Laser welding method and device

Abstract

The invention relates to a laser welding method and device, belongs to the technical field of metal material laser welding, and is particularly suitable for welding a special-shaped contact surface between a workpiece with a porous structure and a connecting piece of the workpiece. The method comprises the steps that a to-be-welded part is placed in a welding clamp, space coordinates of a contact face are obtained, and an NURBS curve equation of a welding path is generated; molten pool images are collected in real time, and the molten pool area is calculated; the real-time power of the laser is dynamically adjusted according to the change of the area of the molten pool, so that the real-time power of the laser is reduced along with the increase of the area of the molten pool and is finally kept at a constant power value; a local air curtain is adopted to protect a welding area, and the flow of protective gas is dynamically and reasonably adjusted. According to the invention, the accuracy of welding path positioning can be improved, the laser welding quality is improved, and the consumption of protective gas is saved.

Description

Technical Field

[0001] The present invention belongs to the technical field of laser welding of metal materials, and specifically relates to a laser welding method and device, which is particularly suitable for welding the special-shaped contact surface of a workpiece with a porous structure and its connecting parts. Background Art

[0002] In the welding of complex metal structures, especially in the laser welding of the special-shaped contact surfaces of workpieces with porous structures and their connectors, the laser welding of porous workpieces (such as workpieces with mesh or grid structures) and their connectors usually faces the following technical difficulties: complex path planning, geometric mutations at the edges of the holes make the positioning accuracy of the welding path low, which may cause large errors and require repeated positioning; uneven heat conduction, abnormal heat flow distribution due to the porous structure, often leading to inconsistent penetration depth and weld porosity problems, and damage to the metal surface due to poor heat input control; certain highly active metals (such as titanium alloys, aluminum alloys, etc.) oxidize quickly at high temperatures, and when the formation of an oxide layer is suppressed by a shielding gas to avoid the formation of an oxide layer on the periphery of the weld surface after laser welding, there is also a problem of shielding gas utilization efficiency; how to avoid various welding defects as much as possible during the welding process, such as porosity, cracks, slag inclusions, etc.

[0003] In the Chinese patent with publication number CN102363247B and title “Laser welding method and laser welding device”, the rotating platform relies on mechanical clamping, which is prone to vibration when rotating at high speed and further causes the weld to shift, and the laser head only performs linear reciprocating motion and cannot adapt to the welding of special-shaped contact surfaces.

[0004] It is particularly noteworthy that in traditional laser welding, although shielding gas is usually used as a protective measure to prevent the formation of an oxide layer on the weld surface, the shielding gas is continuously sprayed when the laser processing head moves to the welding position, which will lead to waste of shielding gas or reduced utilization efficiency. Especially in the case of poor welding path planning or poor positioning accuracy, excessive shielding gas spraying is often required to maintain stable coverage. Summary of the invention

[0005] The technical problem that the present invention attempts to solve is how to accurately plan the welding path and control the laser power to improve the welding quality for metal laser welding of special-shaped contact surfaces, and how to save the consumption of shielding gas as much as possible without affecting the welding quality.

[0006] In order to overcome the deficiencies in the prior art and solve the above-mentioned technical problems, the present invention provides a laser welding method and device, and adopts the following technical solutions.

[0007] In a first aspect, the present invention provides a laser welding method suitable for welding a special-shaped contact surface between a workpiece having a porous structure and a connecting part thereof, the method comprising:

[0008] S100, placing a workpiece with a porous structure and a connecting piece thereof in a welding fixture, obtaining the spatial coordinates of the welding contact surfaces of the two, and generating a NURBS curve equation of a welding path;

[0009] S200, real-time acquisition of a molten pool image and calculation of a molten pool area, the process comprising: acquisition of a molten pool image, preprocessing of the molten pool image, segmentation of the molten pool image, extraction of molten pool image features, and calculation of a molten pool area;

[0010] S300, dynamically adjusting the real-time power of the laser according to the change of the molten pool area, wherein the real-time power decreases as the molten pool area increases and eventually remains at a constant power value;

[0011] S400. A local air curtain is used to protect the welding area, and the flow rate of the shielding gas is dynamically adjusted. When the temperature of the molten pool increases, the flow rate of the shielding gas increases on the basis of the baseline flow rate. When the temperature of the molten pool decreases, the flow rate of the shielding gas decreases on the basis of the baseline flow rate.

[0012] Optionally, the real-time power in step S300 satisfies the calculation formula In this formula, P real is the real-time power of the laser, P1 is the preset reference power, the value range is set to 300-800W, α1 is the gain coefficient, the value range is set to 0.1-2.0, S real is the real-time molten pool area, S target is the target molten pool area.

