Welding control method for copper and aluminum base materials

Through weld image processing and dynamic parameter adjustment, the problem of low repeatability of parameter settings in existing copper and aluminum welding technologies is solved, and the precise control of welding parameters and the improvement of welding quality is achieved.

CN119016820AActive Publication Date: 2024-11-26SHAOGUAN SKYMEN ULTRASONIC TECH LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411007652.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-11-26
Estimated Expiration
2044-07-25

AI Technical Summary

Technical Problem

The existing copper and aluminum welding technologies have low repeatability in parameter settings, making it difficult to achieve precise control, affecting the quality of welding products.

Method used

Weld image acquisition and processing, dynamic welding control and weld quality inspection and feedback methods are adopted to optimize welding parameters in real time through edge detection, feature extraction and parameter adjustment strategies, ensuring the accuracy and controllability of the welding process sex.

Benefits of technology

It improves the precise control of welding parameters, reduces material waste, enhances welding quality and reliability, reduces the chance of defects and failures, and realizes comprehensive monitoring and optimization of the welding environment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119016820A_ABST
    Figure CN119016820A_ABST
Patent Text Reader

Abstract

The invention discloses a copper and aluminum base material welding control method, and relates to the technical field of welding. A copper and aluminum base material welding control method comprises the steps of weld joint image acquisition and processing, dynamic welding control and weld joint quality inspection and feedback. According to the method, through accurate volume feature extraction of the three-dimensional model of the welding seam, the corresponding amount of brazing materials can be accurately output, material waste can be reduced, and it is ensured that the welding process is economical and efficient; initial welding parameters are determined according to the volume characteristic value of the welding seam and the using amount of the brazing material, so that the welding process can be more accurately adjusted according to specific welding parts and requirements, the welding quality is improved, and the welding process is more controllable; the welding parameters are adjusted in real time in the welding process, the most comprehensive and optimal welding parameter adjustment strategy can be obtained through maximum pooling of the first parameter adjustment strategy and the second parameter adjustment strategy, and comprehensive monitoring and optimization of the welding environment are achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of welding, and in particular to a welding control method for copper and aluminum substrates. Background Art

[0002] Copper and aluminum welding technology is a technology widely used in electrical engineering, refrigeration, air conditioning and other fields that require metal bonding. In the field of electrical engineering, the most commonly used copper and aluminum welding technology is brazing technology, especially soft soldering using silver-based solder. This method is widely used due to its excellent electrical conductivity, good mechanical strength and low processing temperature.

[0003] Existing welding technology includes numerous parameters, such as current, voltage, welding speed, flow rate of shielding gas, etc. However, even for professional technicians, there may be slight deviations when manually setting the same parameters, resulting in reduced repeatability of the welding process and affecting the quality of the welded products. In addition, during the welding process, the welding parameters need to be adjusted according to the degree of change of the brazing material and the base material. If only manual adjustment of the parameters is relied upon, it is difficult to achieve precise control of the welding.

[0004] Therefore, we urgently need to provide a copper and aluminum substrate welding control method that can accurately control and adaptively adjust welding parameters. Summary of the invention

[0005] The present invention aims to provide a copper and aluminum substrate welding control method which can accurately control and adaptively adjust welding parameters, thereby improving the accurate control of welding parameters and thus improving the quality of welding products.

[0006] A copper and aluminum substrate welding control method, comprising: S1. Weld image acquisition and processing Acquire weld seam images of copper substrate and aluminum substrate to obtain images of copper weld seam to be processed and images of aluminum weld seam to be processed; perform edge detection on the images of copper weld seam to be processed and images of aluminum weld seam to be processed to obtain images of copper weld seam edge and images of aluminum weld seam edge; determine the amount of brazing material and initial welding parameters based on the images of copper weld seam edge, aluminum weld seam edge and brazing material selection model; perform welding of copper and aluminum substrates according to the initial welding parameters; welding parameters include laser power, focal position, laser beam diameter, shielding gas flow, welding speed and welding path; S2. Dynamic welding control The welding environment temperature and the weld surface image are obtained at every time interval f; the welding environment temperature is compared with the temperature warning range, and if the welding environment temperature reaches the temperature warning range, a first parameter adjustment strategy is obtained; otherwise, welding is continued; feature extraction is performed based on the weld surface image to obtain a brazing contact angle characteristic value, and the corresponding wettability is matched based on the brazing contact angle characteristic value. If the wettability is lower than the set wettability, the corresponding second parameter adjustment strategy is output in the wettability enhancement library based on the wettability; the first parameter adjustment strategy and the second parameter adjustment strategy are subjected to maximum pooling to obtain a welding parameter adjustment strategy, and the welding process is controlled based on the welding parameter adjustment strategy; the first parameter adjustment strategy and the second parameter adjustment strategy include adjustments to various welding parameters; S3. Weld quality inspection and feedback Obtain an external image of the weld joint and an internal image of the weld joint; perform quality inspection on the weld joint based on the external image of the weld joint, the internal image of the weld joint and the weld joint quality inspection model to obtain a quality inspection result of the weld joint; if the quality inspection result of the weld joint is defect-free, record the actual amount of brazing material and a welding parameter adjustment table, and optimize the brazing material selection model based on the actual amount of brazing material and the welding parameter adjustment table; otherwise, determine the welding problem based on the external image of the weld joint, the internal image of the weld joint and the weld joint quality inspection result, and adjust the welding control method based on the welding problem.

