon gas shielded arc welding; S33. Rust, moisture, oil and other debris should be removed from the groove and the area within 20mm on both sides before welding. When welding in a damp, windy and cold open-air site, the welding area should be properly shielded and protective measures should be taken; S34, welding is performed by workers, the positioning welding length is 40-50mm, the spacing is 350-400mm, and the height of the butt weld layer is 4-5mm; S35, T-type butt joint, arc-starting plate and arc-extinguishing plate are welded at both ends of the weld. After welding, the arc-starting plate and arc-extinguishing plate are removed by gas cutting, and then the cut is polished smooth. Specifically, the material and groove of the arc-starting plate and arc-extinguishing plate are the same as those of the workpiece being welded; S36, the welding gun points to the midpoint of the gap, and the welding gun is tilted forward 70°, with an angle of 30° with the bottom plate to complete the base welding; Specifically, the process parameters of base welding are as follows: S37. The filling cap welding gun swings in a triangular shape, with the counterpoint pointing to the midpoint of the base weld, and the welding gun tilts forward 70° to complete the second weld. The welding is along the intersection of the second weld and the base plate, and the welding gun tilts forward 70°.

[0029] Specifically, the process parameters for filling and capping are as follows: Furthermore, it also includes a block node robot intelligent welding system, which includes intelligent weld reverse extraction technology, welding real-time tracking technology, intelligent weld measurement technology, intelligent welding process development and intelligent welding decision-making technology. Specifically, the intelligent weld extraction technology includes drawing file analysis, intelligent weld contour reverse modeling technology, weld extraction, trajectory planning and automatic positioning; the welding real-time tracking technology includes robot control, anti-arc algorithm, trajectory correction algorithm and multi-layer and multi-channel tracking; the intelligent weld measurement technology includes laser imaging, weld tracking, dimension measurement, layer planning and process control; the intelligent welding process development includes basic process database development, self-learning expert database, process matching algorithm, database management and software; the intelligent welding decision-making technology includes real-time measurement of swing state, reasoning expert database, trajectory planning prediction technology and multi-layer and multi-channel measurement technology.

[0030] Furthermore, for the welding scenarios of flat workpieces without CAD models, the intelligent weld reverse extraction technology uses a large-field-of-view laser sensor to perform 3D scanning on the workpiece surface, and uses the intelligent weld reverse extraction technology to automatically extract the weld position information; the point cloud model is denoised and optimized to remove ambient light interference and reduce stitching errors; at the same time, based on the Al algorithm, the point cloud data is processed and analyzed to extract the contour and calculate the weld position data. Specifically, the large-field-of-view laser is installed on the tooling at the end of the robot flange, and the workpiece is scanned as the gantry moves on the x-axis and y-axis.

[0031] Furthermore, unlike traditional arc tracking, which has restrictions on scenes, real-time weld tracking technology uses visual real-time tracking and optimizes the optical path of the laser sensor and filters external light interference during the process. At the same time, based on the Al algorithm, deep learning of the image is performed to significantly filter out external light interference and completely extract the contour curve of the groove under multi-layer and multi-channel conditions.

[0032] Furthermore, the intelligent weld contour reverse modeling technology accurately scans the weld contour, reconstructs the 3D data of the weld contour, measures the size of the groove, and performs weld layer planning and process matching. The scanning speed is more than 20 times the welding speed, and a 3D contour calculation algorithm is used. At the same time, the equal area method and equal height method are used to plan the layer of the groove measurement results.

[0033] Furthermore, the development of intelligent welding process is different from the manual welding process library. Robot welding requires precise welding process parameters. Therefore, the development route in the development process is: material-based process database + industry scenario-based matching model, that is, process development is divided into basic process database and reasoning process database. Experiments are conducted based on fixed-precision materials, and a basic process database is formed through unified processing. The corresponding welding parameters are obtained by measuring the shape of the weld bead to solve the problem of high threshold for setting welding process parameters. For example, when the workpiece precision is fixed, the energy, posture, speed and other process parameters used in carbon steel, 20mm thickness, and 45° groove are obtained through experiments to form a basic process database; on the basis of the process database, combined with the intelligent reasoning model (decoupling, grading, and weight adjustment of various welding parameters), the basic process database is obtained through experimental accumulation, and the experiment cannot exhaustively enumerate the pain point of standard changes of all workpieces, so that the process library has the ability of self-learning. For example, the basic process database already includes the process parameters used in carbon steel, 20mm thickness, and 45° groove. Through the reasoning model, when the workpiece precision is fixed, the process database under thickness changes such as 22mm, 24mm... can be generated autonomously.

