Automatic welding system and welding method thereof

Abstract

The invention discloses an automatic welding system and a welding method thereof.The automatic welding system comprises a welding platform, a welding device, a moving device, a visual scanning module, a welding monitoring module and a laser deslagging module; the welding platform is used for bearing a to-be-welded workpiece; the welding device is used for welding a to-be-welded workpiece; the moving device is used for driving the welding device to move relative to the welding platform in at least one direction of the X axis, the Y axis and the Z axis. The visual scanning module is used for scanning a groove structure of a to-be-welded workpiece and generating welding parameters according to a scanning result; the welding monitoring module is used for scanning a welding seam in real time in the welding process and generating correction parameters according to a scanning result; the laser deslagging module is used for conducting laser deslagging treatment on the front portion of the welding path according to the correction parameters. According to the automatic welding system, welding parameters can be adjusted in real time in the welding process, automatic deviation correction is achieved, the automatic welding system is suitable for layered welding treatment of different narrow gap structures, and the welding quality is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of welding technology, and in particular, to an automatic welding system and an automatic welding method using the automatic welding system. Background Art

[0002] Currently, for the narrow-gap welding treatment with high-precision requirements, an automatic welding is usually performed by using a narrow-gap welding system. However, the traditional narrow-gap welding system simply obtains the structural parameters of the workpiece to be welded through image scanning, and then generates corresponding welding parameters according to the database for welding. The selection of its welding parameters can only rely on the support of a huge database, and it cannot adaptively adjust the welding parameters during the welding process. Moreover, even if the database stores more data, it cannot ensure accurate adaptation to the groove structures of all workpieces to be welded, and cannot guarantee the welding quality of each weld. In addition, as the working time increases, the working parameters of the welding torch will also change to a certain extent, resulting in the inconsistency between the actual working parameters and the preset welding parameters, thereby reducing the welding quality. Summary of the Invention

[0003] The primary object of the present invention is to provide an automatic welding system to solve the technical problem that the existing narrow-gap welding system cannot guarantee the welding quality of narrow-gap welding.

[0004] The present invention also provides an automatic welding method using the above automatic welding system.

[0005] According to one aspect of the present invention, an automatic welding system is provided, including a welding platform, a welding device, a moving device, a visual scanning module, a welding monitoring module, and a laser slag removal module;

[0006] The welding platform is used to carry the workpiece to be welded;

[0007] The welding device is installed above the welding platform and is used to weld the workpiece to be welded on the welding platform;

[0008] The moving device is connected to the welding platform and / or the welding device, and the moving device is used to drive the welding device to move relative to the welding platform along at least one of the X-axis, Y-axis, and Z-axis;

[0009] The visual scanning module is used to scan the groove structure of the workpiece to be welded on the welding platform, generate welding parameters according to the scanning result, and transmit the welding parameters to the welding device and the moving device;

[0010] The welding monitoring module is used to perform real-time scanning on the weld during the welding process, generate correction parameters according to the scanning result, and transmit the correction parameters to the welding device, the moving device, and the laser slag removal module;

[0011] The laser slag removal module is used to perform laser slag removal on the front of the welding path according to the correction parameters.

[0012] Preferably, the moving device includes an X-axis driving mechanism, a Y-axis driving mechanism, and a Z-axis driving mechanism. The Z-axis driving mechanism includes a first lifting component and a second lifting component;

[0013] The X-axis driving mechanism is arranged at the bottom of the welding platform and is used to drive the welding platform to move along the X-axis direction;

[0014] The first lifting component is connected to the welding device and is used to drive the welding device to move up and down along the Z-axis direction. The second lifting component is connected to the welding platform and is used to drive the welding platform to move up and down along the Z-axis direction;

[0015] The Y-axis driving mechanism is connected to the welding device and is used to drive the welding device to move along the Y-axis direction.

[0016] Preferably, the moving device further includes a rotating mechanism. The rotating mechanism is arranged between the second lifting component and the welding platform, and the rotating mechanism is used to drive the welding platform to rotate.

