Welding control method, device, equipment and storage medium

Abstract

The present application discloses a welding control method, apparatus, equipment, and storage medium, relating to the technical field of girth weld welding. The welding control method comprises: acquiring weld images; splicing the weld images to obtain an initial weld curve; smoothing the initial weld curve using a B-spline curve to obtain a target weld curve; and completing welding control based on the target weld curve. The present application acquires weld images, generates a complete weld curve using image splicing technology, and removes noise and minor errors through B-spline curve smoothing to generate a smooth target weld curve; controls the movement and operation of the welding equipment based on the target weld curve, thereby achieving precise welding of the weld, improving welding consistency and quality, reducing defects and errors, and increasing production efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of girth weld welding, and in particular to welding control methods, devices, equipment and storage media. Background Art

[0002] Submerged arc welding (SAW) is widely used in industrial production for welding girth welds on pipelines, pressure vessels, and other large structural components requiring circular connections due to its high efficiency and stability. Traditional welding methods, such as manual argon arc welding (MTAW) and automatic gas shielded arc welding (GMAW), often require a high level of skill and experience when welding girth welds on these large workpieces, and face challenges such as low production efficiency and difficulty maintaining consistent weld quality. Furthermore, the labor intensity and health risks associated with these operations further drive the demand for more automated and precise welding methods.

[0003] Currently, manual argon arc welding and automatic gas shielded welding remain the most commonly used welding methods. Manual argon arc welding relies on the welder's skill and experience. Although it can achieve good weld quality in some cases, it is inefficient and difficult to maintain weld consistency in all positions. Automatic gas shielded welding has improved the degree of welding automation to some extent, but manual adjustment of the welding gun position and angle is still required. In particular, in vertical and overhead welding positions, welding parameters are difficult to accurately control. In addition, although traditional submerged arc welding can provide efficient and high-quality welding, existing equipment is mostly used in the flat welding position, has poor maneuverability, and is not suitable for girth welds. Low flux recovery efficiency also leads to resource waste.

[0004] Although existing methods have solved welding needs to a certain extent, they still have many shortcomings. Manual argon arc welding is inefficient and labor-intensive, gas shielded welding has limited automation, and weld quality is difficult to guarantee. Traditional submerged arc welding equipment has poor maneuverability and is difficult to adapt to the complex welding needs of girth welds. In addition, the flux recovery efficiency is low, resulting in waste of resources. In order to improve welding quality and efficiency, a new type of welding equipment and technology that can automatically and accurately control the welding process is urgently needed. Therefore, how to accurately weld the weld has become an urgent problem to be solved.

[0005] The above content is only used to assist in understanding the technical solution of this application and does not constitute an admission that the above content is prior art. Summary of the Invention

[0006] The purpose of this application is to provide a welding control method, device, equipment and storage medium, aiming to solve the technical problem of how to accurately weld welds.

[0007] To achieve the above objectives, the present application proposes a welding control method, which includes:

[0008] Acquire weld seam images;

[0009] splicing the weld images to obtain an initial weld curve;

[0010] Smoothing the initial weld curve using a B-spline curve to obtain a target weld curve;

[0011] Welding control is performed according to the target weld curve.

[0012] In one embodiment, the step of smoothing the initial weld curve using a B-spline curve to obtain a target weld curve includes:

[0013] performing noise analysis, cleaning, and filtering on the initial weld curve to obtain an intermediate weld curve;

[0014] Obtaining weld feature data by performing edge detection on the middle weld curve;

[0015] Detecting the intermediate weld curve by Hough transform to obtain weld contour information;

[0016] Constructing a B-spline curve according to the weld feature data and the weld profile information;

[0017] Evaluating the smoothness and continuity of the B-spline curve and the degree of closeness to the weld characteristic data to obtain an evaluation result;

[0018] The control points or node vectors of the B-spline curve are adjusted according to the evaluation result, and the adjusted B-spline curve is used as the target weld curve.

[0019] In one embodiment, the step of constructing a B-spline curve based on the weld feature data and the weld profile information includes:

[0020] Selecting the order of the B-spline basis function according to the weld profile information;

[0021] Constructing a node vector according to the weld characteristic data and determining a control point;

[0022] Constructing a B-spline basis function according to the B-spline basis function order and the node vector;

[0023] Optimizing the control points by using the least squares method to obtain optimized control points;

[0024] A B-spline curve is constructed according to the B-spline basis function and the optimized control points.

[0025] In one embodiment, the step of completing welding control according to the target weld curve includes:

[0026] Obtaining a weld gap distance and a weld gap centerline according to the target weld curve;

[0027] Controlling the moving trajectory of the welding gun according to the center line of the weld gap;

[0028] The moving speed and flux amount of the welding gun are controlled according to the weld gap distance to complete welding control.

[0029] In one embodiment, the step of obtaining the weld gap distance and the weld gap centerline according to the target weld curve includes:

[0030] Determine the upper boundary curve and the lower boundary curve of the weld according to the target weld curve;

[0031] The vertical distance between the upper boundary curve and the lower boundary curve is used as the weld gap distance;

[0032] The midpoints of the weld gap distances are connected to obtain the weld gap midline.

[0033] In one embodiment, the step of splicing the weld images to obtain an initial weld curve includes:

[0034] Extracting feature points from the weld image using a scale-invariant feature transformation algorithm to obtain feature points and feature point descriptors;

[0035] Matching the feature points of different images according to the feature point descriptors using a fast nearest neighbor library algorithm to obtain feature point pairs;

[0036] Obtaining an optimal transformation matrix based on the feature point pairs by a random sampling consensus algorithm;

[0037] aligning the weld images according to the optimal transformation matrix through three-dimensional affine transformation to obtain the aligned weld images;

[0038] The aligned weld images are fused according to the Poisson equation, and the splicing traces are smoothed to obtain a panoramic image, which is used as the initial weld curve.

[0039] In one embodiment, the step of acquiring the weld image includes:

[0040] Acquire initial weld images;

[0041] Draw a rectangular frame for the weld area in the initial weld image, and use the portion within the rectangular frame as a region of interest;

[0042] Performing threshold segmentation on the region of interest and converting the image into a grayscale image;

[0043] The grayscale image is filtered to obtain a weld image.

[0044] In addition, to achieve the above objectives, the present application also proposes a welding control device, which includes:

[0045] An acquisition module, used for acquiring weld images;

[0046] A stitching module, used for stitching the weld images to obtain an initial weld curve;

[0047] A fitting module is used to smooth the initial weld curve using a B-spline curve to obtain a target weld curve;

[0048] The control module is used to complete welding control according to the target weld curve.

[0049] In addition, to achieve the above-mentioned purpose, the present application also proposes a welding control device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the welding control method described above.

[0050] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium, and when the computer program is executed by the processor, the steps of the welding control method described above are implemented.