[0007] As a preferred technical solution of the present invention, the brazing material selection model includes a feature recognition layer, a material layer, a parameter output layer and a result output layer; The feature recognition layer is used to construct a three-dimensional model of the weld using stereo vision technology for the edge images of the copper weld and the edge images of the aluminum weld; the volume feature of the three-dimensional model of the weld is extracted to obtain the volume feature value of the weld; The material layer is used to output the corresponding brazing material amount according to the weld volume characteristic value; The parameter output layer is used to match the amount of brazing materials from the brazing material standard library to obtain the welding standard parameters corresponding to the amount of brazing materials with the highest similarity, which are the initial welding parameters; the brazing material standard library contains the amount of brazing materials and the corresponding welding standard parameters; The result output layer is used to output the brazing material consumption and initial welding parameters.

[0008] As a preferred technical solution of the present invention, the specific steps of constructing a brazing material selection model include: Acquire a plurality of weld edge images with known brazing material dosage, each weld edge image including a copper weld edge image and an aluminum weld edge image; combine the plurality of weld edge images to obtain a brazing material selection training set; The CNN model is selected as the initial model of the brazing material selection model, the brazing material selection training set is input into the brazing material selection initial model for model training with the brazing material dosage, and the loss function value is output; the accuracy of the loss function value and the true value is calculated to obtain the accuracy evaluation result; if the accuracy evaluation result is within the error standard range, the trained brazing material selection initial model is output, that is, the brazing material selection model; otherwise, the brazing material selection training set is used to continue model training.

[0009] As a preferred technical solution of the present invention, the specific steps of step S2 include: S21. Obtain the welding environment temperature and weld surface image at each time interval f, compare the welding environment temperature with the temperature warning range, and if the welding environment temperature reaches the temperature warning range, proceed to step D2; otherwise, continue welding; S22. Calculate the difference between the welding environment temperature and the welding standard temperature to obtain the welding temperature difference ΔT; input the welding temperature difference ΔT into the welding thermodynamic model for digital twin simulation to obtain a first adjustment strategy; the welding standard temperature is the optimal temperature for welding; S23. Perform enhanced feature extraction on the weld surface image to obtain a brazing contact angle characteristic value; perform wettability matching based on the brazing contact angle characteristic value to obtain a wettability; if the wettability is lower than a set wettability, output a corresponding second parameter adjustment strategy in a wettability enhancement library based on the wettability; S24. Performing maximum pooling processing on the first parameter adjustment strategy and the second parameter adjustment strategy, and selecting the welding parameter adjustment strategy with the largest adjustment value for each parameter; S25. Perform real-time control of the welding instrument based on the welding parameter adjustment strategy.

[0010] As a preferred technical solution of the present invention, the specific steps of performing enhanced feature extraction on the weld surface image in step S23 include: Denoising the weld surface image to obtain a denoised weld surface image; Grayscale the denoised weld surface image and enhance the edge to obtain an enhanced edge weld surface image; Perform contour extraction on the surface image of the reinforced edge weld to obtain a brazing contact angle contour image; The characteristic value of the brazing contact angle is obtained by performing feature analysis on the brazing contact angle profile image.

[0011] As a preferred technical solution of the present invention, the specific steps of quality inspection of the weld joint in step S3 include: T1. Obtaining external image and internal image of weld joint; T2. Input the external image of the weld joint and the internal image of the weld joint into the weld joint quality inspection model for quality inspection to obtain the weld joint quality inspection result; T3. If the weld joint quality inspection result is qualified, record the actual amount of brazing material and welding parameter adjustment table and proceed to step T4; otherwise, proceed to step T5; T4. Take the average value of each parameter in the welding parameter adjustment table to obtain the welding standard parameters, and input the actual amount of brazing material and the corresponding welding standard parameters into the brazing material standard library; T5. Based on the weld joint quality inspection results output by the weld joint quality inspection model, determine the welding problems and adjust the welding control method according to the welding problems.