[0013] Optionally, the real-time power in step S300 satisfies the calculation formula In this formula, P real is the real-time power of the laser, P2 is the preset reference power, the value range is set to 300-800W, α2 is the gain coefficient, the value range is set to 0-2.0, S real is the real-time molten pool area, S target is the target molten pool area.

[0014] Optionally, the protective gas flow rate in step S400 satisfies the calculation formula In this formula, Q real is the real-time gas flow rate, Q1 is the preset reference flow rate, set to 15-25L / min, β1 is the temperature change response factor, and its value range is set to 0.5-2.0. is the rate of change of molten pool temperature.

[0015] Optionally, the protective gas flow rate in step S400 satisfies the calculation formula In this formula, Q real is the real-time gas flow rate, Q2 is the preset reference flow rate, set to 15-25L / min, β2 is the temperature change response factor, and its value range is set to 0.1-1.0. is the rate of change of molten pool temperature.

[0016] Optionally, the preprocessing of the molten pool image in step S200 includes:

[0017] S221, grayscale conversion: converting a color image into a grayscale image;

[0018] S222, Image enhancement: Improve image quality by contrast enhancement or histogram equalization;

[0019] S223, noise removal: use median filtering or Gaussian filtering to remove image noise.

[0020] Optionally, the segmentation of the molten pool image in step S200 includes:

[0021] S231, threshold segmentation: separating the molten pool from the background color;

[0022] S232, edge detection: find out the contour of the molten pool;

[0023] S233, region growing: merge adjacent pixels and form a molten pool area.

[0024] Optionally, the extraction of the molten pool image features in step S200 includes:

[0025] S241, contour extraction: extracting the contour of the molten pool using a contour tracking algorithm;

[0026] S242, Morphological operation: remove small noise points and fill internal holes.

[0027] Optionally, the calculation of the molten pool area in step S200 includes a pixel counting method and a contour integration method.

[0028] In a second aspect, the present invention provides a laser welding device suitable for welding a special-shaped contact surface between a workpiece having a porous structure and a connecting part thereof, the device comprising:

[0029] M100, laser welding fixture: used to fix the porous structure workpiece to be welded and its connecting parts, made of high temperature resistant and high strength materials, with a structure with adjustable clamping size;

[0030] M200, Laser Welding Unit: It includes a laser generator, a coaxial vision system, and a laser output mirror group. The laser output power of the laser generator can be adjusted in terms of magnitude.

[0031] M300, Imaging Monitoring Unit: It is used for imaging the weld pool between the porous structure workpiece and its connecting part.

[0032] M400, Gas Supply Unit: It is used to supply shielding gas to the welding area, and its gas output can be adjusted in terms of magnitude.

[0033] M500, Temperature Measurement Unit: It is used to measure the temperature of the weld pool in real time.

[0034] M600, Central Control Unit: It controls and implements the laser welding method described in any one of claims 1 - 9.

[0035] Regarding the beneficial effects produced by the technical solution of the present invention, it is described as follows.

[0036] First, in combination with the porous characteristics of the workpiece, the welding path of the laser is dynamically adjusted through the NURBS curve algorithm to bypass the hole area, improving the positioning accuracy of the path and preventing ineffective laser irradiation caused by holes. This not only reduces the need for re - welding but also reduces the risk of heat accumulation on the welding surface.

[0037] Second, the state change of the weld pool area is monitored in real time, and the real - time power of the welding laser is dynamically adjusted from high to low. Through the optimization and matching process of the heat output process of the weld point, it avoids the splashing of liquid metal foam caused by too rapid increase in the weld point temperature, further avoids the formation of local pits or burrs at the weld point, improves the consistency of the weld penetration of the weld point, and also reduces the false - welding rate of the weld point.

[0038] Third, by dynamically adjusting the shielding gas flow rate, the oxygen content in the weld seam is reduced, the formation of oxides is inhibited, and at the same time, the reasonable consumption of the shielding gas is achieved. Especially when the temperature rises rapidly, the amount of shielding gas is increased to cope with the higher oxidation risk. Description of the Drawings

[0039] Figure 1 : Block diagram of the method of the present invention;

[0040] Figure 2 : Process block diagram for collecting weld pool images and calculating the weld pool area;

[0041] Figure 3 : One of the embodiments of the real - time laser temperature change curve;

[0042] Figure 4 : Another embodiment of the real - time laser temperature change curve;

[0043] Figure 5 : Block diagram of the preprocessing process of the melt pool image;

[0044] Figure 6 : Block diagram of the segmentation process of the melt pool image;

[0045] Figure 7 : Block diagram of the melt pool image feature extraction process. DETAILED DESCRIPTION

[0046] In order to more clearly illustrate the features of the technical solution of the present invention, the present invention is further described in detail below through specific implementation methods and in conjunction with the accompanying drawings.