[0051] One or more technical solutions proposed in this application have at least the following technical effects:

[0052] The present application obtains weld images; splices the weld images to obtain an initial weld curve; smoothes the initial weld curve using a B-spline curve to obtain a target weld curve; and completes welding control according to the target weld curve. This application obtains weld images and uses a camera or other image acquisition device to obtain an initial image of the weld, providing visual data of the weld as a basis for subsequent processing and analysis; stitching the weld images, and using image stitching technology to merge multiple weld images to generate a complete weld curve, thereby forming a continuous weld contour for subsequent analysis and processing; smoothing the initial weld curve through a B-spline curve, and using a B-spline curve algorithm to smooth the initial weld curve to remove noise and slight errors, generate a smooth target weld curve, eliminate irregularities in the curve, and make the weld path smoother and more controllable; completing welding control according to the target weld curve, and using the smoothed target weld curve to control the movement and operation of the welding equipment, including the movement trajectory, speed and welding parameters of the welding gun, to ensure that the welding gun strictly follows the smooth weld path for welding, thereby achieving precise welding of the weld, improving the consistency and quality of welding, reducing welding defects and errors, and improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0055] Figure 1 A schematic diagram of a flow chart of the first embodiment of the welding control method of the present application;

[0056] Figure 2 This is a schematic diagram of the overall structure of the welding control device according to an embodiment of the present application;

[0057] Figure 3 This is a schematic diagram of a flux feeding and adjusting device of a welding control device according to an embodiment of the present application;

[0058] Figure 4 A schematic diagram of a flow chart provided for the second embodiment of the welding control method of the present application;

[0059] Figure 5 This is a schematic diagram of the module structure of the welding control device according to an embodiment of the present application;

[0060] Figure 6 Schematic diagram of the equipment structure of the hardware operating environment involved in the welding control method in the embodiment of the present application.

[0061] Description of Figure Numbers:

[0062] 1: Hydraulic cylinder; 2: Ring rail; 3: Robotic arm module; 4: Rotating stage; 5: Welding hopper; 6: Driving gear; 7: BP rotating motor; 8: Wire feeding mechanism; 9: Robotic arm; 10: Wire feeding mechanism; 11: Industrial camera; 12: Submerged arc welding gun; 13: Flux feeding adjustment device; 14: Rack; 15: Workpiece to be welded; 16: Workpiece displacement rotating base.

[0063] The purpose, features and advantages of this application will be further explained with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0064] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0065] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0066] Currently, manual argon arc welding and automatic gas shielded welding are commonly used welding methods. Manual argon arc welding relies on the welder's skills and experience. Although it can achieve good welding quality in some cases, it is inefficient and difficult to maintain consistency in all-position welding. Although automatic gas shielded welding has improved the degree of automation, it still requires manual adjustment of the welding gun position and angle, especially in the vertical and overhead welding positions, where it is difficult to accurately control the welding parameters. In addition, although traditional submerged arc welding is efficient and has good welding quality, the equipment is mainly used in the flat welding position, has poor maneuverability and is not suitable for girth weld welding, and the flux recovery efficiency is low, resulting in waste of resources. In general, although the existing methods have partially met the welding needs, they still have problems such as low efficiency, limited degree of automation, unstable weld quality and waste of resources. In order to improve welding quality and efficiency, there is an urgent need for a new type of welding equipment and technology that can automatically and accurately control the welding process.

[0067] The main solution of the embodiment of the present application is: first, use the visual inspection module to capture the weld image, and extract the initial features of the weld through edge detection technology, then apply B-spline curve fitting technology to smooth the weld curve to obtain an accurate actual weld curve, and determine the movement trajectory of the welding gun and the amount of flux coverage based on the fitted weld curve, thereby ensuring the accuracy of the welding process and the quality of the weld.

[0068] It should be noted that the execution subject of the embodiments of the present application can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, mobile phone, etc., or an electronic device, submerged arc welding system, welding equipment, etc. that can realize the above functions. The submerged arc welding system is used as an example to illustrate this embodiment and the following embodiments.

[0069] Based on this, the embodiment of the present application provides a welding control method, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the welding control method of the present application.

[0070] In this embodiment, the welding control method includes steps S10 to S40:

[0071] Step S10, acquiring a weld image;

[0072] It should be noted that weld images refer to images of the welding area collected by the visual inspection module. Specifically, these images are high-resolution images of the circumferential weld area, which can clearly show the edges and gaps of the welds. These images are used for subsequent image processing and analysis steps, such as edge detection, weld feature extraction and stitching, to generate a preliminary weld curve model. These images capture the weld characteristics and welding gaps before welding, and provide the basic data required for B-spline curve fitting and precise flux deposition control.

[0073] Please refer to Figure 2 and Figure 3 , Figure 2 This is a schematic diagram of the overall structure of the welding control device according to an embodiment of the present application. Figure 3This is a schematic diagram of the flux feeding and adjusting device of the welding control device according to an embodiment of the present application. The welding control device comprises a visual inspection module, a lifting module, a welding gun rotation and movement module, a robotic arm module, a welding workpiece displacement module, and a flux feeding module. The device primarily includes a hydraulic cylinder 1, a telescopic rod 2, a ring rail 3, a rotating stage 4, a flux hopper 5, a drive gear 6, a brushless permanent magnet (BP) rotating motor 7, a welding wire feeder 8, a robotic arm 9, a welding wire feeder 10, an industrial camera 11, a submerged arc welding gun 12, a flux feeding and adjusting device 13, a rack 14, a workpiece to be welded 15, and a workpiece displacement and rotating base 16. These components work together to achieve an automated welding process. The lifting module mainly includes a hydraulic cylinder 1 and a telescopic rod 2. The visual inspection module adopts a hand-eye vision system (Eye-in-Hand), with an industrial camera 9 installed at the end of the robotic arm 7. The use of Eye-in-Hand can greatly improve welding accuracy and flexibly adapt to welds in different positions. The welding gun rotation and movement module consists of a BP motor 7, a ring rail 3, a worktable 4, and a ring rail rack 14. The BP motor 7 drives the drive gear 6 connected to it to rotate through its output shaft, so that the rotating worktable 4 connected to the rack can achieve circumferential movement. The flux feeding module consists of a flux funnel 5, a flux discharge device 13, and a submerged arc welding gun 12. The flux discharge device 13 consists of a motor, a conveyor belt, and a flux flow rate range rotary paddle. The flux flow rate range rotary paddle has three gears: large, medium, and low. The appropriate gear is automatically matched according to the width of the weld to be detected. In this embodiment, the submerged arc welding system first places the workpiece to be welded on the workpiece displacement rotating base 16 to ensure that the workpiece is stable and in the correct welding position. Then, the required flux and welding wire are loaded through the flux funnel 5 and the wire reel 8. The motor is controlled by the motion control card to rotate, thereby driving the welding gun to rotate the mobile module to achieve uniform rotation of the welding area. The device is equipped with a sensor to ensure that it can trigger the camera to take pictures every 60° rotation. The weld is photographed using a high-resolution industrial camera 11 to capture a clear image of the weld. The device first extracts the edge features of the weld before welding and uses advanced image processing technology to accurately capture the contour information of the weld. Subsequently, the obtained weld feature data is modeled using a B-spline curve fitting algorithm to generate a smooth and accurate curve model. Based on the fitting curve, the average distance of the weld (weld gap before welding) is calculated, and this key parameter is fed back to the flux control system in real time to ensure that the flux can accurately cover the weld area, thereby optimizing the welding process. Furthermore, the device dynamically plans the movement trajectory of the submerged arc welding torch during the welding process based on the B-spline curve fitting results. By precisely controlling the position and moving speed of the welding gun, uniform and stable welding of the weld seam is achieved, thus improving welding quality and efficiency.In addition, the device also has real-time monitoring and adjustment functions, which can automatically adjust the welding gun path and flux supply according to the actual changes in the weld, ensuring the stability of the entire welding process and the consistency of the weld quality.