[0012] As a preferred technical solution of the present invention, the weld joint quality inspection model includes an image classification layer, an external image detection layer, an internal image detection layer and a result output layer; The image classification layer is used to classify the input image, input the external image of the weld joint into the external image detection layer, and input the internal image of the weld joint into the internal image detection layer; The external image detection layer is used to perform defect detection on the external image of the weld joint and output the external defect detection results; The internal image detection layer is used to perform defect detection on the internal image of the weld joint and output the internal defect detection results; The result output layer is used to output the weld joint quality inspection results. If both the external defect detection results and the internal defect detection results are defect-free, the weld joint quality inspection result is qualified; otherwise, the weld joint quality inspection result is unqualified, and the weld joint quality inspection reason is output.

[0013] As a preferred technical solution of the present invention, the specific steps of performing defect detection on the external image and the internal image of the weld joint include: In the external image detection layer, a weld joint external quality inspection model is deployed; the weld joint external image is input into the weld joint external quality inspection model for image analysis to obtain the external defect detection result; The YOLOv5 model is used as the initial model of the weld joint external quality inspection model, and m external weld joint defect images and external weld joint defect normal images are obtained, and the weld joint defect images are classified and labeled, and the external weld joint defect normal images are labeled as defect-free; the data enhancement method is used for q external weld joint normal images to obtain q new external weld joint defect images; the data enhancement method is used for r external weld joint defect images to obtain p new external weld joint defect images; the p+q new external weld joint defect images are combined with the m external weld joint defect images and the external weld joint defect normal images to obtain the external weld joint defect training set, and the external weld joint defect training set is used to train the YOLOv5 model to obtain the weld joint external quality inspection model; In the internal image detection layer, a weld joint internal quality inspection model is deployed to obtain the weld joint internal image, perform image annotation processing on the weld joint internal image, and extract the image in the annotation box to obtain the weld joint internal image feature image; the image annotation target is the darker or brighter area in the weld joint internal image; the weld joint internal image feature image is input into the weld joint internal quality inspection model for image analysis to obtain the internal defect detection result; The YOLOv5 model is used as the initial model of the weld joint internal quality inspection model, g weld joint internal image feature images are obtained, and the internal weld joint internal image feature images are classified and labeled for defects. The feature enhancement method is used for the g weld joint internal image feature images to obtain h new weld joint internal image feature images. The h new weld joint internal image feature images and the g weld joint internal image feature images are combined to obtain the internal weld joint defect training set. The internal weld joint defect training set is used to train the YOLOv5 model to obtain the weld joint internal quality inspection model.

[0014] The present invention has the following advantages: 1. The present invention extracts the precise volume features of the weld stereo model, and the material layer can accurately output the corresponding brazing material dosage. This accuracy helps reduce material waste and ensures that the welding process is economical and efficient. The parameter output layer determines the initial welding parameters according to the weld volume characteristic value and the brazing material dosage, so that the welding process can be more accurately adjusted according to the specific welding position and needs, which not only improves the welding quality, but also makes the welding process more controllable. Through precise material selection and parameter setting, the quality of the weld joint can be improved, the probability of defects and failures can be reduced, and the reliability of welding can be improved. By adjusting the welding parameters in real time according to the ambient temperature and the brazing contact angle characteristic value, it is helpful to modify the welding strategy when the temperature reaches the warning range, and reduce the potential damage of excessive ambient temperature to the equipment and welding results. By adjusting the parameters to achieve a preset wetting degree, the welding quality can be improved. Through the maximum pooling of the first parameter adjustment strategy and the second parameter adjustment strategy, the most comprehensive and optimal welding parameter adjustment strategy can be obtained, and the comprehensive monitoring and optimization of the welding environment can be achieved.

[0015] 2. The present invention can immediately discover possible problems in the welding process by performing quality inspection on the external and internal images of the weld joint, so as to correct them in time. By learning from errors and failed experiences, the welding quality can be continuously improved. The actual amount of brazing material used and the welding parameter adjustment table can be recorded. These data can drive the algorithm to optimize the brazing material selection model. The welding parameters that have an important impact on the welding quality can be fine-tuned with the help of these data, thereby optimizing the welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The present invention is a schematic flow chart of a copper and aluminum substrate welding control method used in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Example