[0034] Furthermore, intelligent welding decision-making technology uses artificial intelligence technology to conduct big data model training and actual welding tests on welding processes, measurement results, image information, process models and other information to determine the optimal matching relationship. Finally, based on the inference process database, it combines the scene understanding of each industry, process quality requirements, and working conditions to form a matching model, and generates a matching model with real-time correction capabilities, thereby realizing intelligent decision-making in the welding process.

[0035] Furthermore, the model-based welding process, the reverse welding process and the block node robot intelligent welding system all transmit data at high speed through 5G communication, and use three security monitoring systems at the same time. When personnel break into the working area, the alarm is linked, which can effectively ensure the safety of on-site operators; the robot's motion posture can be monitored in real time, which is convenient for operators to grasp the welding progress; and digital twin technology is used to dynamically match the production process screen with the actual production process screen in real time; and based on large model data, online fault diagnosis and preventive maintenance functions are performed, and molten pool monitoring technology is used as a technical means of visual monitoring, so that workers can monitor the molten pool away from the welding site, perform data processing and analysis based on images, and adjust welding parameters. The molten pool monitoring system is equipped with various necessary sensors to monitor the molten pool in real time during the welding process. The monitoring data is calculated optically and acoustically to obtain the size, shape, temperature, composition, fluidity and other parameters of the molten pool in real time, and the welding parameters are adjusted synchronously in real time according to the real-time parameters, control algorithms, and databases to ensure the stability and consistency of welding quality.

[0036] In the production process, for multi-workpiece and multi-weld task scenarios, multi-robot and multi-workstation intelligent task allocation technology is adopted to automatically plan robot welding tasks and collaborative plans. The idle waiting rate of each robot does not exceed 10%, and the welding efficiency is optimized. It is equipped with a temperature sensor to monitor the preheating temperature and interlayer temperature before welding, and display them in real time on the console. It compares the preset preheating / interlayer temperature and issues an alarm prompt. At the same time, it is equipped with preheating-related hardware equipment, which can automatically complete the preheating before welding in conjunction with the temperature detection function, and pass harmful smoke into the purification system through the suction duct. The purification system blocks large particles of sparks and flammable substances as soon as possible. The hose of the welding smoke purifier can be bound to the welding gun and cutting gun to purify the welding smoke at the source, which is flexible and convenient.

[0037] The block node robot intelligent welding system includes an interaction layer, a control layer, and an execution layer. Each layer completes the deployment of different areas according to different functions to achieve decentralized control and comprehensive coordination. The robot system consists of: one robot body, six Kuka robot external axis motors, one welding machine, and two robot control cabinets. One controls the robot system and the x-axis and axis external axis, and the other controls the y-axis external axis and the positioner external axis. The two control cabinets communicate through EtherCAT, which enables the robot and the positioner to achieve linkage control. The gantry is powered by a system drag chain. The drag chain system is installed on the steel beam of the factory, which can effectively ensure the power supply and gas supply. The gantry and positioner are all piped and wired by drag chains. The robot cable and other related cables are all flexible cables to improve the service life. The gantry and positioner are connected by aviation plugs. At the same time, the robot system is equipped with two sets of programs, one with a positioner and the other without a positioner. Customers can choose to activate different programs according to their own needs to achieve different welding application scenarios. The ground wire of the welding machine is connected to the positioner by a copper rod sliding contact connection, which has a simple structure, convenient maintenance, and strong conductivity, which can effectively ensure the realization of the welding process. Four photoelectric sensors are installed on the gantry, which can more accurately determine the position of the gantry and effectively prevent the gantry from going astray and colliding. The laser sensor has a weld tracking function in weld tracking, which can control the welding gun centering and arc length parameters in real time, and the weld tracking accuracy is better than ±0.2mm; and weld measurement is performed: it can realize "adaptive welding", that is, according to the real-time changing actual weld space position and actual weld contour information, the "welding process parameters" are adjusted in real time to adapt to the groove gap, misalignment, The uneven and discontinuous changes in the angle of the groove surface realize automatic adaptive welding. The whole process does not require manual programming and manual operation, thereby realizing an efficient, economical and practical single-sided welding and double-sided forming process, providing strong technical support for the safe, efficient and economic construction of offshore oil projects.

[0038] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A single-sided welding and double-sided forming process for a K-type weld of a block node, characterized in that: Including model-based welding process and reverse-based welding process, the model-based welding process includes the following steps: S10, model import; S11, generating a robot path program; S12, the workpiece is clamped in place; S13, move the gantry to above the workpiece; S14, visual positioning; S15, weld scanning measurement; S16, layer planning and process matching; S17, generating a welding program; S18, perform welding seam tracking and parameter monitoring; S19, welding is completed; The reverse-based welding process includes the following steps: S20, the workpiece is placed in place; S21, move the gantry to above the workpiece; S22, laser scanning reverse modeling; S23, automatically extracting welds and manually confirming welds; S24, generating a robot path program; S25, weld scanning measurement; S26, layer planning and process matching; S27, generating a welding program; S28, performing welding seam tracking and parameter monitoring; S29, welding is completed.

2. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 1 is characterized in that: Both the model-based welding process and the reverse-based welding process include multi-layer and multi-pass processes, which are between performing welding seam tracking, parameter monitoring and generating welding programs.

3. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 2 is characterized in that: The block nodes in the model-based welding process and the reverse-based welding process are composed of three groups of I-beams.

4. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 3 is characterized in that: The block node welding method in the model-based welding process and the reverse-based welding process includes the following steps: S30, butt weld is full penetration weld; S31. The groove processing method is flame or plasma cutting, the groove angle error is between 50°±5°, and varies randomly with the weld, the root gap range is between 3-5mm, and varies randomly with the weld, and single-sided welding and double-sided forming and no backing are required, the inner / outer skin misalignment range is between 0-2mm, and varies randomly with the weld, the blunt edge thickness range is between 0-2mm, and varies randomly with the weld, and the weld position deviation is ≥3mm; S32, select K-type groove to form symmetrical welding, so that the joint forms a symmetrical stress field, the welding position is horizontal butt, the preheating requirements must reach the minimum preheating temperature and interlayer temperature specified in WPS, arrange the welding sequence to control deformation, the welding method is robot fully automatic welding, and the welding is carried out by melting extremely rich argon gas shielded arc welding; S33. Rust, moisture, oil and other debris should be removed from the groove and within 20mm on both sides before welding; S34, welding is performed by workers, the positioning welding length is 40-50mm, the spacing is 350-400mm, and the height of the butt weld layer is 4-5mm; S35, T-type butt joint, arc-starting plate and arc-extinguishing plate are welded at both ends of the weld. After welding, the arc-starting plate and arc-extinguishing plate are removed by gas cutting, and then the cut is polished smooth; S36, the welding gun points to the midpoint of the gap, and the welding gun is tilted forward 70°, with an angle of 30° with the bottom plate to complete the base welding; S37. The filling cap welding gun swings in a triangular shape, with the counterpoint pointing to the midpoint of the base weld, and the welding gun tilts forward 70° to complete the second weld. The welding is along the intersection of the second weld and the base plate, and the welding gun tilts forward 70°.

5. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 1 is characterized in that: It also includes a block node robot intelligent welding system, which includes intelligent weld reverse extraction technology, welding real-time tracking technology, intelligent weld measurement technology, intelligent welding process development and intelligent welding decision-making technology.

6. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 5 is characterized in that: For flat workpiece welding scenarios without CAD models, the intelligent weld reverse extraction technology uses a large-field-of-view laser sensor to perform 3D scanning on the workpiece surface, and uses intelligent weld reverse extraction technology to automatically extract the weld position information. The point cloud model is denoised and optimized to remove ambient light interference and reduce stitching errors. At the same time, based on the Al algorithm, the point cloud data is processed and analyzed to extract the contour and calculate the weld position data.

7. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 5, characterized in that: The real-time welding seam tracking technology uses visual real-time tracking, optimizes the optical path of the laser sensor, and filters external light interference in the process. At the same time, based on the Al algorithm, it conducts deep learning on the image, greatly filters out external light interference, and completely extracts the contour curve of the groove under multi-layer and multi-channel conditions.

8. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 5, characterized in that: Intelligent weld contour reverse modeling technology accurately scans the weld contour, reconstructs the 3D data of the weld contour, measures the size of the groove, and performs weld layer planning and process matching. The scanning speed is more than 20 times the welding speed, and a 3D contour calculation algorithm is used. At the same time, the equal area method and equal height method are used to plan the layer of the groove measurement results.

9. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 5, characterized in that: The development route of intelligent welding process development is: material-based process database + industry scenario-based matching model, that is, process development is divided into basic process database and reasoning process database, experiments are carried out based on fixed-precision materials, and the basic process database is formed through unified processing. By measuring the weld shape, the corresponding welding parameters are obtained; combined with the intelligent reasoning model on the basis of the process database.

10. The single-sided welding and double-sided forming process of the K-type weld of the block node according to claim 5, characterized in that: Intelligent welding decision-making technology uses artificial intelligence technology to conduct big data model training and actual welding tests on welding processes, measurement results, image information, process models and other information to determine the optimal matching relationship. Finally, based on the inference process database, it combines the scene understanding of each industry, process quality requirements, and working conditions to form a matching model, and generates a matching model with real-time correction capabilities, thereby realizing intelligent decision-making in the welding process.

Images