[0074] It can be understood that the visual inspection module is used to capture high-resolution images of the circumferential weld area, clearly display the weld edges and gaps, and generate images for subsequent edge detection, weld feature extraction and stitching, providing basic data for the creation of weld curve models and precise flux deposition control.

[0075] As an example, the steps of obtaining a weld image include: collecting an initial weld image; drawing a rectangular frame on the weld area in the initial weld image, and taking the part within the rectangular frame as the region of interest; performing threshold segmentation on the region of interest and converting it into a grayscale image; filtering the grayscale image to obtain a weld image.

[0076] The initial weld image refers to the raw image captured directly from the weld area, an unprocessed weld photograph. This is the first image of the weld area captured by the system's visual inspection module. The weld region refers to the portion of the image containing the actual weld, i.e., the area where the metal materials melt and connect during the welding process. This area is the focus of detection and analysis. The rectangular frame is a rectangular boundary drawn in the initial weld image to enclose the weld region. This frame helps identify and isolate the weld portion in the image, making subsequent processing more focused and efficient. The region of interest (ROI) is the portion of the image within the rectangular frame, which includes the weld region and its surrounding areas and is the focus of subsequent image processing. Threshold segmentation compares the pixel values ​​in the image with a set threshold and classifies the pixels into two categories based on the comparison result: one category with pixel values ​​above the threshold and the other with pixel values ​​below the threshold. This method is used to separate the weld area from the background. The grayscale image converts the ROI into an image containing only grayscale levels (black and white). The color information of each pixel is represented by different grayscale values ​​from black to white. This conversion simplifies image processing and makes edge detection and feature extraction more efficient. First, an initial image of the welding area is collected through the visual inspection module. Then, a rectangular frame is drawn in the initial image to frame the weld area, so that the part within the frame is the region of interest. Next, this region of interest is threshold segmented to divide the image into two parts: the weld and the background. At the same time, it is converted into a grayscale image, that is, only the grayscale information of each pixel is retained. Finally, the generated grayscale image is filtered to remove noise and enhance details, ultimately obtaining a clear weld image.

[0077] Step S20, stitching the weld images to obtain an initial weld curve;

[0078] It should be noted that the initial weld curve refers to a continuous curve of the weld path formed by stitching together multiple weld images in sequence. This curve represents the actual trajectory of the weld during the entire welding process and can be used to analyze and evaluate the quality and consistency of the weld.

[0079] As you can understand, multiple processed weld images are first arranged and stitched together in sequence. These images represent the weld area after previous acquisition and processing. Next, an image stitching algorithm aligns the edges of these images to form a continuous weld image. The weld centerline, or the main weld path, is then extracted from this continuous weld image. Ultimately, this centerline or path forms the initial weld curve, representing the actual weld trajectory throughout the welding process. This curve can be used to assess weld quality, detect welding defects, and perform weld tracking and analysis.

[0080] As an example, the steps of splicing the weld images to obtain an initial weld curve include: extracting feature points from the weld images through a scale-invariant feature transformation algorithm to obtain feature points and feature point descriptors; matching the feature points of different images according to the feature point descriptors through a fast nearest neighbor library algorithm to obtain feature point pairs; obtaining an optimal transformation matrix based on the feature point pairs through a random sampling consistency algorithm; aligning the weld images through a three-dimensional affine transformation according to the optimal transformation matrix to obtain the aligned weld images; fusing the aligned weld images according to the Poisson equation, and smoothing the splicing marks to obtain a panoramic image, and using the panoramic image as the initial weld curve.

[0081] The Scale-Invariant Feature Transform (SIFT) algorithm is used to detect and describe local features in images. The features it extracts are invariant to changes in scale, rotation, and brightness, and are commonly used for image matching and stitching. Feature points are salient and recognizable points in an image, typically corners, edges, or other pixels with local variations, representing key locations in the image. Feature point descriptors are vectorized representations that describe the image information surrounding a feature point. They encode the local image patch around the feature point into a high-dimensional vector for easy matching and comparison. The Fast Nearest Neighbor Library (FLANN) algorithm is an efficient algorithm for nearest neighbor search in high-dimensional space and is commonly used for image feature point matching to accelerate the matching process. Feature point pairs are corresponding feature points found in two images, representing the location of the same object in different images. These correspondences can be used for image alignment and stitching. The Random Sample Consensus Algorithm (RANSAC) is an iterative algorithm for estimating model parameters from a set of data. It is particularly well-suited for data containing a large number of outliers (noise). It uses random sampling and model validation to find the best-fitting model. The optimal transformation matrix is ​​a mathematical matrix used to transform points in one image to another image so that the corresponding feature points in the two images are aligned as much as possible. It is the best transformation parameter calculated by the algorithm. The RANSAC algorithm is used to estimate the optimal transformation matrix so that its feature points are aligned with the feature points of the reference image. For each randomly selected feature point pair {i, j}, the RANSAC algorithm formula is as follows:

[0082]

[0083] Where H is the transformation matrix, and T(H,pi) is the point p after applying the transformation matrix H. i The position of , N is the number of feature point pairs.

[0084] The three-dimensional affine transformation is a linear transformation used to rotate, scale, and translate images in three-dimensional space. It can perform geometric transformations on images to achieve image alignment and stitching. Using the three-dimensional affine transformation, images taken from different angles can be aligned so that they can accurately match when stitched to form a continuous, seamless panoramic image. In addition, the affine transformation allows the image to maintain its size ratio during the transformation, which is crucial for maintaining the true size and shape of the weld gap. When stitching images, it is necessary to fuse the overlapping areas of the images to eliminate the seams. The three-dimensional affine transformation helps to accurately determine these overlapping areas and ensure the smoothness of the fusion process. Finally, in the weld image, the shape and position of the circumferential weld vary. If the camera has posture changes during the shooting process (such as pitch or roll), the three-dimensional affine transformation can model and correct these changes. For each non-reference image, a change matrix is ​​estimated to align its feature points with the feature points of the reference image. Assume that we have two sets of feature points: one set is the feature points P of the reference image. ref , the other group is the feature points P of the image to be transformed src The three-dimensional affine transformation matrix A can be expressed as:

[0085]

[0086] where a 11 ,a 22 ,a 33 is the scaling factor, a 12 ,a 13 ,a 21 ,a 23 ,a 31 ,a 32 is the rotational component, t x ,t y ,t z is the translation component. Since the image is taken along the circumference, the main transformation is the rotation around the center of the circle, so we can simplify the transformation matrix and mainly consider rotation and translation. The rotation matrix R can be expressed as:

[0087]

[0088] Where θ is the rotation angle. For every 60° image, θ can be 0, 60, 120, 180, 240, 300 degrees in radians. The translation vector t can be calculated based on the radius r of the circle and the rotation angle θ:

[0089]

[0090] Each image is transformed using the estimated transformation matrix A, adjusting their position and orientation so that they are aligned around the reference image.