[0018] The brazing technology is usually used for copper and aluminum substrate welding. The brazing material is usually silver-based brazing filler metal. Taking the brazing control method under the brazing technology using silver-based brazing filler metal as raw material as an example, a copper and aluminum substrate welding control method is shown in Figure 1. Figure 1 As shown, including: S1. Weld image acquisition and processing Acquire weld seam images of copper substrate and aluminum substrate to obtain images of copper weld seam to be processed and images of aluminum weld seam to be processed; perform edge detection on the images of copper weld seam to be processed and images of aluminum weld seam to be processed to obtain images of copper weld seam edge and images of aluminum weld seam edge; determine the amount of brazing material and initial welding parameters based on the images of copper weld seam edge, aluminum weld seam edge and brazing material selection model; perform welding of copper and aluminum substrates according to the initial welding parameters; welding parameters include laser power, focal position, laser beam diameter, shielding gas flow, welding speed and welding path; A camera is installed in the welding furnace to capture the welding process. The edge detection algorithm is the Canny edge detection algorithm. The Canny edge detection algorithm first uses a Gaussian filter to reduce image noise, then uses a multi-directional gradient operator to determine the strength and direction of the edge, and finally applies non-maximum suppression and hysteresis threshold technology to determine the position of the edge. The use of the Canny edge detection algorithm can effectively eliminate the influence of other irrelevant factors in the welding furnace on image recognition. Silver-based solder is welded by laser welding. Laser welding can control the size, shape and energy of the solder joint very accurately, which makes it ideal for processing small or precision parts. Laser welding can achieve targeted local heating without causing excess heat to the surrounding copper and aluminum substrates, thereby protecting the original properties of the welding substrate. The initial welding parameters are set for laser welding. The output power of the laser directly affects the welding speed and penetration depth. The focal position of the laser on the surface of the material has a significant impact on the quality and penetration of the weld. The speed of the laser head movement is usually measured in mm / s. The welding speed needs to be reasonably set according to the melting point and thermal conductivity of the material to obtain the ideal welding effect. In the laser welding process, inert gas is usually used to protect the welding area, prevent oxidation and assist heat transfer. The flow rate of the shielding gas is usually measured in L / min, and the specific value depends on the size and shape of the weld. For complex welding tasks, the planning of the movement of the laser welding head along a specific path is also one of the key parameters. Penetration depth refers to the melting depth of the welded metal caused by the welding heat source during the welding process. It is an important indicator to measure the strength of welding penetration. The brazing material selection model includes a feature recognition layer, a material layer, a parameter output layer and a result output layer; The feature recognition layer is used to construct a three-dimensional model of the weld using stereo vision technology for the edge images of the copper weld and the edge images of the aluminum weld; the volume feature of the three-dimensional model of the weld is extracted to obtain the volume feature value of the weld; The material layer is used to output the corresponding brazing material amount according to the weld volume characteristic value; The parameter output layer is used to match the amount of brazing materials from the brazing material standard library to obtain the welding standard parameters corresponding to the amount of brazing materials with the highest similarity, which are the initial welding parameters; the brazing material standard library contains the amount of brazing materials and the corresponding welding standard parameters; The result output layer is used to output the amount of brazing material and initial welding parameters; In the brazing process, using too much brazing material may cause accumulation and waste, while too little may cause loose joints or holes. By extracting the precise volume features of the weld three-dimensional model, the material layer can accurately output the corresponding amount of brazing material. This accuracy helps reduce material waste and ensures that the welding process is economical and efficient. The parameter output layer determines the initial welding parameters based on the weld volume feature value and the amount of brazing material, so that the welding process can be adjusted more accurately according to the specific welding location and needs, which not only improves the welding quality, but also makes the welding process more controllable. The brazing material standard library is established by professional technicians based on the actual usage of historical brazing materials and the welding parameter adjustment table; The specific steps to build a brazing material selection model include: Acquire a plurality of weld edge images with known brazing material dosage, each weld edge image including a copper weld edge image and an aluminum weld edge image; combine the plurality of weld edge images to obtain a brazing material selection training set; The CNN model is selected as the initial model of the brazing material selection model, and the brazing material selection training set is input into the brazing material selection initial model for model training with the brazing material dosage, and the loss function value is output; the accuracy of the loss function value and the true value is calculated to obtain the accuracy evaluation result; if the accuracy evaluation result is within the error standard range, the trained brazing material selection initial model is output, that is, the brazing material selection model; otherwise, the brazing material selection training set is used to continue model training; Through accurate material selection and parameter setting, the quality of weld joints can be improved, the probability of defects and failures can be reduced, and the reliability of welding can be improved; through accurate material usage evaluation, the waste of brazing materials can be reduced, and production costs can be reduced. At the same time, high-quality welding reduces rework and scrap generation; the consistency of model output helps to standardize the production process, which is the key to improving the consistency of weld joints and reducing operational complexity; S2. Dynamic welding control The welding environment temperature and the weld surface image are obtained at every time interval f; the welding environment temperature is compared with the temperature warning range, and if the welding environment temperature reaches the temperature warning range, a first parameter adjustment strategy is obtained; otherwise, welding is continued; feature extraction is performed based on the weld surface image to obtain a brazing contact angle characteristic value, and the corresponding wettability is matched based on the brazing contact angle characteristic value. If the wettability is lower than the set wettability, the corresponding second parameter adjustment strategy is output in the wettability enhancement library based on the wettability; the first parameter adjustment strategy and the second parameter adjustment strategy are subjected to maximum pooling to obtain a welding parameter adjustment strategy, and the welding process is controlled based on the welding parameter adjustment strategy; the first parameter adjustment strategy and the second parameter adjustment strategy include adjustments to various welding parameters; The time interval f is set by professionals based on actual conditions; The welding environment temperature is obtained by the temperature sensor, and the temperature warning range is set based on the welding threshold temperature. The welding threshold temperature is set by the melting point of the base material and the brazing material. The environment temperature must be lower than the melting point of all base materials. Otherwise, if the environment temperature is higher than the melting point of the base material, the base material will be melted to generate oxides. The oxides attached to the weld will affect the strength of the weld joint, and too high an environment temperature will affect the quality of the weld joint. The temperature warning range is set by professional technicians based on actual conditions. The specific steps of step S2 include: S21. Obtain the welding environment temperature and weld surface image at each time interval f, compare the welding environment temperature with the temperature warning range, and if the welding environment temperature reaches the temperature warning range, proceed to step D2; otherwise, continue welding; S22. Calculate the difference between the welding environment temperature and the welding standard temperature to obtain the welding temperature difference ΔT; input the welding temperature difference ΔT into the welding thermodynamic model for digital twin simulation to obtain the first adjustment strategy; the welding standard temperature is the optimal temperature for welding; the first adjustment strategy includes adjustments to various welding parameters, such as changing the laser power, changing the welding speed, and increasing the shielding gas flow rate; The welding thermodynamic model is set by professional technicians. The welding thermodynamic model can describe the relationship between parameters such as temperature, material properties, heat transfer, heat consumption and time. The welding temperature difference ΔT is input into the welding thermodynamic model for digital twin simulation. The welding temperature difference ΔT is infinitely close to 0 for calculation, and the first adjustment strategy is obtained. S23. Perform enhanced feature extraction on the weld surface image to obtain a brazing contact angle characteristic value; perform wettability matching based on the brazing contact angle characteristic value to obtain a wettability; if the wettability is lower than a set wettability, output a corresponding second parameter adjustment strategy in a wettability enhancement library based on the wettability; The wetting degree enhancement library contains the current wetting degree and the corresponding second parameter adjustment strategy for enhancing the wetting degree. The second parameter adjustment strategy is set by professional technicians based on multiple welding experiments. The second parameter adjustment strategy includes adjustments to various welding parameters, such as changing the focal position, changing the diameter of the laser beam, etc. The wetting degree