[0017] As a second aspect, the present invention further provides an automatic welding method. Using the above automatic welding system, the automatic welding method includes the following steps:

[0018] S100: Position the workpiece to be welded on the welding platform;

[0019] S200: Scan the groove structure of the workpiece to be welded on the welding platform through the vision scanning module, generate welding parameters according to the scanning results, and transmit the welding parameters to the welding device and the moving device;

[0020] S300: Operate the welding device and the moving device according to the welding parameters provided by the vision scanning module, weld the workpiece to be welded on the welding platform through the welding device, and adjust the welding position of the workpiece to be welded in real time through the moving device;

[0021] S400: Scan the weld seam in real time during the welding process through the welding monitoring module, generate correction parameters according to the scanning results, and correct the welding parameters of the welding device and the moving device in real time;

[0022] S500: Operate the laser slag removal module according to the correction parameters provided by the welding monitoring module, and perform laser slag removal on the front of the welding path through the laser slag removal module.

[0023] Preferably, step S200 specifically includes:

[0024] S201: Use the visual scanning module to scan the groove structure of the workpiece to be welded on the welding platform, compare the scanning result with the database. If corresponding data exists in the database, directly select the corresponding welding parameters according to the database. If corresponding data does not exist in the database, proceed to the next step;

[0025] S202: Perform image slicing on the scanning result, extract feature points layer by layer and generate a 3D model. Then divide the 3D model longitudinally into multiple layers, and a set of welding parameters is generated for each layer structure to perform layer-by-layer and pass-by-pass welding on the workpiece to be welded, and input the 3D model and the corresponding multiple layers of welding parameters into the database.

[0026] Preferably, in step S201, after using the visual scanning module to scan the groove structure of the workpiece to be welded on the welding platform, use formula one to remove noise and interference in the image, and use formula one to perform sharpening processing on the image;

[0027] Formula one:

[0028] Formula two: I(x,y) = I(x + 1,y) + I(x - 1,y) + I(x,y + 1) + I(x,y - 1) - 4I(x,y)

[0029] Where f is the input image, g is the filter signal, (i, j) are the pixel coordinates of the input image, and (x, y) are the pixel coordinates of the output image.

[0030] Preferably, in step S202, performing image slicing on the scanning result, extracting feature points layer by layer and generating a 3D model specifically includes:

[0031] Generate the feature point coordinates P(x,y,z) of the point cloud data through scanning. The feature points identify the key geometric features of the groove according to formula three, and then convert the point cloud data into a 3D model according to formula four;

[0032] Formula three:

[0033] Formula four:

[0034] Where X is the point cloud data and μ is the mean of the point cloud.

[0035] Preferably, step S400 specifically includes:

[0036] Use the welding monitoring module to perform real-time scanning on the weld during welding, use formula five to specify the welding path f(n), and use formula six to dynamically adjust the welding parameters u(t);

[0037] Formula five: f(n) = g(n) + h(n)

[0038] Formula Six:

[0039] where g(n) is the actual path from the starting point to node n, h(n) is the heuristic estimated path from node n to the target node, e(t) is the error signal, and K p is the proportionality coefficient, K i is the integral coefficient, and K d is the differential coefficient.

[0040] Preferably, in step S400, the welding monitoring module predicts the welding quality using Formula Seven and predicts the welding defects using Formula Eight;

[0041] Formula Seven: y = β 0 + β 1 x 1 + β 2 x 2 +... + β n x n + ε

[0042] Formula Eight:

[0043] where y is the welding quality, x 1 、x 2 ...x n are the welding parameters, β 1 、β 2 ...β n are the regression coefficients, f is the activation function, ω i is the weight, and b is the bias.

[0044] Preferably, in step S500, the working energy E of the laser slag removal module is calculated using Formula Seven and adjusted using Formula Eight;

[0045] Formula Seven:

[0046] Formula Eight: E pulse =P peak ·t pulse

[0047] where P is the laser power, A is the acting area of the laser, P peak is the peak power, and t pulse is the pulse width.

[0048] The present invention has the following beneficial effects:

[0049] In the automatic welding system provided by the present invention, the groove structure of the workpiece to be welded can be scanned by the vision scanning module first, and welding parameters can be generated according to the scanning results to provide initial operating parameters for the welding device and the moving device. During the welding process, the welding seam can be scanned in real time by the welding monitoring module to perform position finding and tracking during the welding process, and correction parameters can be generated according to the scanning results to achieve adaptive control of the groove size, gap size, and data fluctuations during the welding process, so as to be able to achieve automatic deviation correction and dynamically adjust the operating parameters of the welding device and the moving device, making it better adapt to the actual welding situation and effectively improving the welding quality. Secondly, the automatic welding system is also equipped with a laser slag removal module, which can perform laser slag removal treatment on the front of the welding path through the laser slag removal module, and plan the slag removal route by using the correction parameters provided by the welding monitoring module to ensure the slag removal accuracy and effect, and can accurately remove impurities such as the slag and rust on the lower layer during multi-layer welding to ensure the multi-layer welding effect.