[0091] Poisson's equation is a partial differential equation used for image fusion and restoration in image processing. By solving the Poisson equation, the edges of the stitched images can be smoothed so that the stitching marks are not obvious. A panoramic image is a wide-angle image stitched together from multiple images. By stitching the weld images into a panoramic image, a complete weld trajectory can be obtained. This image is the initial weld curve. Images that have undergone three-dimensional affine transformation may produce discontinuities or obvious marks at the seams when stitched. Image fusion can smooth these marks to make the stitching look more natural. There may be differences in color and brightness between different images. Image fusion can adjust these differences so that the entire stitched image is visually consistent. Using the Poisson equation to fuse images can well preserve the edge information of the image and produce a smooth transition in the overlapping area. The objective function of Poisson fusion can be expressed as:

[0092]

[0093] I1, I2 refer to the two images that need to be fused, μ refers to the fusion function, represents the gradient operator, Ω is the pixel domain of the image, and λ is the regularization parameter. Poisson fusion can be used to smooth the spliced ​​weld images. Since the weld area may have irregularities, Poisson fusion can achieve a smooth transition of the image while maintaining the sharpness of the weld edge, which is very important for subsequent B-spline fitting.

[0094] First, the scale-invariant feature transform (SIFT) algorithm is used to extract feature points and feature point descriptors from each weld image. These feature points identify the salient areas in the weld image and their local image features. Then, the fast nearest neighbor library (FLANN) algorithm is applied to find similar feature points in different weld images based on these feature point descriptors to form feature point pairs. Then, the random sample agreement (RANSAC) algorithm is used to screen the most stable matching points from these feature point pairs and calculate the optimal transformation matrix. This matrix is ​​used for three-dimensional affine transformation to accurately align different weld images so that they overlap correctly in space. Next, the aligned weld images are fused using the Poisson equation to eliminate splicing marks and smooth transition areas, ultimately obtaining a seamless panoramic image. This panoramic image fully displays the overall picture of the weld, called the initial weld curve, and provides a continuous and clear weld trajectory, which facilitates subsequent analysis and processing.

[0095] Step S30, smoothing the initial weld curve using a B-spline curve to obtain a target weld curve;

[0096] It should be noted that the B-spline curve refers to a mathematical curve interpolation method, commonly used to smooth and approximate a set of discrete data points. The target weld curve refers to the final weld curve obtained after B-spline smoothing, which is the curve trajectory or path of the smoothed weld. During the submerged arc welding process of girth welds, due to the limitations of the submerged arc welding process, the weld is deposited with flux during welding, making it impossible to correct the welding gun's movement trajectory in real time. Ensuring that the welding gun moves precisely along the weld center is critical to ensuring welding quality. The B-spline curve fitting method can accurately fit the control points of the initial weld curve obtained from the visual inspection module, generating a continuous weld curve and providing an accurate movement trajectory for the welding gun. In addition, the local control characteristics of the B-spline curve allow for fine-tuning of specific areas of the weld curve without having to refit the entire curve. Furthermore, the weld image obtained through B-spline curve fitting can accurately calculate the average weld width, thereby achieving refined control of the flux dosage, which is crucial for avoiding excessive or insufficient solder during the welding process.

[0097] As you can understand, first, the discrete data points in the initial weld curve are interpolated using the characteristics of the B-spline curve to obtain a smooth curve. During this process, the B-spline curve can ensure smooth transitions in various parts of the curve and stay as close to the original data points as possible by controlling the position of the points and the degree of the curve, thereby eliminating any possible noise or discontinuities. The resulting target weld curve is smoother and more continuous, making it suitable for subsequent analysis and application.

[0098] Step S40: completing welding control according to the target weld curve.

[0099] It should be noted that welding control refers to the analysis and understanding of the target weld curve, using automated equipment and control algorithms to accurately guide the welding equipment to perform welding operations along this curve. This process includes adjusting the welding equipment's speed, angle, and welding parameters (such as current and voltage) to ensure that the welding process strictly follows the target weld curve, ultimately achieving high-quality, precise welding results. Through welding control, problems such as weld offset and welding defects can be effectively avoided, ensuring the strength and appearance quality of the weld.

[0100] It can be understood that, first, the specific coordinates and direction of the welding path are determined by analyzing the target weld curve; then, automated welding equipment, such as a robotic welding arm, is used in combination with advanced control algorithms to accurately guide the equipment according to the analyzed weld curve; during the process, the control system will continuously adjust the welding parameters (such as welding speed, current, voltage, etc.) to ensure that the welding head moves smoothly along the predetermined curve and welds; the real-time monitoring and feedback mechanism will ensure that any deviations in the welding process are corrected in time, thereby completing high-precision, high-quality welding operations; finally, the welding equipment will complete the welding task in strict accordance with the target weld curve to ensure the structural strength and aesthetics of the weld.

[0101] As an example, the steps of completing welding control according to the target weld curve include: obtaining the weld gap distance and the weld gap centerline according to the target weld curve; controlling the movement trajectory of the welding gun according to the weld gap centerline; controlling the movement speed and flux amount of the welding gun according to the weld gap distance to complete welding control.

[0102] The weld gap distance refers to the distance or gap between the two pieces of material to be welded during the welding process, which determines the amount of welding material filling and the setting of welding parameters; the weld gap centerline refers to the center line of the weld gap distance, which is usually used as the reference line of the moving path of the welding gun to ensure accurate positioning during the welding process; the welding gun is a part of the welding equipment used to apply welding material and heat to melt and connect materials. The welding gun can be a handheld or automated device mounted on a robot; the moving trajectory is the path that the welding gun moves along the weld gap centerline during the welding process, which determines the exact application position of the welding material; the moving speed refers to the speed at which the welding gun moves during the welding process. The moving speed affects the melting and cooling time of the welding material, thereby affecting the welding quality; the flux amount refers to the amount of welding material applied during the welding process. The flux amount must match the weld gap distance to ensure the strength and integrity of the welded connection. First, the weld gap distance and weld gap centerline are extracted based on the target weld curve. The weld gap distance refers to the distance between the two pieces of material to be welded, while the weld gap centerline is the centerline of these distances and serves as a reference path for the welding gun's movement. Next, the welding gun's movement trajectory is controlled based on the weld gap centerline, ensuring that the welding gun welds along the predetermined path. Simultaneously, the welding gun's movement speed and flux amount are adjusted based on the weld gap distance to ensure that the amount of welding material applied matches the gap size. This precise control enables high-quality welding results and ultimately completes welding control.