setting is set by professional technicians based on actual experience. The contact angle is the physical angle formed by the liquid droplet contacting the solid surface, specifically the angle between the edge of the liquid droplet formed on the solid surface and the solid surface; a small contact angle means that the liquid wets the solid better, while a large contact angle means poor wettability. In welding, if the contact angle of the solder on the substrate is small, then the wettability is good, otherwise the contact angle is large and the wettability is poor; for liquid solder, a high wettability means that it can diffuse and spread better on the substrate surface to form a uniform weld; S24. Performing maximum pooling processing on the first parameter adjustment strategy and the second parameter adjustment strategy, and selecting the welding parameter adjustment strategy with the largest adjustment value for each parameter; S25. Real-time control of welding equipment based on welding parameter adjustment strategy; The specific steps of performing enhanced feature extraction on the weld surface image in step S23 include: De-noising the weld surface image to obtain a de-noised weld surface image; the Gaussian filter algorithm can be used to de-noise the weld surface image. De-noising is to reduce irrelevant information or background noise in the image, which can help to see the true shape of the weld more clearly; The denoised weld surface image is grayed and the edges are enhanced to obtain an enhanced edge weld surface image; commonly used graying methods include averaging method, weighting method, etc.; the weld area can then be highlighted by enhancing the edge of the image; the color image is converted to a grayscale image to reduce the computational complexity, while the grayscale image retains most of the important information reflecting the weld edge; the contrast between the weld and the background is enhanced, making the recognition of the weld edge more obvious; The contour of the reinforced edge weld surface image is extracted to obtain the brazing contact angle contour image; the findContours function in the OpenCV library can be used to obtain the edge contour of the image. The extracted contour can be used to generate an accurate morphological representation of the weld, which is the basis for calculating the weld eigenvalue; Performing feature analysis on the brazing contact angle profile image to obtain a characteristic value of the brazing contact angle; fitting the contact angle by a fitting method, such as a least square method, and calculating the characteristic value of the brazing contact angle; S3. Weld quality inspection and feedback Obtain an external image of the weld joint and an internal image of the weld joint; perform quality inspection on the weld joint based on the external image of the weld joint, the internal image of the weld joint and the weld joint quality inspection model to obtain a quality inspection result of the weld joint; if the quality inspection result of the weld joint is defect-free, record the actual amount of brazing material and the welding parameter adjustment table, and optimize the brazing material selection model based on the actual amount of brazing material and the welding parameter adjustment table; otherwise, determine the welding problem based on the external image of the weld joint, the internal image of the weld joint and the weld joint quality inspection result, and adjust the welding control method based on the welding problem; The weld joint is the area where the copper and aluminum substrates are connected through the welding process. By performing quality inspections on the external and internal images of the weld joint, problems that may arise during the welding process can be immediately discovered and corrected in a timely manner. By learning from errors and failures, the welding quality can be continuously improved. The actual amount of brazing material used and the welding parameter adjustment table are recorded. These data can drive the algorithm to optimize the brazing material selection model. The welding parameters that have an important impact on the welding quality can be fine-tuned with the help of these data to optimize the welding process. The specific steps of quality inspection of the weld joint in step S3 include: T1. Obtain the external image and internal image of the weld joint; the external image of the weld joint is taken by a high-definition camera, and the internal image of the weld joint is obtained by X-ray imaging; T2. Input the external image of the weld joint and the internal image of the weld joint into the weld joint quality inspection model for quality inspection to obtain the weld joint quality inspection result; T3. If the weld joint quality inspection result is qualified, the actual amount of brazing material and the welding parameter adjustment table are recorded, and the process proceeds to step T4; otherwise, the process proceeds to step T5; the welding parameter adjustment table contains a record of each adjustment of the welding parameters during the welding process; T4. Take the average value of each parameter in the welding parameter adjustment table to obtain the welding standard parameters, and input the actual amount of brazing material and the corresponding welding standard parameters into the brazing material standard library; T5. Based on the weld joint quality inspection results output by the weld joint quality inspection model, the welding problems are judged and the welding control method is adjusted according to the welding problems. The weld joint quality inspection results include the causes of weld defects. According to the causes of weld defects, professional technicians can judge the welding problems and adjust the welding control method accordingly. Common causes of weld joint defects include porosity, slag inclusion, incomplete penetration, lack of fusion, cracks, cold seams, heat-affected zone hardening, burn-through, slag splashing, sun scars and walkways; The weld joint quality inspection model includes an image classification layer, an external image detection layer, an internal image detection layer and a result output layer; The image classification layer is used to classify the input image, input the external image of the weld joint into the external image detection layer, and input the internal image of the weld joint into the internal image detection layer; The external image detection layer is used to perform defect detection on the external image of the weld joint and output the external defect detection results; The internal image detection layer is used to perform defect detection on the internal image of the weld joint and output the internal defect detection results; The result output layer is used to output the weld joint quality inspection results. If both the external defect detection results and the internal defect detection results are defect-free, the weld joint quality inspection result is qualified; otherwise, the weld joint quality inspection result is unqualified, and the weld joint quality inspection reason is output; The specific steps of defect detection on the external image and the internal image of the weld joint include: In the external image detection layer, a weld joint external quality inspection model is deployed; the weld joint external image is input into the weld joint external quality inspection model for image analysis to obtain the external defect detection result; The YOLOv5 model is used as the initial model of the weld joint external quality inspection model, and m external weld joint defect images and external weld joint defect normal images are obtained, and the weld joint defect images are classified and labeled, and the external weld joint defect normal images are labeled as defect-free; the data enhancement method is used for q external weld joint normal images to obtain q new external weld joint defect images; the data enhancement method is used for r external weld joint defect images to obtain p new external weld joint defect images; the p+q new external weld joint defect images are combined with the m external weld joint defect images and the external weld joint defect normal images to obtain the external weld joint defect training set, and the external weld joint defect training set is used to train the YOLOv5 model to obtain the weld joint external quality inspection model; In the internal image detection layer, a weld joint internal quality inspection model is deployed to obtain the weld joint internal image, perform image annotation processing on the weld joint internal image, and extract the image in the annotation box to obtain the weld joint internal image feature image; the image annotation target is the darker or brighter area in the weld joint internal image; the weld joint internal image feature image is input into the weld joint internal quality inspection model for image analysis to obtain the internal defect detection result; The YOLOv5 model is used as the initial model of the weld joint internal quality inspection model, g weld joint internal image feature images are obtained, and the internal weld joint internal image feature images are labeled for defect classification, and the g weld joint internal image feature images are subjected to feature enhancement method to obtain h new weld joint internal image feature images; the h new weld joint internal image feature images and the g weld joint internal image feature images are combined to obtain an internal weld joint defect training set, and the internal weld joint defect training set is used to train the YOLOv5 model to obtain a weld joint internal quality inspection model; In the target detection model, a large number of image samples are required for model training, but the number of normal images is much larger than the number of defective images, so it is necessary to use data enhancement methods to amplify the collected data set to increase the diversity of weld defects; in the external quality inspection model of weld joints, the Copy-Pasting data enhancement method and the Mosaic+Mixup data enhancement method can be used to increase the external weld joint defect images; the Copy-Pasting data enhancement method is to paste the defective part with the annotation box into the normal image to artificially create the defective image; the Mosaic+Mixup data enhancement method is to first use the Mosaic data enhancement method to segment and reorganize a defective image and three additional defective images to obtain a new defective image, and the Mixup data enhancement method is to add the two images output by the Mosaic data enhancement method in proportion to obtain a new defective image; in the internal quality inspection model of weld joints, contrast enhancement algorithms, edge enhancement algorithms, denoising algorithms and sharpening algorithms can be used to enhance the feature details of the internal image feature images of weld joints; different data enhancement methods have greatly increased the number of weld defect samples in the training set and improved the accuracy of target detection model training.