[0047] As a first embodiment, the present invention provides a laser welding method suitable for welding a special-shaped contact surface between a workpiece having a porous structure and a connecting part thereof, the method comprising:

[0048] S100, placing a workpiece with a porous structure and a connecting piece thereof in a welding fixture, obtaining the spatial coordinates of the welding contact surfaces of the two, and generating a NURBS curve equation of a welding path;

[0049] S200, real-time acquisition of a molten pool image and calculation of a molten pool area, the process comprising: acquisition of a molten pool image, preprocessing of the molten pool image, segmentation of the molten pool image, extraction of molten pool image features, and calculation of a molten pool area;

[0050] S300, dynamically adjusting the real-time power of the laser according to the change of the molten pool area, wherein the real-time power decreases as the molten pool area increases and eventually remains at a constant power value;

[0051] S400. A local air curtain is used to protect the welding area, and the flow rate of the shielding gas is dynamically adjusted. When the temperature of the molten pool increases, the flow rate of the shielding gas increases on the basis of the baseline flow rate. When the temperature of the molten pool decreases, the flow rate of the shielding gas decreases on the basis of the baseline flow rate.

[0052] A schematic block diagram of the method process of the present invention is shown in FIG. Figure 1 As shown in the figure, this method involves four parts: creating a welding path, calculating the molten pool area, dynamically adjusting the laser power, and dynamically adjusting the flow rate of the shielding gas.

[0053] It should be noted that in the production practice of metal laser welding, before the welding operation, the parts to be welded often involve necessary pretreatment steps, such as surface cleaning of the parts to be welded, including the removal of impurities such as surface oil and oxide layer, etc., which can be done by chemical cleaning, mechanical grinding and other methods.

[0054] In step S100, there are various specific implementation methods for the method of obtaining the spatial coordinates of the welding contact surface between the workpiece and its connecting piece in the prior art. For example:

[0055] First, the laser scanning method. Using the principle of laser ranging, by emitting laser light to an object and measuring the time of the reflected light, the three-dimensional coordinates of the object are calculated.

[0056] Second, the structured light scanning method. By projecting a specific light pattern (such as stripes) onto an object and using a camera to capture the deformed pattern due to the change in the shape of the object, the three-dimensional shape of the object is calculated.

[0057] Third, the optical triangulation method. Using multiple cameras with known angles to capture images of an object, and calculating the three-dimensional coordinates through the principle of triangulation.

[0058] Fourth, the stereo vision method. Using two or more cameras to capture the same scene from different angles, and calculating the depth information through the principle of parallax.

[0059] Fifth, the time-of-flight method. By measuring the time required for light to travel from emission to return to calculate the distance, and then obtaining the three-dimensional coordinates.

[0060] Sixth, the mechanical three-dimensional scanning method. Using a robotic arm equipped with a contact or non-contact sensor for scanning.

[0061] The NURBS curve equation described in step S100 is expressed as:

[0062]

[0063] In the above formula, C(u) is the point on the curve, is the p-th B-spline basis function defined on the knot vector, P i is the vertex of the control polygon, i.e., the control point, W i is the weight factor associated with the control point P i u is the parameter, usually taking values within the interval defined by the knot vector, for example, u ∈ [0, 1], n is the value obtained by subtracting 1 from the number of control points, and p is the degree of the B-spline curve.

[0064] The above B-spline basis function itself is defined recursively as follows:

[0065]

[0066] The knot vector [u0, u1, …, u m is a non-decreasing sequence, where m = n + p + 1.

[0067] The NURBS curve (Non-Uniform Rational B-Splines) here, as a mathematical model in geometric figures, is used as the curve equation for generating and representing the spatial coordinates of the contact surface between two workpieces in the technical solution of this application.

[0068] The NURBS curve is currently widely used in fields such as animation production. It can accurately represent conic curves (circles, ellipses, hyperbolas, and parabolas) and polynomial curves. The continuity and smoothness of this curve can be controlled by the selection of control points and knot vectors. By adjusting the weights, the shape of the curve can be easily locally controlled.