[0103] As an example, the steps of obtaining the weld gap distance and the weld gap centerline according to the target weld curve include: determining the weld upper boundary curve and the lower boundary curve according to the target weld curve; taking the vertical distance between the upper boundary curve and the lower boundary curve as the weld gap distance; and connecting the midpoints of the weld gap distances to obtain the weld gap centerline.

[0104] The upper boundary curve refers to the curve profile of the upper edge of the weld in the target weld curve; the lower boundary curve refers to the curve profile of the lower edge of the weld in the target weld curve; the vertical distance refers to the vertical distance between the upper boundary curve and the lower boundary curve, that is, the gap distance between the two sides of the weld; the weld gap distance refers to the above vertical distance, that is, the vertical distance between the boundary curves on both sides of the weld. The vertical distance is calculated using the vertical distance formula:

[0105]

[0106] All calculated vertical distances d(u) are accumulated and the sum is divided by the number of sampling points N to obtain the average vertical distance D avg :

[0107] In this embodiment, the average vertical distance D avg Set the control threshold of the flux amount and set a tolerance range ±∈ for the control threshold to adapt to the fluctuations in the welding process. During the welding process, the vertical distance of the weld is detected in real time. According to the comparison between the monitored distance and the control threshold, the flux supply amount is automatically adjusted; if the monitored distance is greater than D avg +∈(the upper limit of the control threshold), increase the flux supply to ensure that the wide weld can be properly filled. If the monitored distance is less than D avg -∈(lower limit of the control threshold), the flux supply should be reduced to avoid excessive flux filling in narrow welds. This embodiment uses a belt drive system driven by a small motor to automatically adjust the flux flow rate to achieve precise control. The system includes a rotary paddle designed with three fixed positions, corresponding to the three flux flow gears of large, medium and low. By developing a motor control strategy, the speed of the small motor is automatically adjusted according to the comparison between the real-time monitored weld width data and the preset control threshold, and the rotary paddle is driven to the corresponding gear to accurately control the flux supply. In addition, the system is equipped with a feedback adjustment mechanism, which uses sensors to monitor the actual flux supply to ensure consistency with the control threshold.

[0108] The weld gap centerline is a straight line obtained by connecting the midpoints of the weld gap distance, which is used to control the movement trajectory of the welding gun. First, according to the target weld curve, the upper boundary curve and the lower boundary curve of the weld are determined. These curves represent the edge contours of the weld above and below, respectively; then, the vertical distance between the upper boundary curve and the lower boundary curve is calculated. This distance is called the weld gap distance, which is used to describe the width of the gap between the boundaries on both sides of the weld; then, the midpoint of the weld gap distance is found, and the two midpoints are connected to obtain the weld gap centerline; the weld gap centerline plays an important role in welding control. It is used to guide the movement trajectory of the welding gun and ensure that the position control of the welding gun during welding is accurate and stable. The midpoint line is a virtual curve located between the two boundary curves. It divides the weld gap into two equal parts. Since the pipe girth weld is along a three-dimensional path, the three coordinate axes x, y and z need to be considered when calculating the midpoint line. Note P upper (u)=(x upper (u),y uppee (u),z upper (u)) is the point on the upper boundary curve at parameter u, P lower (u)=(x lower (u),y lower (u),z lower (u)) is a point on the lower boundary curve at parameter u. Then the midpoint line P mid (u) can be calculated as follows:

[0109]

[0110] To ensure P mid (u) is continuous and smooth in the entire parameter range. The value range of u should cover the entire length of the weld. As the parameter u changes, a series of points P are generated. mid (u), these points define the midpoint line C mid (u), using the generated midpoint line C mid (u) to guide the welding gun to move along the center of the weld.

[0111] This embodiment includes a lifting module, a visual monitoring module, a welding gun rotating and moving module, a robotic arm module, a welding workpiece displacement module, and a flux feeding module. The lifting module mainly includes a hydraulic cylinder and a telescopic rod. The hydraulic cylinder drives the telescopic rod to perform telescopic movement by providing power, thereby adjusting the vertical position of the welding gun or camera to meet the welding requirements of different heights. The visual inspection module recognizes the weld image. The visual inspection module is arranged on the welding gun rotating and moving platform and is used to identify the weld (welding gap before welding) and complete the extraction of weld edge features. The hardware composition of the visual inspection module includes an industrial camera, a lens, and an image acquisition card. The working mode of the workstation camera is real-time video streaming, and the image data is transmitted to the visual host in real time via a gigabit network cable. The welding gun rotation module uses a servo motor as its power source, providing precise and stable driving force to meet the requirements for precise positioning of the welding gun during the welding process. The servo motor drives the welding gun rotation module through gear meshing, achieving precise rotational positioning of the welding gun on the ring track to adapt to welding requirements in different positions. The welding gun rotation module is intelligently controlled by a programmable logic controller (PLC), which cooperates with an advanced motion controller to achieve precise control of the servo motor. The control system also includes a sensor system for real-time detection and adjustment of the welding gun position, as well as a computing module for image processing and data analysis. The robotic arm module consists of a six-axis robotic arm. The robotic arm can perform multi-dimensional movement through precise control and cooperates with the wire feeding mechanism to achieve precise wire feeding and welding operations. The end of the robotic arm is equipped with a welding gun and a camera. The welding workpiece positioner module consists of a base, a workpiece placement table, a rotation mechanism, a positioning mechanism, and a control module. The base serves as the support structure for the positioner, ensuring the stability of the entire equipment. The worktable is used to place the workpiece and can be fixed, rotated, or tilted to accommodate different processing requirements. The rotation mechanism is controlled by a servo motor and uses a ball screw system to achieve precise rotation of the worktable. The control module, which integrates a PLC, is responsible for precisely controlling the rotation of the workpiece positioner and is coordinated with the welding gun rotation module to ensure the optimal positional relationship between the welding gun and the workpiece during welding. The flux feeder module consists of a flux conveyor, a flux hopper, and a flux flow control device. The flux conveyor uses an external gas source to continuously deliver flux directly to the welding point. The flux hopper is used to hold the flux. The flux feeder module is a key component for achieving precise flux deposition and consists of several key components: a flux hopper, a flux discharge device, and a submerged arc welding gun. The flux hopper's primary function is to guide and store the flux, ensuring smooth flow to the discharge device. The flux feeder is the core component that controls the flux flow. It is driven by a motor and delivers the flux from the hopper via a conveyor belt. The speed of the conveyor belt is controlled by the motor, and the flux flow rate can be adjusted according to the actual needs of the welding process.The flux flow rate selector rotary paddle is a key component of the flux feeding device. It is designed with three fixed positions, corresponding to high, medium, and low flux flow settings. These positions can be automatically adjusted based on the detected weld width to match the most appropriate flux quantity. For example, when the weld is wide, the system automatically selects a high flow rate position to ensure that the weld is fully filled; when the weld is narrow, a smaller flow rate position is selected to avoid excessive flux accumulation. The entire flux feeding module is designed to achieve precise control of the flux supply. Through an automated adjustment mechanism, it ensures uniform distribution and optimized use of the flux during the welding process, thereby improving welding quality and efficiency while reducing material waste.