[0019] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A copper and aluminum substrate welding control method, characterized in that: include: S1. Weld image acquisition and processing Acquire weld seam images of the copper substrate and the aluminum substrate to obtain a copper weld seam image to be processed and an aluminum weld seam image to be processed; Perform edge detection on the copper weld image to be processed and the aluminum weld image to be processed to obtain the copper weld edge image and the aluminum weld edge image; determine the amount of brazing material and initial welding parameters based on the copper weld edge image, the aluminum weld edge image and the brazing material selection model; perform copper and aluminum substrate welding according to the initial welding parameters; the welding parameters include laser power, focus position, laser beam diameter, shielding gas flow, welding speed and welding path; S2. Dynamic welding control The welding environment temperature and the weld surface image are obtained at every time interval f; the welding environment temperature is compared with the temperature warning range, and if the welding environment temperature reaches the temperature warning range, a first parameter adjustment strategy is obtained; otherwise, welding is continued; feature extraction is performed based on the weld surface image to obtain a brazing contact angle characteristic value, and the corresponding wettability is matched based on the brazing contact angle characteristic value. If the wettability is lower than the set wettability, the corresponding second parameter adjustment strategy is output in the wettability enhancement library based on the wettability; the first parameter adjustment strategy and the second parameter adjustment strategy are subjected to maximum pooling to obtain a welding parameter adjustment strategy, and the welding process is controlled based on the welding parameter adjustment strategy; the first parameter adjustment strategy and the second parameter adjustment strategy include adjustments to various welding parameters; S3. Weld quality inspection and feedback Obtain an external image of the weld joint and an internal image of the weld joint; perform quality inspection on the weld joint based on the external image of the weld joint, the internal image of the weld joint and the weld joint quality inspection model to obtain a quality inspection result of the weld joint; if the quality inspection result of the weld joint is defect-free, record the actual amount of brazing material and a welding parameter adjustment table, and optimize the brazing material selection model based on the actual amount of brazing material and the welding parameter adjustment table; otherwise, determine the welding problem based on the external image of the weld joint, the internal image of the weld joint and the weld joint quality inspection result, and adjust the welding control method based on the welding problem.