[0050] In addition to the purposes, features, and advantages described above, the present invention has other purposes, features, and advantages. The following will refer to the drawings to further elaborate on the present invention in detail. Brief Description of the Drawings

[0051] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0052] Figure 1 is a schematic structural diagram of the automatic welding system provided by the embodiment of the present invention;

[0053] Figure 2 is a schematic structural diagram of the welding device in the automatic welding system provided by the embodiment of the present invention;

[0054] Figure 3 is a flowchart of generating welding parameters in the automatic welding method provided by the embodiment of the present invention.

[0055] Legend Explanation:

[0056] 1. Welding platform; 2. Welding device; 3. Moving device; 31. X-axis driving mechanism; 32. Y-axis driving mechanism; 33. Z-axis driving mechanism; 331. First lifting component; 4. Vision scanning module; 5. Welding monitoring module; 6. Laser slag removal module; 7. Control box. Detailed Description of the Embodiment

[0057] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, the present invention can be implemented in many different ways defined and covered by the following. Examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0058] Those skilled in the art of the present technology can understand that, unless specifically stated otherwise, the term "including" used in the specification of the present invention means the presence of the described features, integers, steps, operations, components and / or assemblies, but does not exclude the presence or addition of one or more other features, integers, steps, operations, components, assemblies and / or combinations thereof. It should be understood that when we say a component is "connected" to another component, it can be directly connected to other components or connected through intermediate components. The phrase "and / or" used here includes all or any unit and all combinations of one or more related listed items. The terms "first" and "second" etc. in the specification and claims of the present invention are used to distinguish different objects and not to describe a specific order.

[0059] Figure 1 and Figure 2 collectively show an automatic welding system provided by an embodiment of the present invention, which is used for automatically welding a narrow-gap groove of a workpiece to be welded and can adjust welding parameters in real time during the welding process to improve the welding effect.

[0060] Please refer to Figure 1 and Figure 2, the automatic welding system includes a welding platform 1, a welding device 2, a moving device 3, a vision scanning module 4, a welding monitoring module 5, and a laser slag removal module 6; the welding platform 1 is used to carry the workpiece to be welded; the welding device 2 is installed above the welding platform 1 and is used to weld the workpiece to be welded on the welding platform 1; the moving device 3 is connected to the welding platform 1 and / or the welding device 2, and the moving device 3 is used to drive the welding device 2 to move relative to the welding platform 1 along at least one of the X-axis, Y-axis, and Z-axis, so as to flexibly adjust the working position of the welding device 2 in the three-dimensional coordinate system, that is, adjust the welding position of the workpiece to be welded; the vision scanning module 4 is used to scan the groove structure of the workpiece to be welded on the welding platform 1, generate welding parameters according to the scanning results, and transmit the welding parameters to the welding device 2 and the moving device 3; the welding monitoring module 5 is used to scan the weld seam in real time during the welding process, generate correction parameters according to the scanning results, and transmit the correction parameters to the welding device 2, the moving device 3, and the laser slag removal module 6; the laser slag removal module 6 is used to perform laser slag removal treatment on the front of the welding path according to the correction parameters.

[0061] In the automatic welding system, the groove structure of the workpiece to be welded can be scanned by the vision scanning module 4 first, and welding parameters are generated according to the scanning results to provide initial operating parameters for the welding device 2 and the moving device 3; during the welding process, the weld seam is scanned in real time by the welding monitoring module 5, and position finding and tracking are performed during the welding process. Correction parameters are generated according to the scanning results to realize adaptive control of the groove size, gap size, and data fluctuations during the welding process, so as to be able to realize automatic deviation correction and dynamically adjust the operating parameters of the welding device 2 and the moving device 3 to better adapt to the actual welding situation and effectively improve the welding quality. Secondly, the automatic welding system is also equipped with a laser slag removal module 6. Through the laser slag removal module 6, laser slag removal treatment can be performed on the front of the welding path. The slag removal line is planned by using the correction parameters provided by the welding monitoring module 6 to ensure the slag removal accuracy and effect. When welding in layers, the slag, rust and other impurities on the lower layer can be accurately removed to ensure the multi-layer welding effect.