[0069] In this application, for the porous structure of the workpiece, an algorithm created using the NURBS curve can dynamically correct the welding path. By automatically avoiding the void area, the positioning accuracy of the path is improved, and laser dry burning caused by holes is avoided. Of course, further, the high-precision positioning naturally reduces the repair welding operation, thereby also reducing the possibility of heat accumulation on the welding surface.

[0070] In step S200, collecting the molten pool image is the basis for calculating the molten pool area and further dynamically adjusting the laser power. It is achieved through a camera system including a visible light camera or an infrared camera: the former can directly capture the dynamic changes of the molten pool during the welding process, preferably a coaxial high-speed camera; while the latter obtains the image by detecting the thermal radiation difference between the molten pool and the surrounding environment.

[0071] Regarding the process block diagram for collecting the molten pool image and calculating the molten pool area, as Figure 2 shown.

[0072] When collecting the molten pool image, a camera and an optical system (such as a filter, lighting, etc.) are used to obtain a clear image of the molten pool, and it should be ensured that the imaging time and frame rate of the camera are suitable for the dynamic changes of the molten pool during the welding process.

[0073] When preprocessing the molten pool image, the purpose is to improve the quality of the image, making it easier for computer processing and analysis. This includes reducing the complexity of the image data, improving the visual effect of the image, and enhancing the clarity and readability of the image.

[0074] When segmenting the molten pool image, the purpose is to subdivide the digital image into several regions or objects to simplify the image information, facilitate the understanding of the image, and be conducive to feature extraction and target recognition, etc.

[0075] When extracting image features, representative information that helps in image analysis and recognition needs to be extracted from the image.

[0076] When calculating the molten pool area, it is necessary to quantitatively calculate the size of the target area in the image.

[0077] Step S300 defines the calculation method for adjusting the laser power magnitude.

[0078] During the laser welding process of the present application, the contradiction between breaking through thermal damage and welding strength is balanced through gradient energy input. On the one hand, during the welding process, due to excessive heat input, the microstructure of the material changes, thus affecting the performance of the welded joint, namely the so-called thermal damage. The changes in this thermal damage include: the growth of metal grains caused by high temperature, thereby reducing the strength and toughness of the joint; the generation of cracks and pores in the material caused by high temperature, and these defects will reduce the overall strength of the welded joint, etc. On the other hand, in order to obtain a high-strength welded joint, it is necessary to ensure that the weld area has a uniform and fine grain structure and no defects.

[0079] During the laser welding process of the solution of the present application, along the length or depth direction of the weld, the energy input of the laser is not constant but varies. By using a higher energy input at the beginning of the weld, it is possible to ensure the formation of a metallurgical bond at the welded joint and increase the penetration depth; at the same time, the gradient energy input helps to control the size and shape of the molten pool, so that the molten pool can form a uniform grain structure during the cooling process; and in different regions of the weld, by adjusting the energy input, it is possible to reduce the cracks generated due to excessive thermal stress and the pores generated due to too rapid cooling of the molten pool; also, the gradient energy input can achieve different cooling rates in different parts of the weld, which helps to refine the grains and improve the strength of the welded joint.

[0080] In addition, too high a laser power density will cause the metal to evaporate rapidly, forming a high pressure that pushes the molten liquid metal to splash. Therefore, by gradually decreasing the laser output power, the occurrence of welding spatter can be prevented, and at the same time, the laser oscillation welding method can be used to further reduce the generation of spatter.

[0081] In summary, by dynamically adjusting the laser power through gradient energy input, the matching of the molten pool area and power is achieved. Without affecting the welding strength, the thermal damage on the metal surface can be reduced, thereby improving the consistency of the penetration depth and the stability of the welding.

[0082] Step S400 defines the calculation method for adjusting the shielding gas magnitude.

[0083] During the laser welding process, the purpose of using a shielding gas is usually to prevent the welding area from reacting with oxygen, nitrogen, and other gases in the air, thereby avoiding oxidation, nitridation, and other unwanted chemical reactions. Usually, in the form of local shielding gas coverage, the oxygen content in the weld can be reduced, and the formation of metal oxides can be inhibited. The commonly used shielding gases in current laser welding include: argon, helium, nitrogen, hydrogen, and mixed gases (such as the mixture of argon and helium, the mixture of argon and carbon dioxide, etc.).

[0084] Compared with the traditional continuous gas supply mode, through the dynamic adjustment of the shielding gas flow rate in this application, while achieving basic protection, the consumption of the shielding gas is saved to realize the rational utilization of the shielding gas. Among them, the faster the temperature rises, the more shielding gas is provided. This is considered based on the fact that the higher the temperature, especially the faster the temperature rises, the greater the risk of metal oxidation, and the more shielding gas is needed. And reducing the shielding gas supply during the cooling process is considered from the perspective of gas conservation.