2. A copper and aluminum substrate welding control method according to claim 1, characterized in that: The brazing material selection model includes a feature recognition layer, a material layer, a parameter output layer and a result output layer; The feature recognition layer is used to construct a three-dimensional model of the weld using stereo vision technology for the edge images of the copper weld and the edge images of the aluminum weld; the volume feature of the three-dimensional model of the weld is extracted to obtain the volume feature value of the weld; The material layer is used to output the corresponding brazing material amount according to the weld volume characteristic value; The parameter output layer is used to match the amount of brazing materials from the brazing material standard library to obtain the welding standard parameters corresponding to the amount of brazing materials with the highest similarity, which are the initial welding parameters; the brazing material standard library contains the amount of brazing materials and the corresponding welding standard parameters; The result output layer is used to output the brazing material consumption and initial welding parameters.

3. A copper and aluminum substrate welding control method according to claim 2, characterized in that: The specific steps to build a brazing material selection model include: Acquire a plurality of weld edge images with known brazing material dosage, each weld edge image including a copper weld edge image and an aluminum weld edge image; combine the plurality of weld edge images to obtain a brazing material selection training set; The CNN model is selected as the initial model of the brazing material selection model, the brazing material selection training set is input into the brazing material selection initial model for model training with the brazing material dosage, and the loss function value is output; the accuracy of the loss function value and the true value is calculated to obtain the accuracy evaluation result; if the accuracy evaluation result is within the error standard range, the trained brazing material selection initial model is output, that is, the brazing material selection model; otherwise, the brazing material selection training set is used to continue model training.

4. A copper and aluminum substrate welding control method according to claim 1, characterized in that: The specific steps of step S2 include: S21. Obtain the welding environment temperature and weld surface image at each time interval f, compare the welding environment temperature with the temperature warning range, and if the welding environment temperature reaches the temperature warning range, proceed to step D2; otherwise, continue welding; S22. Calculate the difference between the welding environment temperature and the welding standard temperature to obtain the welding temperature difference ΔT; input the welding temperature difference ΔT into the welding thermodynamic model for digital twin simulation to obtain a first adjustment strategy; the welding standard temperature is the optimal temperature for welding; S23. Perform enhanced feature extraction on the weld surface image to obtain a brazing contact angle characteristic value; perform wettability matching based on the brazing contact angle characteristic value to obtain a wettability; if the wettability is lower than a set wettability, output a corresponding second parameter adjustment strategy in a wettability enhancement library based on the wettability; S24. Performing maximum pooling processing on the first parameter adjustment strategy and the second parameter adjustment strategy, and selecting the welding parameter adjustment strategy with the largest adjustment value for each parameter; S25. Perform real-time control of the welding instrument based on the welding parameter adjustment strategy.