[0062] Preferably, the mobile device 3 includes an X-axis driving mechanism 31, a Y-axis driving mechanism 32, and a Z-axis driving mechanism 33. The Z-axis driving mechanism 33 includes a first lifting assembly 331 and a second lifting assembly (not shown in the figure, the same below). The X-axis driving mechanism 31 is disposed at the bottom of the welding platform 1 and is used to drive the welding platform 1 to move along the X-axis direction. The first lifting assembly 331 is connected to the welding device 2 and is used to drive the welding device 2 to move up and down along the Z-axis direction. The second lifting assembly is connected to the welding platform 1 and is used to drive the welding platform 1 to move up and down along the Z-axis direction. The Y-axis driving mechanism 32 is connected to the welding device 2 and is used to drive the welding device 2 to move along the Y-axis direction. It should be understood that the X-axis direction, the Y-axis direction, and the Z-axis direction refer to the three directions in a three-dimensional rectangular coordinate system, where the X-axis is the horizontal axis, the Y-axis is the vertical axis, and the Z-axis is the vertical axis.

[0063] Further, the Y-axis driving mechanism 32 is installed between the first lifting assembly 331 and the welding device 2, and the first lifting assembly 331 is used to drive the Y-axis driving mechanism 32 and the welding device 2 as a whole to move up and down together; or the first lifting assembly 331 is disposed between the Y-axis driving mechanism 32 and the welding device 2, and the Y-axis driving mechanism 32 is used to drive the first lifting assembly 331 and the welding device 2 as a whole to move along the Y-axis direction.

[0064] The mobile device 3 realizes the multi-directional movement of the welding device 2 relative to the welding platform 1 through the coordinated cooperation of the X-axis driving mechanism 31, the Y-axis driving mechanism 32, and the Z-axis driving mechanism 33, enabling the welding device 2 to flexibly move to any position in three-dimensional space for welding operations and ensuring welding accuracy. Secondly, since the Z-axis driving mechanism 33 includes a first lifting assembly 331 connected to the welding device 2 and a second lifting assembly connected to the welding platform 1, the lifting stroke is effectively increased, the applicability is improved, and the lifting position can also be selected according to welding requirements. For example, when it is necessary to keep the welding device 2 stationary to stabilize the welding parameters, the welding platform 1 can be driven to move up and down by the second lifting assembly to adjust the welding height; when it is necessary to keep the workpiece to be welded stationary to stabilize the welding position, the welding device 2 can be driven to move up and down by the first lifting assembly to adjust the welding height.

[0065] Specifically, the X-axis drive mechanism 31, the Y-axis drive mechanism 32, the first lifting assembly 331, and / or the second lifting assembly include a linear drive assembly and a guiding assembly. The linear drive assembly includes a drive motor and a lead screw nut assembly. The guiding assembly includes a slide rail and a slider. The slider is embedded in the slide rail and is used to move along the slide rail for guiding. The lead screw nut assembly includes a lead screw and a nut threadedly connected to the lead screw. The output shaft of the drive motor is connected to the lead screw and is used to drive the lead screw to rotate. The nut is connected to the slider. The nut is used to drive the slider to move axially along the lead screw through threaded engagement when the lead screw rotates, thereby driving the slider to slide along the slide rail to achieve linear drive. The drive motor preferably uses a servo motor to ensure the drive accuracy. In other embodiments, the linear drive assembly may also use a cylinder, a hydraulic cylinder, or a linear motor.

[0066] As Figure 1 shown, the moving device 3 further includes a rotating mechanism 34. The rotating mechanism 34 is disposed between the second lifting assembly and the welding platform 1. The rotating mechanism 34 is used to drive the welding platform 1 to rotate. By driving the welding platform 1 to rotate through the rotating mechanism 34, the workpiece to be welded on the welding platform 1 is driven to rotate, adapting to the automatic welding operation of rotary workpieces, and the welding angle can be flexibly adjusted to meet different welding requirements.

[0067] Further, the rotating mechanism 34 includes a rotating motor. The output shaft of the rotating motor is directly connected to the welding platform 1 or connected to the welding platform 1 through a speed reducer to ensure stable drive.

[0068] Further, a high-precision encoder and a torque sensor are provided on the welding platform 1 to monitor the rotation angle and load distribution of the welding platform 1 in real time, and automatically adjust the driving speed of the moving device 3 to adapt to the welding requirements of different workpieces to be welded.

[0069] Further, the welding platform 1 is provided in a disc shape, and a weight flange protruding upward is provided at the edge of the welding platform 1, so that the self-weight of the welding platform 1 can be evenly distributed along the axial direction of the rotation axis of the rotating mechanism 34 to avoid the occurrence of center of gravity deviation, and the workpiece to be welded can also be limited by the weight flange to prevent the workpiece to be welded from slipping off the welding platform 1.