[0085] As one of the specific embodiments, the real-time power in step S300 satisfies the calculation formula In this formula, P real is the real-time power of the laser, P1 is the preset reference power, and its value range is set to 300 - 800W. α1 is the gain coefficient, and its value range is set to 0.1 - 2.0. S real is the real-time molten pool area, and S target is the target molten pool area.

[0086] According to this calculation formula, the real-time power of the laser shows a trend of changing from large to small. Initially, the value of the area S real of the molten pool surface is 0, and the real-time laser power P real = P1·(1 + α1). As the temperature of the metal surface spot position continues to rise, the area of the molten pool surface gradually increases, and the volume of the liquid metal in the molten pool also becomes more and more. If the laser power remains unchanged at a high temperature, the rapid increase in the temperature of the metal surface at the spot is likely to cause the splashing of liquid metal foam and form pits or burrs locally at the solder joint. Here, by dynamically adjusting the laser power, the power of the laser is gradually reduced as the molten pool area increases. When the real-time molten pool area reaches the target molten pool area, that is, S target - S real = 0, the real-time laser power P real= P1. Here, by gradually reducing the laser power, the aim is to slow down the heating rate of the metal surface at the light spot, avoiding the splashing of liquid metal foam. Meanwhile, by slowing down the heating rate of the molten pool, the change process of the molten pool becomes easier to control, resulting in better consistency of the penetration depth, reducing the penetration depth volatility, and also reducing the porosity (false soldering rate) of the welded solder joints.

[0087] It should be noted here that when S target -S real = 0, the real-time power P real linearly decays to P1 and remains at this power value at the end of the welding process of one solder joint, rather than immediately completely stopping the heat output. The purpose is to suppress the solidification cracks of the solder joint.

[0088] As the second specific embodiment, the real-time power described in step S300 satisfies the calculation formula In this formula, P real is the real-time power of the laser, P2 is the preset reference power, and its value range is set to 300 - 800 W. α2 is the gain coefficient, and its value range is set to 0 - 2.0. S real is the real-time molten pool area, and S target is the target molten pool area.

[0089] According to this calculation formula, the real-time power of the laser also shows a decreasing trend from large to small. Initially, the area S real value of the molten pool surface is 0, and the real-time laser power at this time As the temperature at the light spot position on the metal surface continuously increases, the area of the molten pool surface gradually increases. Eventually, when S target -S real = 0, P real = P2, that is, the real-time power of the laser reaches the reference power and remains at this power value at the end of the welding process of one solder joint.

[0090] In the above two embodiments regarding the calculation of the laser real-time power, the real-time power values both show a characteristic of decreasing as the molten pool area S real continuously increases. The difference is that: referring to Figure 3 , the calculated values in the first embodiment decrease uniformly in a linear manner; referring to Figure 4 , the calculated values in the second embodiment decrease at an exponential rate, that is, the laser real-time power is fast first and then slow during the entire decreasing process.

[0091] As the first specific embodiment, the shielding gas flow rate described in step S400 satisfies the calculation formula In this formula, Q realis the magnitude of the real-time gas flow rate, Q1 is a preset reference flow rate, set to 15 - 25 L / min, β1 is the temperature change response factor, and its value range is set to 0.5 - 2.0, is the rate of change of the molten pool temperature.

[0092] According to this calculation formula, when the temperature of the molten pool rises, that is, is positive, the flow rate of the shielding gas increases based on the reference flow rate, and the faster the temperature rises, the greater the increase in the gas flow rate; conversely, when the temperature of the molten pool drops, that is, is negative, the flow rate of the shielding gas decreases based on the reference flow rate, and the faster the temperature drops, the greater the decrease in the gas flow rate.

[0093] As the second specific embodiment, the shielding gas flow rate described in step S400 satisfies the calculation formula In this formula, Q real is the magnitude of the real-time gas flow rate, Q2 is a preset reference flow rate, set to 15 - 25 L / min, β2 is the temperature change response factor, and its value range is set to 0.1 - 1.0, is the rate of change of the molten pool temperature.

[0094] In the above two embodiments regarding the real-time flow rate calculation of the shielding gas, the shielding gas flow rate shows a characteristic of increasing with the rise of the molten pool temperature and decreasing with the drop of the molten pool temperature. Only in the first embodiment, the increase or decrease of the shielding gas flow rate changes linearly with the temperature change rate, while in the second embodiment, it changes exponentially.