[0112] This embodiment provides a welding control method, which includes acquiring weld images; splicing the weld images to obtain an initial weld curve; smoothing the initial weld curve using a B-spline curve to obtain a target weld curve; and completing welding control based on the target weld curve. This embodiment obtains weld images and uses a camera or other image acquisition device to obtain an initial image of the weld, providing visual data of the weld as a basis for subsequent processing and analysis; stitching the weld images, and using image stitching technology to merge multiple weld images to generate a complete weld curve, thereby forming a continuous weld contour for subsequent analysis and processing; smoothing the initial weld curve using a B-spline curve, and using a B-spline curve algorithm to smooth the initial weld curve to remove noise and slight errors, generate a smooth target weld curve, eliminate irregularities in the curve, and make the weld path smoother and more controllable; completing welding control according to the target weld curve, and using the smoothed target weld curve to control the movement and operation of the welding equipment, including the movement trajectory, speed and welding parameters of the welding gun, to ensure that the welding gun strictly follows the smooth weld path for welding, thereby achieving precise welding of the weld, improving the consistency and quality of welding, reducing welding defects and errors, and improving production efficiency.

[0113] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above embodiment 1 can be referred to the above introduction and will not be described in detail later. Figure 4 , Figure 4 This is a flow chart of the second embodiment of the welding control method of the present application. Step S30 of the welding control method includes steps S31 to S36:

[0114] Step S31, performing noise analysis, cleaning and filtering on the initial weld curve to obtain an intermediate weld curve;

[0115] It should be noted that noise analysis refers to the inspection and evaluation of the initial weld curve to identify and quantify the noise components in the curve. These noises may be due to interference or inaccurate data generated during the image acquisition process. Cleaning refers to the removal of unnecessary noise and error data from the initial weld curve. This process may include deleting outliers, smoothing mutation points, or correcting errors, with the aim of improving the quality and accuracy of the curve. Filtering refers to the application of mathematical filtering technology to process the weld curve to further eliminate remaining high-frequency noise and subtle fluctuations, thereby making the curve smoother and more stable. Commonly used filtering techniques include low-pass filtering and Gaussian filtering. The intermediate weld curve refers to the weld curve obtained after noise analysis, cleaning and filtering. It is smoother and more accurate than the initial weld curve, but may require further processing to become the final target weld curve for welding control.

[0116] It can be understood that first, a noise analysis is performed on the initial weld curve to identify and quantify the noise components in the curve and clarify which parts need to be processed; then, unnecessary noise and error data are removed through a cleaning step, which includes deleting outliers, smoothing mutation points, and correcting deviations in the data; then, filtering techniques are applied, such as low-pass filters or Kalman filters, to further eliminate remaining high-frequency noise and subtle fluctuations, making the curve smoother and more stable; finally, after these processes, a higher quality and more accurate intermediate weld curve is obtained, providing a reliable data basis for subsequent welding control.

[0117] Step S32, obtaining weld feature data by performing edge detection on the middle weld curve;

[0118] It should be noted that edge detection refers to the use of image processing algorithms to identify and extract edge information in the middle weld curve. These edges represent the significant features and change points of the weld. Commonly used algorithms include Canny edge detection, Sobel operator, etc.; weld feature data refers to the key feature points and geometric parameters of the weld extracted through edge detection, including the weld contour, width, position and other important information that can describe the weld shape and structure.

[0119] It can be understood that by performing edge detection on the intermediate weld curve, image processing algorithms (such as Canny edge detection or Sobel operator) are used to identify and extract the edge information of the weld. These edges represent the significant features and change points of the weld. After edge detection, key characteristic data of the weld is obtained, including geometric parameters such as the weld profile, width, and position. This data can describe the shape and structure of the weld in detail, providing accurate information for subsequent welding control and quality assessment.

[0120] Step S33, detecting the middle weld curve by Hough transform to obtain weld contour information;

[0121] It should be noted that the Hough Transform is an image processing algorithm used to detect geometric shapes in images, such as lines and circles. It maps points in the image into a parameter space in order to identify sets of points that conform to certain specific shapes. Weld contour information refers to the weld boundary and outline detected by the Hough Transform, including detailed information such as the weld shape, location, length, and direction. This information can be used to accurately describe the weld's geometric characteristics, providing an important basis for adjusting the welding process and controlling quality.

[0122] It can be understood that the Hough transform is used to detect the intermediate weld curve and identify and extract the weld boundary and contour using the Hough transform algorithm. This transform maps points in the image to parameter space and determines the shape and position of the weld by detecting high-density areas in the parameter space. Ultimately, the weld contour information is obtained, including the weld's geometric characteristics such as shape, location, length, and direction. This information provides an important basis for adjusting the welding process and controlling quality.

[0123] Step S34, constructing a B-spline curve based on the weld feature data and the weld profile information;

[0124] It should be noted that before performing B-spline fitting, data preparation is required to obtain a set of coordinate points. These coordinate point sets are used to create a mathematical model of an object or phenomenon:

[0125] P={(x1,y1),(x2,y2),…,(x m ,y m )}

[0126] Select the appropriate order k of the B-spline basis function. The order k determines the smoothness of the curve and the flexibility of fitting. The choice of order k depends on the characteristics of the data, the accuracy requirements of the fitting, the smoothness of the curve, and the complexity of the calculation. The knot vector plays a vital role in the B-spline curve. It determines the local control characteristics and shape of the curve. The knot vector is a set of non-decreasing parameter values ​​that defines the scope of the B-spline basis function. The total number of knots in the knot vector is determined by the formula n+k+1, where n is the control

[0127] ″″″

[0128] The number of points, and k is the order of B-spline. Let C={(x1,y1),(x2,y2),…,(x n ,y n )} is a set of control points, and then the least squares method is used to adjust the control points to minimize the fitting error and reduce the difference between the B-spline curve and the actual data points. The minimization objective function is as follows:

[0129]

[0130] Where P(u i ) is the B-spline curve with parameter u i The point at which.

[0131] The equation of the B-spline curve is as follows:

[0132]

[0133] Among them C i (i=0,1,...,n) is the control vertex, N i,k (t)(i=1,2,…,n) is called B-spline basis function, and its recursive formula is:

[0134]

[0135] Where k is the order of the B-spline basis function.

[0136] It can be understood that, first, the weld feature data is used as control points, and the spatial position and characteristics of these points are used to determine the key shape and path of the weld; then, the weld contour information is used to correct and optimize these control points to ensure that the curve can accurately reflect the actual shape of the weld; then, the B-spline curve algorithm is applied to generate a smooth curve through these optimized control points. The algorithm calculates a smooth and continuous curve based on the position of the control points, making the weld path more accurate and smooth; finally, the constructed B-spline curve can adapt to the various details and complex shapes of the weld, providing an accurate welding trajectory for the welding robot, thereby improving welding quality and efficiency.