[0070] Preferably, the welding platform 1 further includes clamping jaws and springs. The clamping jaws are arranged inside the welding platform 1 and are circumferentially spaced around the welding platform 1. The springs are provided in one-to-one correspondence with the clamping jaws. The first end of the spring is connected to the weight flange, and the second end of the spring is connected to the clamping jaw. The spring is used to apply an elastic force to the clamping jaw in the direction towards the center of the welding platform 1 to drive the clamping jaw to elastically clamp the workpiece to be welded.

[0071] Further, the automatic welding system further includes a control box 7. A processor is provided inside the control box 7. The welding platform 1, the welding device 2, the moving device 3, the visual scanning module 4, the welding monitoring module 5, and the laser slag removal module 6 are all connected to the control box 7. The control box 7 is used to process the scanning and detection signals of the visual scanning module 4 and the welding monitoring module 5, and respectively output corresponding welding parameters to the welding platform 1, the welding device 2, the moving device 3, and the laser slag removal module 6.

[0072] As Figure 2 shown, the welding device 2 includes a welding torch 21, a wire feeder, and a gas supply pipeline respectively connected to the welding torch 21. The wire feeder is used to convey welding wire to the welding torch 21, and the gas supply pipeline is used to convey welding protection gas to the welding torch 21. A tension sensor and a speed feedback component are provided on the wire feeder. The wire feeder is used to automatically adjust the wire feeding speed and tension according to the welding parameters generated by the visual scanning module or according to the correction parameters generated by the welding monitoring module to ensure the stable conveyance of the welding wire; a flow sensor and a pressure controller are provided on the gas supply pipeline. The gas supply pipeline is used to automatically adjust the gas flow rate and pressure according to the welding parameters generated by the visual scanning module or according to the correction parameters generated by the welding monitoring module to ensure the uniform coverage of the welding protection gas.

[0073] Further, the laser slag removal module 6 includes a high-frequency laser. The high-frequency laser uses pulsed laser technology and can perform precise laser slag removal treatment on the area to be welded before welding to ensure the slag removal effect.

[0074] As a second aspect, the present invention also provides an automatic welding method using the above automatic welding system. The automatic welding method includes the following steps:

[0075] S100: Position the workpiece to be welded on the welding platform 1;

[0076] S200: Scan the groove structure of the workpiece to be welded on the welding platform 1 through the visual scanning module 4, generate welding parameters according to the scanning results, and transmit the welding parameters to the welding device 2 and the moving device 3;

[0077] S300: Operate the welding device 2 and the moving device 3 according to the welding parameters provided by the vision scanning module 4, weld the workpiece to be welded on the welding platform 1 through the welding device 2, and adjust the welding position of the workpiece to be welded in real time through the moving device 3;

[0078] S400: During the welding process, scan the weld seam in real time through the welding monitoring module 5, generate correction parameters according to the scanning results, and correct the welding parameters of the welding device 2 and the moving device 3 in real time;

[0079] S500: Operate the laser slag removal module 6 according to the correction parameters provided by the welding monitoring module 5, and perform laser slag removal treatment on the front of the welding path through the laser slag removal module 6.

[0080] The automatic welding method first scans the groove structure of the workpiece to be welded through the vision scanning module 4, generates welding parameters according to the scanning results, and provides initial operating parameters for the welding device 2 and the moving device 3; during the welding process, the welding monitoring module 5 scans the weld seam in real time, performs position finding and tracking during the welding process, generates correction parameters according to the scanning results, realizes adaptive control of the groove size, gap size and data fluctuations during the welding process, so as to be able to realize automatic deviation correction, dynamically adjust the operating parameters of the welding device 2 and the moving device 3, make it better adapt to the actual welding situation, and effectively improve the welding quality. Secondly, the automatic welding method also performs laser slag removal treatment on the front of the welding path through the laser slag removal module 6, plans the slag removal line by using the correction parameters provided by the welding monitoring module 6, ensures the slag removal accuracy and effect, and can accurately remove impurities such as slag and rust on the lower layer during multi-layer welding, ensuring the multi-layer welding effect.