[0095] As a specific embodiment, the preprocessing of the molten pool image described in step S200 includes:

[0096] S221, Grayscale conversion: Convert the color image into a grayscale image;

[0097] S222, Image enhancement: Improve the image quality through methods such as contrast enhancement or histogram equalization;

[0098] S223, Noise elimination: Use median filtering or Gaussian filtering to remove image noise.

[0099] In this specific embodiment, the image preprocessing process is as Figure 5 shown.

[0100] In the grayscale conversion of converting the color image into a grayscale image, the three color channels of red, green, and blue included in the color image are converted into one channel included in the grayscale image to simplify the computational amount of subsequent data processing. It is required that the important feature information of the image is not lost during the conversion process and the brightness information of the original image is reflected.

[0101] The image is enhanced by two methods: contrast enhancement or histogram equalization to improve the image quality. Among them, the former method is to adjust the contrast of the image to make the dark areas in the image darker and the bright areas brighter, thereby enhancing the contrast of the image; the latter method is to change the histogram distribution of the image to make the gray-scale distribution of the image more uniform, thereby improving the overall contrast, especially when both the background and foreground of the image are too bright or too dark.

[0102] Two methods, median filtering or Gaussian filtering, are used to remove image noise. Among them, the former filtering method is to select the median value in the neighborhood around each pixel in the image to replace the value of that pixel. This method is particularly effective for removing salt-and-pepper noise; the latter filtering method is to use a filter with a Gaussian function as the convolution kernel. This method can effectively remove Gaussian noise in the image while preserving the edge information of the image. Note that while removing noise, the damage to the image edges and details should be minimized.

[0103] As a specific embodiment, the segmentation of the molten pool image described in step S200 includes:

[0104] S231. Threshold segmentation: Separating the molten pool from the background color;

[0105] S232. Edge detection: Finding out the contour of the molten pool;

[0106] S233. Region growing: Merging adjacent pixels and forming the molten pool region.

[0107] In this specific embodiment, the image segmentation process is as Figure 6 shown.

[0108] During threshold segmentation, the gray-scale characteristics of the image are used to separate the target region from the background region in the image. Specifically, according to the gray-scale difference between the molten pool and the background, a suitable threshold is selected to separate the molten pool from the background. The segmented image should reflect the true shape and size of the molten pool as accurately as possible.

[0109] During edge detection, common edge detection algorithms such as Canny and Sobel can be used. The contour of the molten pool in the image usually represents the boundary of the molten pool.

[0110] During region growing, starting from the seed points, adjacent pixels are merged according to the similarity criterion to form the molten pool region. The selection of the seed points should be based on the known features of the molten pool, such as gray-scale value, texture, etc., to ensure the correctness of the region growing.

[0111] As a specific embodiment, the extraction of the molten pool image features described in step S200 includes:

[0112] S241. Contour extraction: Use a contour tracing algorithm to extract the contour of the molten pool.

[0113] S242. Morphological operations: Remove small noise points and fill internal holes.

[0114] In this specific embodiment, the feature extraction process is as Figure 7 shown.

[0115] During contour extraction: Use a contour tracing algorithm, such as the chain code method, the boundary tracking method, etc., to extract the contour of the molten pool. The purpose is to obtain the boundary information of the molten pool in the image. The extracted contour of the molten pool should be continuous without breaks, so as to facilitate subsequent shape analysis.

[0116] During morphological operations: Use morphological operations such as opening operation and closing operation to remove small noise points and fill internal holes. It aims at a series of image processing techniques based on shape, which are used to improve the image quality and highlight the features of interest in the image.

[0117] As a specific embodiment, the calculation of the molten pool area described in step S200 includes the pixel counting method and the contour integration method.

[0118] In this specific embodiment, when calculating the molten pool area, it is necessary to quantify the size of the target area in the image. Here, the pixel counting method or the contour integration method can be used to calculate the molten pool area. The former is to calculate the number of pixels belonging to the molten pool area and then multiply by the area of a single pixel to obtain the molten pool area. This method is a simple and direct method for calculating the area of the target area in the image and is usually used for binary images. The latter is to use a numerical integration method, such as Simpson's rule, etc., for the extracted contour of the molten pool to calculate the molten pool area. This method is a method for calculating the area enclosed by a closed contour using numerical integration methods and is suitable for more accurate area measurement.

[0119] The following gives a specific application of the method of the present invention in the laser welding between the electrode titanium mesh and its rib plate during the manufacture of the electrolytic cell.