[0137] As an example, the steps of constructing a B-spline curve based on the weld feature data and the weld profile information include: selecting the order of the B-spline basis function based on the weld profile information; constructing a node vector based on the weld feature data and determining the control points; constructing a B-spline basis function based on the order of the B-spline basis function and the node vector; optimizing the control points by the least squares method to obtain the optimized control points; and constructing a B-spline curve based on the B-spline basis function and the optimized control points.

[0138] The order of a B-spline basis function is a parameter of a B-spline curve. It refers to the order of the basis function, which controls the smoothness and continuity of the curve. A higher order results in a smoother curve, but also increases the computational complexity. The knot vector is a sequence that defines the B-spline basis functions and contains the parameter values ​​used in the curve generation process. It determines the domain and shape of the basis functions, influencing the curve's form and the influence range of the control points. Control points are key elements in B-spline curve construction; they define the curve's shape. The positions of these points allow the B-spline basis functions to generate a smooth curve, and adjustment of the control points directly affects the curve's direction. B-spline basis functions are the fundamental functions used to generate a B-spline curve. The final B-spline curve is obtained through a weighted combination of these basis functions, which determine the influence of the control points on the curve. Least squares is an optimization technique used to minimize the sum of squared errors between data points and the fitted curve. In B-spline curve construction, the least squares method is used to optimize the positions of control points to ensure that the curve better fits the weld characteristic data. First, the appropriate B-spline basis function order is selected according to the weld profile information. In this embodiment, the order is 3. This step involves evaluating the complexity of the weld and the smoothness requirements of the curve, and selecting the appropriate order to ensure that the curve can be accurately fitted without overfitting. Then, a node vector is constructed according to the weld feature data. The specific steps are to distribute the feature data in the node vector according to certain rules, and distribute the nodes in a uniform or non-uniform manner to ensure that the curve has high accuracy at the key points. The nodes should be evenly distributed within the range of the data points to ensure the local control characteristics of the curve. The number of nodes is usually one more than the control points to ensure the smoothness of the curve. Then, the node vector is constructed according to the weld feature data. The specific steps are to distribute the feature data in the node vector according to certain rules, and distribute the nodes in a uniform or non-uniform manner to ensure that the curve has high accuracy at the key points. The nodes should be evenly distributed within the range of the data points to ensure the local control characteristics of the curve. The number of nodes is usually one more than the control points to ensure the smoothness of the curve. The B-spline basis function is constructed using the selected B-spline basis function order and node vectors, and the value of each basis function is calculated one by one according to the mathematical definition to form the basic construction framework of the curve; then, the position of the control points is optimized using the least squares method, which includes defining an objective function that measures the error between the curve and the data points, and minimizing the error through an iterative algorithm, thereby adjusting the position of the control points to obtain the best fit effect; finally, the final B-spline curve is generated based on the optimized control points and B-spline basis function. By combining the optimized control points and basis functions, the coordinates of each point of the curve are calculated, resulting in a smooth curve that accurately describes the shape of the weld.

[0139] Step S35, evaluating the smoothness and continuity of the B-spline curve and its closeness to the weld characteristic data to obtain an evaluation result;

[0140] It should be noted that smoothness refers to the smoothness of the B-spline curve over its entire domain, that is, the continuity of the curve's derivatives. A smooth curve means that its variation at any given point is continuous. Continuity refers to the smooth transition of the curve when connecting different segments or control points. Specifically, B-spline curves are usually at least second-order continuous, which means that the first-order and second-order derivatives of the curve at the connection points are continuous and have no obvious "corners." Closeness refers to the degree of fit or match between the B-spline curve and the given weld characteristic data. This includes the deviation or error of the curve at a given data point, and evaluates how effectively the curve describes the shape of the actual weld. The evaluation results refer to the conclusions or summaries drawn from the above three aspects of the evaluation of the B-spline curve. These results can include quantitative or qualitative descriptions of the curve's smoothness, continuity, and degree of fit with the actual data, in order to determine whether the curve meets design or engineering requirements.

[0141] It can be understood that the smoothness is evaluated first, that is, the continuity of the derivative of the curve in the entire definition domain to ensure that there are no sudden changes or non-smooth areas; secondly, the continuity of the curve is checked to ensure a smooth transition when connecting different segments or control points, usually at least second-order continuity is required; then the closeness of the curve to the actual weld characteristic data is evaluated, which includes the fitting accuracy and error analysis of the curve at the data points; finally, the evaluation results are obtained based on these evaluation criteria to determine whether the B-spline curve meets the engineering design requirements or needs further adjustment to improve its accuracy and adaptability.

[0142] Step S36: adjusting the control points or node vectors of the B-spline curve according to the evaluation result, and using the adjusted B-spline curve as the target weld curve.

[0143] It should be noted that the target weld curve refers to the final B-spline curve obtained after evaluation and possible adjustment. This curve is considered to be a weld curve that meets the design requirements and evaluation standards. The control points or node vectors of the B-spline curve are adjusted to optimize its smoothness, continuity, and closeness to the actual weld characteristic data to meet design and engineering requirements.

[0144] It can be understood that the control points or node vectors of the B-spline curve are adjusted according to the evaluation results in order to optimize the smoothness, continuity and matching degree of the curve with the actual weld characteristic data. This step usually involves adjusting the position of the control points or adjusting the distribution of the node vectors so that the B-spline curve better meets the design requirements and evaluation criteria. The adjusted B-spline curve is regarded as the target weld curve, which reflects the final curve shape after optimization and adjustment and is suitable for subsequent welding control and analysis.

[0145] This embodiment performs noise analysis, cleaning, and filtering on the initial weld curve to obtain an intermediate weld curve; performs edge detection on the intermediate weld curve to obtain weld feature data; detects the intermediate weld curve using Hough transform to obtain weld profile information; constructs a B-spline curve based on the weld feature data and the weld profile information; evaluates the smoothness, continuity, and proximity of the B-spline curve to the weld feature data to obtain an evaluation result; adjusts the control points or node vectors of the B-spline curve based on the evaluation result, and uses the adjusted B-spline curve as the target weld curve. First, the initial weld curve is subjected to noise analysis, cleaning, and filtering to remove interference information and make the curve smoother and more accurate. By performing edge detection on the intermediate weld curve, the precise contour and geometric features of the weld can be extracted, improving the reliability of the data. The intermediate weld curve is detected using Hough transform technology to obtain detailed weld contour information, which helps to accurately identify the weld shape. A B-spline curve is constructed based on the weld feature data and contour information to ensure that the curve can accurately fit the actual shape of the weld. The smoothness, continuity, and closeness of the B-spline curve to the weld feature data are evaluated to verify the quality and accuracy of the curve. Based on the evaluation results, the control points or node vectors of the B-spline curve are adjusted to further optimize the curve to make it more consistent with the actual weld shape. Finally, the optimized B-spline curve is used as the target weld curve. This processing process improves the accuracy and consistency of the welding path and ensures welding quality.