[0081] As Figure 3 shown, preferably, step S200 specifically includes:

[0082] S201: Scan the groove structure of the workpiece to be welded on the welding platform 1 through the vision scanning module 4, compare the scanning results with the database. If corresponding data exists in the database, directly select the corresponding welding parameters according to the database. If corresponding data does not exist in the database, enter the next step;

[0083] S202: Perform image slicing processing on the scanning results, extract feature points layer by layer and generate a three-dimensional model, then divide the three-dimensional model longitudinally into multiple layers, and generate a set of welding parameters for each layer structure to perform multi-layer and multi-pass welding on the workpiece to be welded, and record the three-dimensional model and the corresponding multi-layer welding parameters into the database.

[0084] Specifically, according to the scanning results of the visual scanning module 4, a large amount of past experience stored in the system database can be utilized to quickly confirm the welding parameters, select appropriate welding parameters such as current, voltage, gas flow rate, welding speed, and moving speed, ensuring the accuracy and reliability of the welding parameters. When the scanning results do not match the database, image slicing processing can be performed on the scanning results, feature points can be extracted layer by layer and a three-dimensional model can be generated. Then, the three-dimensional model is divided into multiple layers longitudinally, simplifying the complex three-dimensional structure into a multi-layer planar structure. Thus, a set of welding parameters can be generated corresponding to each layer structure to perform layer-by-layer and multi-pass welding on the workpiece to be welded, that is, automatic multi-pass layout according to the actual groove, and the three-dimensional model and the corresponding multi-layer welding parameters are entered into the database to automatically expand the database, so that when the same groove structure is scanned next time, the welding parameters can be directly confirmed according to the database, reducing the data processing process and improving the welding efficiency and welding effect.

[0085] Preferably, in step S201, after the visual scanning module 4 scans the groove structure of the workpiece to be welded on the welding platform 1, formula one is used to remove noise and interference in the image, and formula one is used to perform sharpening processing on the image;

[0086] Formula one:

[0087] Formula two: I(x,y) = I(x + 1,y) + I(x - 1,y) + I(x,y + 1) + I(x,y - 1) - 4I(x,y)

[0088] Where f is the input image, g is the filter signal, (i, j) are the pixel coordinates of the input image, and (x, y) are the pixel coordinates of the output image. An image processing algorithm based on a convolutional neural network (CNN) is used for image processing. Among them, sharpening uses the Laplace operator. Through formula one and formula two, noise and interference in the scanned image can be automatically identified and removed. Gaussian filtering or median filtering is used for denoising, and appropriate sharpening processing can be performed, effectively improving the accuracy and speed of image processing.

[0089] Preferably, in step S202, the specific steps of performing image slicing processing on the scanning results, extracting feature points layer by layer, and generating a three-dimensional model include:

[0090] The coordinate P(x,y,z) of the feature points of the point cloud data is generated through scanning. The feature points identify the key geometric features of the groove according to formula three, and then the point cloud data is converted into a three-dimensional model according to formula four;

[0091] Formula three:

[0092] Formula four:

[0093] Among them, X is the point cloud data, and μ is the mean value of the point cloud. The point cloud data is generated by laser scanning as P(c, u, x). The feature points use principal component analysis to identify the key geometric features of the groove. Through triangulation or voxelization, the point cloud data is transformed into a three-dimensional model V(x, y, z), which can quickly construct a three-dimensional model. Then, according to the algorithm, layer-by-layer and pass-by-pass welding processing is carried out to ensure accurate selection of welding parameters.

[0094] Preferably, step S400 specifically includes:

[0095] During the welding process, the welding monitoring module 5 scans the weld seam in real time, adopts the welding path g(m) of formula five specifications, and dynamically adjusts the welding parameters u(t) using formula six;

[0096] Formula five: g(m) = h(n) + h(n)

[0097] Formula six:

[0098] Among them, g(n) is the actual path from the starting point to node n, h(n) is the heuristic estimated path from node n to the target node, e(t) is the error signal, K p is the proportionality coefficient, K i is the integral coefficient, K f is the differential coefficient.

[0099] In the automatic welding method, according to the actual path from the welding starting point to node n and the heuristic estimated path from node n to the target node, the best welding path of the welding torch 21 can be accurately obtained, real-time feedback data is given, and the PID control algorithm (Proportional, Integral, Differential) is used to dynamically adjust the welding position and angle during the welding process, effectively ensuring the accuracy and timeliness of welding parameter correction and improving the welding effect.

[0100] Preferably, in step S400, the welding monitoring module 5 predicts the welding quality using formula seven and predicts the welding defects using formula eight;

[0101] Formula seven: y = β 0 + β 1 x 1 + β 2 x 2 +... + β n x n + ε

[0102] Formula eight:

[0103] Among them, y is the welding quality, x1 , x 2 ...x n are welding parameters, and β 1 , β 2 ...β n are regression coefficients, f is an activation function, ω i is a weight, and b is a bias.