[0120] During the operation of the electrolytic cell, the performance and stability of the electrode are crucial. At present, there are some problems in the welding of the electrolytic cell electrode, such as: the strength of the welded part is insufficient, and it is easy to fall off or be damaged during the electrolysis process, affecting the normal operation and service life of the electrolytic cell; the existing welding process may be difficult to ensure the uniformity and consistency of welding, resulting in uneven current distribution in the electrode during electrolysis and reducing the electrolysis efficiency.

[0121] In particular, in the current laser welding of electrolytic cell electrodes, especially when using a titanium mesh as the anode of the electrolytic cell and laser welding it with the supporting rib plate, it is a common problem that it is difficult to position the small holes in the titanium mesh. The high-precision positioning algorithm in the present invention can effectively adapt to its micro features and is particularly suitable for solving key problems such as positioning deviation, thermal damage, and oxidation pollution in the precision welding of the titanium mesh. In addition, in view of the thermal conductivity of titanium, a thermal input control strategy with a gradient real-time change can prevent damage to the titanium surface. Therefore, the method of the present invention is particularly suitable for the laser welding of the titanium mesh of the electrolytic cell electrode and its rib plate.

[0122] In the electrolytic cell electrode assembly, the connection between the titanium mesh with a porous structure and its rib plate needs to meet the requirements of electrical conductivity, corrosion resistance, and mechanical strength at the same time. However, the laser welding between the traditional electrolytic cell electrode titanium mesh and its supporting rib plate usually has the following problems:

[0123] One, insufficient positioning accuracy: Currently, the common pore diameters of the titanium mesh of the electrolytic cell electrode are several specifications such as 1X2 (mm), 2X4 (mm), 3X6 (mm), etc. Based on this small pore diameter property, the laser welding head often causes the solder joint to shift due to positioning errors (such as more than ±0.1 mm), and increases the virtual welding rate of poor micro-fusion, especially when the contact surface of the rib plate is a curved surface.

[0124] Two, out-of-control thermal input: Whether it is pure titanium or titanium alloy, its thermal conductivity is relatively low compared with other metals. This relatively low thermal conductivity makes pure titanium or titanium alloy have better heat insulation performance in some applications, but the heat accumulation during welding is very likely to cause its welding thermal damage, resulting in the peeling of the surface layer, and is prone to generate pores and thermal cracks.

[0125] Three, high-temperature oxidation: Titanium absorbs oxygen rapidly above 400 °C and easily reacts with oxygen to form a brittle oxide layer (TiO2). If the protective gas coverage is insufficient, it is easy to cause embrittlement of the weld. In order to make the protective gas better protect the oxidation of the welding surface while ensuring that the protective gas will not cause unnecessary waste, in the present invention, it is very necessary to increase the supply of the protective gas through the dynamic adjustment of the protective gas supply, especially in the high-temperature state of the welding surface.

[0126] Applying the method of the present invention to the welding between the titanium mesh of the electrolytic cell electrode and its rib plate can improve the strength and stability of the welding part, reduce the risk of the electrode titanium mesh falling off or being damaged during the electrolysis process; by reasonably and precisely controlling the input of welding energy, ensuring the uniformity and consistency of welding, making the current distribution on the anode more uniform during the electrolysis process, and improving the electrolysis efficiency; due to the improvement of welding quality, the stability of the electrode titanium mesh is enhanced, thereby extending the service life of the electrolytic cell and reducing the maintenance cost.

[0127] As a second embodiment, the present invention provides a laser welding device suitable for welding a special-shaped contact surface between a workpiece having a porous structure and a connecting piece thereof, the device comprising:

[0128] M100, laser welding fixture: used to fix the porous structure workpiece to be welded and its connecting parts, made of high temperature resistant and high strength materials, with a structure with adjustable clamping size;

[0129] M200, laser welding unit: including a laser generator, a coaxial vision system and a laser output mirror set, the laser output power of the laser generator can be adjusted;

[0130] M300, imaging monitoring unit: used for imaging the molten pool of the welding spot between the porous structure workpiece and its connecting part;

[0131] M400, gas supply unit: used to provide protective gas to the welding area, and its gas output can be adjusted;

[0132] M500, temperature measurement unit: used to measure the molten pool temperature in real time;

[0133] M600, central control unit: controls and implements the laser welding method as described in any one of claims 1-9.

[0134] Finally, it should be noted that although the present invention is exemplarily described through specific implementation methods, it does not constitute a limitation on the scope of patent protection of the present invention. Technical personnel in the relevant technical field should understand that various equivalent substitutions and optimization improvements can still be made to the specific embodiments of the present invention, and any substitutions and improvements that do not deviate from the spirit of the present invention should be covered by the scope of patent protection of the present invention.