[0146] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the welding control method of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0147] This application also provides a welding control device, please refer to Figure 5 , the welding control device comprises:

[0148] An acquisition module 10 is used to acquire a weld image;

[0149] A stitching module 20 is used to stitch the weld images to obtain an initial weld curve;

[0150] A fitting module 30 is used to smooth the initial weld curve using a B-spline curve to obtain a target weld curve;

[0151] The control module 40 is used to complete welding control according to the target weld curve.

[0152] The welding control device provided in this application, utilizing the welding control method described in the aforementioned embodiments, can address the technical problem of precisely welding welds. Compared to the prior art, the welding control device provided in this application achieves the same beneficial effects as the welding control method described in the aforementioned embodiments. Other technical features of the welding control device are the same as those disclosed in the aforementioned embodiments and are not further elaborated upon here.

[0153] The present application provides a welding control device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the welding control method in the above-mentioned embodiment 1.

[0154] Reference below Figure 6 , which shows a schematic structural diagram of a welding control device suitable for implementing the embodiments of the present application. The welding control device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 6 The welding control device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0155] like Figure 6As shown, the welding control device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the welding control device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input devices 1007 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and communication device 1009. Communication device 1009 can allow the welding control device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a welding control device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems may be implemented or have alternatively.

[0156] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program comprising program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0157] The welding control device provided in this application, utilizing the welding control method described in the aforementioned embodiment, can solve the technical problem of precisely welding welds. Compared to the prior art, the welding control device provided in this application achieves the same beneficial effects as the welding control method described in the aforementioned embodiment. Other technical features of the welding control device are the same as those disclosed in the aforementioned embodiment and are not further elaborated here.

[0158] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0159] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0160] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer program) stored thereon, and the computer-readable program instructions are used to execute the welding control method in the above embodiment.

[0161] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0162] The computer-readable storage medium may be included in the welding control device, or may exist independently without being assembled into the welding control device.

[0163] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the welding control device, the welding control device is caused to: acquire weld images; splice the weld images to obtain an initial weld curve; smooth the initial weld curve using a B-spline curve to obtain a target weld curve; and complete welding control according to the target weld curve.

[0164] Computer program code for performing the operations of the present application may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0165] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, program segment or a part of code, and the module, program segment or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0166] The modules described in the embodiments of the present application may be implemented in software or hardware, wherein the name of a module does not necessarily limit the unit itself.

[0167] The computer-readable storage medium provided herein stores computer-readable program instructions (i.e., a computer program) for executing the aforementioned welding control method, thereby solving the technical problem of precisely welding welds. Compared to the prior art, the beneficial effects of the computer-readable storage medium provided herein are similar to those of the welding control method provided in the aforementioned embodiments and are not further elaborated here.

[0168] The present application also provides a computer program product, comprising a computer program, which implements the steps of the welding control method as described above when the computer program is executed by a processor.

[0169] The computer program product provided in this application can solve the technical problem of how to accurately weld a weld. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the welding control method provided in the above embodiment, and will not be repeated here.

[0170] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A welding control method, characterized in that: The method comprises: Acquire weld seam images; splicing the weld images to obtain an initial weld curve; Smoothing the initial weld curve using a B-spline curve to obtain a target weld curve; Complete welding control according to the target weld curve; The step of splicing the weld images to obtain an initial weld curve comprises: Extracting feature points from the weld image using a scale-invariant feature transformation algorithm to obtain feature points and feature point descriptors; Matching the feature points of different images according to the feature point descriptors using a fast nearest neighbor library algorithm to obtain feature point pairs; Obtaining an optimal transformation matrix based on the feature point pairs by a random sampling consensus algorithm; aligning the weld images according to the optimal transformation matrix through three-dimensional affine transformation to obtain the aligned weld images; The aligned weld images are fused according to the Poisson equation, and the splicing traces are smoothed to obtain a panoramic image, which is used as the initial weld curve.

2. The method according to claim 1, wherein The step of smoothing the initial weld curve using a B-spline curve to obtain a target weld curve comprises: performing noise analysis, cleaning, and filtering on the initial weld curve to obtain an intermediate weld curve; Obtaining weld feature data by performing edge detection on the middle weld curve; Detecting the intermediate weld curve by Hough transform to obtain weld contour information; Constructing a B-spline curve according to the weld feature data and the weld profile information; Evaluating the smoothness and continuity of the B-spline curve and the degree of closeness to the weld characteristic data to obtain an evaluation result; The control points or node vectors of the B-spline curve are adjusted according to the evaluation result, and the adjusted B-spline curve is used as the target weld curve.

3. The method according to claim 2, wherein The step of constructing a B-spline curve according to the weld feature data and the weld profile information comprises: Selecting the order of the B-spline basis function according to the weld profile information; Constructing a node vector according to the weld characteristic data and determining a control point; Constructing a B-spline basis function according to the B-spline basis function order and the node vector; Optimizing the control points by using the least squares method to obtain optimized control points; A B-spline curve is constructed according to the B-spline basis function and the optimized control points.

4. The method according to claim 1, wherein The step of completing welding control according to the target weld curve comprises: Obtaining a weld gap distance and a weld gap centerline according to the target weld curve; Controlling the moving trajectory of the welding gun according to the center line of the weld gap; The moving speed and flux amount of the welding gun are controlled according to the weld gap distance to complete welding control.

5. The method according to claim 4, wherein The step of obtaining the weld gap distance and the weld gap centerline according to the target weld curve comprises: Determine the upper boundary curve and the lower boundary curve of the weld according to the target weld curve; The vertical distance between the upper boundary curve and the lower boundary curve is used as the weld gap distance; The midpoints of the weld gap distances are connected to obtain the weld gap midline.

6. The method according to claim 1, wherein The step of obtaining the weld image comprises: Acquire initial weld images; Draw a rectangular frame for the weld area in the initial weld image, and use the portion within the rectangular frame as a region of interest; Performing threshold segmentation on the region of interest and converting the image into a grayscale image; The grayscale image is filtered to obtain a weld image.

7. A welding control device, characterized in that: The welding control device is used to implement the welding control method according to any one of claims 1 to 6, and the device includes: An acquisition module, used for acquiring weld images; A stitching module, used for stitching the weld images to obtain an initial weld curve; The splicing module is further used to extract feature points from the weld image using a scale-invariant feature transformation algorithm to obtain feature points and feature point descriptors; Matching the feature points of different images according to the feature point descriptors using a fast nearest neighbor library algorithm to obtain feature point pairs; Obtaining an optimal transformation matrix based on the feature point pairs by a random sampling consensus algorithm; aligning the weld images according to the optimal transformation matrix through three-dimensional affine transformation to obtain the aligned weld images; The aligned weld images are fused according to the Poisson equation, and stitching marks are smoothed to obtain a panoramic image, which is used as an initial weld curve; A fitting module is used to smooth the initial weld curve using a B-spline curve to obtain a target weld curve; The control module is used to complete welding control according to the target weld curve.

8. A welding control device, characterized in that: The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the welding control method according to any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the welding control method according to any one of claims 1 to 6 are implemented.

Images