[0104] Specifically, in the automatic welding method, a linear regression algorithm is used to predict the welding quality, and an SVM algorithm (Support Vector Machines) is used to classify welding defects: where ω is the normal vector of the hyperplane, C is the regularization parameter, and ξ i is the relaxation parameter. At the same time, based on the neural network algorithm, welding defects are predicted. Through algorithm optimization technology, the feedback data during the welding process can be compared with the actual welding results, and the welding parameters and algorithms can be automatically adjusted to ensure the continuous improvement of the automation level of the automatic welding system. It is also equipped with a welding quality prediction model based on historical data, which can predict the welding results according to the real-time data during the welding process and automatically adjust the welding parameters through algorithms to ensure the stability and consistency of the welding quality.

[0105] Preferably, in step S500, the working energy E of the laser slag removal module 6 is calculated using Formula Seven and adjusted using Formula Eight;

[0106] Formula Seven:

[0107] Formula Eight: E pulse = P peak ·t pulse

[0108] where P is the laser power, A is the action area of the laser, P peak is the peak power, and t pulse is the pulse width.

[0109] Through the positioning function of the welding monitoring module 5 and its feedback data, rust, water, oil, welding slag, etc. in the welding area can be captured, and the slag removal path can be planned. The best slag removal path can be automatically planned according to the geometric shape of the product to be welded and the welding requirements, enabling the laser slag removal module 6 to select an appropriate working energy to emit high-density laser energy to achieve the purpose of accurate slag removal, ensuring the slag removal effect while reducing energy consumption and time waste during the slag removal process.

[0110] Furthermore, after step S500, it further includes:

[0111] S600: After welding is completed, detect the weld flaw detection, compare the weld flaw detection with the predicted results of welding quality, calculate the error, and then optimize the welding parameters according to the error. Enter the optimized welding parameters into the database.

[0112] Further optimize the welding parameters in the database through the actual flaw detection structure of the weld, realize the automatic improvement of the database, make the selection of the next welding parameters more accurate and reliable, and improve the welding effect.

[0113] The automatic welding method is equipped with multi-sensor fusion technology, which can collect various parameters during the welding process (such as parameters like current, voltage, temperature, gas flow rate, etc.). By introducing big data analysis technology, the system can analyze the collected data in real time, predict the welding quality in advance, and based on the welding defect recognition algorithm, can automatically identify defects such as lack of fusion and slag inclusion during the welding process, and perform real-time tracking and capture through the image data of the welding monitoring module 5 to ensure the stability of welding quality.

[0114] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An automatic welding system, characterized in that: It comprises a welding platform (1), a welding device (2), a moving device (3), a visual scanning module (4), a welding monitoring module (5) and a laser slag removal module (6); The welding platform (1) is used to carry a workpiece to be welded; The welding device (2) is mounted above the welding platform (1) and is used to weld the workpiece to be welded on the welding platform (1); The moving device (3) is connected to the welding platform (1) and / or the welding device (2), and the moving device (3) is used to drive the welding device (2) to move relative to the welding platform (1) along at least one direction of the X-axis, the Y-axis, and the Z-axis; The visual scanning module (4) is used to scan the groove structure of the workpiece to be welded on the welding platform (1), generate welding parameters according to the scanning results, and transmit the welding parameters to the welding device (2) and the moving device (3); The welding monitoring module (5) is used to scan the weld in real time during the welding process, generate correction parameters according to the scanning results, and transmit the correction parameters to the welding device (2), the moving device (3) and the laser slag removal module (6); The laser slag removal module (6) is used to perform laser slag removal processing on the front of the welding path according to the correction parameters.

2. The automatic welding system according to claim 1, characterized in that: The moving device (3) comprises an X-axis driving mechanism (31), a Y-axis driving mechanism (32) and a Z-axis driving mechanism (33); the Z-axis driving mechanism (33) comprises a first lifting component (331) and a second lifting component; The X-axis driving mechanism (31) is arranged at the bottom of the welding platform (1) and is used to drive the welding platform (1) to move along the X-axis direction; The first lifting assembly (331) is connected to the welding device (2) and is used to drive the welding device (2) to move up and down along the Z-axis direction; the second lifting assembly is connected to the welding platform (1) and is used to drive the welding platform (1) to move up and down along the Z-axis direction; The Y-axis driving mechanism (32) is connected to the welding device (2) and is used to drive the welding device (2) to move along the Y-axis direction.