Claims

1. A laser welding method applicable to welding of a special-shaped contact surface between a workpiece with a porous structure and its connecting member, characterized in that The method includes: S100. Place the workpiece with a porous structure and its connecting piece in a welding fixture, obtain the spatial coordinates of the welding contact surface between the two, and generate the NURBS curve equation of the welding path; S200. Collect the molten pool image in real time and calculate the area of the molten pool. The process includes: collection of the molten pool image, preprocessing of the molten pool image, segmentation of the molten pool image, extraction of the features of the molten pool image, and calculation of the area of the molten pool; S300. Dynamically adjust the real-time power of the laser according to the change of the area of the molten pool. The real-time power decreases as the area of the molten pool becomes larger and finally remains at a constant power value; S400. Protect the welding area with a local air curtain and dynamically adjust the size of the flow rate of the shielding gas. When the temperature of the molten pool rises, the flow rate of the shielding gas increases based on the reference flow rate, and when the temperature of the molten pool drops, the flow rate of the shielding gas decreases based on the reference flow rate.

2. The laser welding method according to claim 1, wherein The real-time power described in step S300 satisfies the calculation formula In this formula, P real is the real-time power of the laser, P1 is the preset reference power, and its value range is set to 300 - 800 W. α1 is the gain coefficient, and its value range is set to 0.1 - 2.

0. S real is the real-time molten pool area, and S target is the target molten pool area.

3. The laser welding method according to claim 1, characterized in that, The real-time power described in step S300 satisfies the calculation formula In this formula, P real is the real-time power of the laser, P2 is the preset reference power, and its value range is set to 300 - 800 W. α2 is the gain coefficient, and its value range is set to 0 - 2.

0. S real is the real-time molten pool area, and S target is the target molten pool area.

4. The laser welding method according to claim 1, characterized in that, The shielding gas flow rate described in step S400 satisfies the calculation formula In this formula, Q real is the magnitude of the real-time gas flow rate, Q1 is the preset reference flow rate, set to 15 - 25 L / min, β1 is the temperature change response factor, and its value range is set to 0.5 - 2.0, is the rate of change of the molten pool temperature.

5. The laser welding method according to claim 1, characterized in that, The shielding gas flow rate described in step S400 satisfies the calculation formula In this formula, Q real is the magnitude of the real-time gas flow rate, Q2 is the preset reference flow rate, set to 15 - 25 L / min, β2 is the temperature change response factor, and its value range is set to 0.1 - 1.0, is the rate of change of the molten pool temperature.

6. The laser welding method according to claim 1, characterized in that, In step S200, the preprocessing of the molten pool image described includes: S221. Gray conversion: Convert the color image into a grayscale image; S222. Image enhancement: Improve the image quality by methods such as contrast enhancement or histogram equalization; S223. Noise elimination: Remove the image noise using median filtering or Gaussian filtering.

7. The laser welding method according to claim 1, characterized in that, In step S200, the segmentation of the molten pool image described includes: S231. Threshold segmentation: Separate the molten pool from the background color; S232. Edge detection: Find out the contour of the molten pool; S233. Region growing: Merge adjacent pixels and form the molten pool region.

8. The laser welding method according to claim 1, characterized in that, In step S200, the extraction of the features of the molten pool image described includes: S241. Contour extraction: Extract the contour of the molten pool using a contour tracing algorithm; S242. Morphological operation: Remove small noise points and fill internal holes.

9. The laser welding method according to claim 1, wherein In step S200, the calculation of the area of the molten pool described includes the pixel counting method and the contour integration method.

10. A laser welding device is applicable to welding the dissimilar-shaped contact surfaces between a workpiece with a porous structure and its connecting member, and is characterized in that The device includes: M100. Laser welding fixture: Used to fix the workpiece with a porous structure to be welded and its connecting piece, made of high-temperature resistant and high-strength materials, and having a structure with adjustable clamping dimensions; M200. Laser welding unit: Includes a laser generator, a coaxial vision system, and a laser output mirror group. The laser output power of the laser generator can be adjusted in size; M300. Imaging monitoring unit: Used for imaging the solder joint molten pool between the workpiece with a porous structure and its connecting piece; M400. Gas supply unit: Used to supply shielding gas to the welding area, and its gas output can be adjusted in size; M500. Temperature measurement unit: Used to measure the temperature of the molten pool in real time; M600. Central control unit: Controls and implements the laser welding method described in any one of claims 1-9.

Images