5. A copper and aluminum substrate welding control method according to claim 4, characterized in that: The specific steps of performing enhanced feature extraction on the weld surface image in step S23 include: Denoising the weld surface image to obtain a denoised weld surface image; Grayscale processing is performed on the denoised weld surface image and the edge is enhanced to obtain an edge-enhanced weld surface image; Perform contour extraction on the surface image of the reinforced edge weld to obtain a brazing contact angle contour image; The characteristic value of the brazing contact angle is obtained by performing feature analysis on the brazing contact angle profile image.

6. A copper and aluminum substrate welding control method according to claim 1, characterized in that: The specific steps of quality inspection of the weld joint in step S3 include: T1. Obtaining external image and internal image of weld joint; T2. Input the external image of the weld joint and the internal image of the weld joint into the weld joint quality inspection model for quality inspection to obtain the weld joint quality inspection result; T3. If the weld joint quality inspection result is qualified, record the actual amount of brazing material and welding parameter adjustment table and proceed to step T4; otherwise, proceed to step T5; T4. Take the average value of each parameter in the welding parameter adjustment table to obtain the welding standard parameters, and input the actual amount of brazing material and the corresponding welding standard parameters into the brazing material standard library; T5. Based on the weld joint quality inspection results output by the weld joint quality inspection model, determine the welding problems and adjust the welding control method according to the welding problems.

7. A copper and aluminum substrate welding control method according to claim 6, characterized in that: The weld joint quality inspection model includes an image classification layer, an external image detection layer, an internal image detection layer and a result output layer; The image classification layer is used to classify the input image, input the external image of the weld joint into the external image detection layer, and input the internal image of the weld joint into the internal image detection layer; The external image detection layer is used to perform defect detection on the external image of the weld joint and output the external defect detection results; The internal image detection layer is used to perform defect detection on the internal image of the weld joint and output the internal defect detection results; The result output layer is used to output the weld joint quality inspection results. If both the external defect detection results and the internal defect detection results are defect-free, the weld joint quality inspection result is qualified; otherwise, the weld joint quality inspection result is unqualified, and the weld joint quality inspection reason is output.

8. A copper and aluminum substrate welding control method according to claim 7, characterized in that: The specific steps of defect detection on the external image and the internal image of the weld joint include: In the external image detection layer, a weld joint external quality inspection model is deployed; the weld joint external image is input into the weld joint external quality inspection model for image analysis to obtain the external defect detection result; The YOLOv5 model is used as the initial model of the weld joint external quality inspection model, and m external weld joint defect images and external weld joint defect normal images are obtained, and the weld joint defect images are classified and labeled, and the external weld joint defect normal images are labeled as defect-free; the data enhancement method is used for q external weld joint normal images to obtain q new external weld joint defect images; the data enhancement method is used for r external weld joint defect images to obtain p new external weld joint defect images; the p+q new external weld joint defect images are combined with the m external weld joint defect images and the external weld joint defect normal images to obtain the external weld joint defect training set, and the external weld joint defect training set is used to train the YOLOv5 model to obtain the weld joint external quality inspection model; In the internal image detection layer, a weld joint internal quality inspection model is deployed to obtain the weld joint internal image, perform image annotation processing on the weld joint internal image, and extract the image in the annotation box to obtain the weld joint internal image feature image; the image annotation target is the darker or brighter area in the weld joint internal image; the weld joint internal image feature image is input into the weld joint internal quality inspection model for image analysis to obtain the internal defect detection result; The YOLOv5 model is used as the initial model of the weld joint internal quality inspection model, g weld joint internal image feature images are obtained, and the internal weld joint internal image feature images are classified and labeled for defects. The feature enhancement method is used for the g weld joint internal image feature images to obtain h new weld joint internal image feature images. The h new weld joint internal image feature images and the g weld joint internal image feature images are combined to obtain the internal weld joint defect training set. The internal weld joint defect training set is used to train the YOLOv5 model to obtain the weld joint internal quality inspection model.

Citation Information

Patent Citations

  • Method for enhancing corrosion resistance of aluminum alloy brazing joint

    CN112894046A

  • Laser welding method for aluminum alloy / nickel-based alloy or nickel dissimilar material

    CN114505577A

  • Plate welding quality real-time monitoring method based on digital twinning

    CN114789307A

  • Welding method based on digital modeling and flexible weld joint recognition

    CN116275742A

  • Temperature control method of welding system for copper and aluminum welding

    CN118357540A