3. The automatic welding system according to claim 2, characterized in that: The moving device (3) further comprises a rotating mechanism (34), wherein the rotating mechanism (34) is arranged between the second lifting assembly and the welding platform (1), and the rotating mechanism (34) is used to drive the welding platform (1) to rotate.

4. An automatic welding method, characterized in that: Using the automatic welding system as described in any one of claims 1 to 3, the automatic welding method comprises the following steps: S100: Positioning the workpiece to be welded on the welding platform (1); S200: Scanning the groove structure of the workpiece to be welded on the welding platform (1) through a visual scanning module (4), generating welding parameters according to the scanning results, and transmitting the welding parameters to the welding device (2) and the moving device (3); S300: operating the welding device (2) and the moving device (3) according to the welding parameters provided by the visual scanning module (4), welding the workpiece to be welded on the welding platform (1) by the welding device (2), and adjusting the welding position of the workpiece to be welded in real time by the moving device (3); S400: Scanning the weld in real time during the welding process through the welding monitoring module (5), generating correction parameters according to the scanning results, and correcting the welding parameters of the welding device (2) and the moving device (3) in real time; S500: operating the laser slag removal module (6) according to the correction parameters provided by the welding monitoring module (5), and performing laser slag removal processing on the front of the welding path through the laser slag removal module (6).

5. The automatic welding method according to claim 4, characterized in that: Step S200 specifically includes: S201: Scanning the groove structure of the workpiece to be welded on the welding platform (1) through the visual scanning module (4), comparing the scanning result with the database, and if the database has corresponding data, directly selecting corresponding welding parameters according to the database; if the database does not have corresponding data, proceeding to the next step; S202: Perform image slicing processing on the scanning result, extract feature points layer by layer and generate a three-dimensional model, then divide the three-dimensional model into multiple layers along the longitudinal direction, generate a set of welding parameters for each layer structure, so as to perform layered and pass welding processing on the workpiece to be welded, and enter the three-dimensional model and the corresponding multi-layer welding parameters into the database.

6. The automatic welding method according to claim 5, characterized in that: In step S201, after scanning the groove structure of the workpiece to be welded on the welding platform (1) by the visual scanning module (4), noise and interference in the image are removed by using formula 1, and the image is sharpened by using formula 1; Formula 1: Formula 2: I(x,y)=I(x+1,y)+I(x-1,y)+I(x,y+1)+I(x,y-1)-4I(x,y) Where f is the input image, g is the filter signal, (i, j) is the pixel coordinate of the input image, and (x, y) is the pixel coordinate of the output image.

7. The automatic welding method according to claim 5, characterized in that: In step S202, the scanning result is subjected to image slicing processing, and feature points are extracted layer by layer to generate a three-dimensional model, which specifically includes: The feature point coordinates P (x, y, z) of the point cloud data are generated by scanning. The feature points identify the key geometric features of the groove according to Formula 3, and then the point cloud data is converted into a three-dimensional model according to Formula 4; Formula 3: Formula 4: Among them, X is the point cloud data and μ is the mean of the point cloud.

8. The automatic welding method according to claim 4, characterized in that: Step S400 specifically includes: The welding seam is scanned in real time during the welding process by means of a welding monitoring module (5), a welding path f(n) is specified by formula 5, and a welding parameter u(t) is dynamically adjusted by means of formula 6; Formula 5: f(n) = g(n) + h(n) Formula 6: Among them, g(n) is the actual path from the starting point to node n, h(n) is the heuristic estimated path from node n to the target node, e(t) is the error signal, and K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient.

9. The automatic welding method according to claim 4, characterized in that: In step S400, the welding monitoring module (5) uses formula 7 to predict welding quality and uses formula 8 to predict welding defects; Formula 7: y=β0+β1x1+β2x2+...+β n x n +e Formula 8: Among them, y is the welding quality, x1, x2...x n are welding parameters, β1, β2...β n is the regression coefficient, f is the activation function, ω i is the weight and b is the bias.

10. The automatic welding method according to claim 4, characterized in that: In step S500, the working energy E of the laser slag removal module (6) is calculated using formula 7 and increased using formula 8; Formula 7: Formula 8: E pulse =p peak ·t pulse Where P is the laser power, A is the laser action area, P peak is the peak power, t pulse is the pulse width.

Images