A modular welding equipment and a welding quality evaluation method

Abstract

This invention relates to the field of steel bridge welding technology, and provides a modular welding equipment and a welding quality evaluation method. The modular welding equipment can meet various processing requirements, can be adjusted according to processing needs, is flexible and diverse, produces high welding quality, saves costs, has multiple functions, and reduces the investment in manpower and other equipment. The welding quality evaluation method includes the following steps: driving an industrial camera to take n weld seam images at equal intervals along a track through a slag cleaning module; calculating the noise intensity using the local variance estimation method, and using weighted nonlocal mean noise reduction filtering; calling preset operators according to priority to perform defect detection, and identifying cracks, porosity, and slag inclusions through different algorithms; calculating the average recognition probability of the images, if it is less than a preset threshold, the welding is qualified, and the equipment moves to the next area; if it is greater than or equal to the threshold, it is judged as unqualified, and the process is repeated after targeted defect cleaning until qualified; this method can improve the accuracy of welding quality evaluation.

Description

Technical Field

[0001] This invention relates to the field of steel bridge welding technology, and in particular to a modular welding equipment and a welding quality evaluation method. Background Technology

[0002] In the past two years, steel structures, especially steel-concrete composite or steel-concrete combined structures, have shown a clear trend of gradually replacing reinforced concrete in bridge construction. This is because they have advantages such as strong spanning capacity, light weight, short construction period, and strong seismic performance, and are particularly suitable for complex projects such as cross-sea (cross-river and cross-river) bridges, cross-highway and cross-railway bridges.

[0003] Currently, manual welding or welding carts are commonly used. However, the following problems exist during the welding process: 1. The welding quality of manual welding depends on the welder's skill level and experience, resulting in poor stability and slow welding speed, making it difficult to meet the needs of mass production; 2. Current welding carts can only meet single welding requirements. Other processes, such as temperature measurement and slag removal, still require manual operation, which is time-consuming, labor-intensive, and inefficient.

[0004] Therefore, to address the above problems, a modular welding equipment and welding quality evaluation method are proposed. Summary of the Invention

[0005] This invention addresses the shortcomings of existing technologies by developing a modular welding equipment and a welding quality evaluation method. This invention meets various processing requirements, can be adjusted according to processing needs, is flexible and diverse, produces high welding quality, saves costs, offers multiple functions, and reduces the investment in manpower and other equipment.

[0006] The technical solution to the technical problem solved by the present invention is as follows: The present invention provides a modular welding equipment, including a temperature measuring module, a temperature adjustment module, a welding module and a slag cleaning module arranged in sequence, and a track arranged along the length of the weld. The temperature measuring module, the temperature adjustment module, the welding module and the slag cleaning module each include a walking unit, a vehicle body, a vertical moving unit, a horizontal moving unit and a swinging unit. Two adjacent vehicle bodies are connected by a connector. The walking unit is located at the bottom of the vehicle body and moves along the track. A vertical moving unit is arranged on the vehicle body. A horizontal moving unit is arranged at the output end of the vertical moving unit. A swinging unit is arranged at the output end of the horizontal moving unit. A temperature measuring gun for temperature measurement or a wind duct for temperature adjustment or a welding gun for welding or a vibratory hammer for slag cleaning is arranged at the end of the swinging unit.

[0007] As an optimization, the walking unit includes a drive motor, a power wheel, a track wheel, and a connecting shaft. The drive motor is located at the bottom of the vehicle body, and the power wheel is located at the output end of the drive motor. The power wheel and the track wheel are connected by a connecting shaft, which is rotatably located at the bottom of the vehicle body.

[0008] As an optimization, the vertical moving unit includes a mounting frame, a hydraulic cylinder, and a mounting plate. The mounting frame is set on the top surface of the vehicle body, the hydraulic cylinder is vertically mounted on the mounting frame, the output end of the hydraulic cylinder is mounted on the mounting plate, and the horizontal moving unit is mounted on the mounting plate.

[0009] As an optimization, the horizontal moving unit includes a second hydraulic cylinder, an ejector plate, and a connecting plate. The second hydraulic cylinder is horizontally mounted on the mounting plate along the vertical length of the vehicle body. The ejector plate is vertically mounted on the output end of the second hydraulic cylinder. The side of the ejector plate away from the first hydraulic cylinder is connected to the connecting plate. The end of the connecting plate away from the ejector plate is equipped with a swing unit.

[0010] As an optimization, the swing unit includes a second drive motor, a swing shaft, and a clamp. The second drive motor is vertically fixed at the end of the connecting plate away from the push-out plate. The swing shaft is set at the output end of the second drive motor. The swing shaft includes a horizontal shaft and an inclined shaft. One end of the horizontal shaft is connected to the second drive motor, and the other end is connected to the inclined shaft. The inclined shaft is set inclined downward. The clamp is set at the end of the inclined shaft away from the horizontal shaft. The clamp is used to hold a temperature measuring gun, a fan, a welding gun, or a vibratory hammer.

[0011] As an optimization, the clamp is a snap-fit ​​and the connector is a bolt.

[0012] As an optimization, a guide component one is set on the mounting frame along the vertical direction. The output end of the guide component one is connected to the support plate one. The top of the support plate one is connected to the mounting plate. A guide component two is set on the side of the support plate one away from the hydraulic cylinder one along the horizontal direction. The output end of the guide component two is connected to the connecting plate.

[0013] As an optimization, both guide component one and guide component two are guide rail slider pairs.

[0014] This invention also provides a welding quality evaluation method, applied to the aforementioned modular welding equipment, characterized in that it includes:

[0015] Step 1: Image acquisition of the weld surface: The industrial camera is driven to be directly above the weld by the swing unit of the slag cleaning module; the second drive motor adjusts the angle of the clamp through the swing shaft so that the optical axis of the industrial camera is perpendicular to the weld surface; the height is adjusted by the vertical movement unit; the lateral offset is finely adjusted by the horizontal movement unit so that the weld is located in the center of the image; n images are taken at equal intervals along the track according to the length of the weld.

[0016] Step 2, Noise Adaptive Filtering: Calculate the image noise intensity using the local variance estimation method; based on the image noise intensity, apply weighted nonlocal mean denoising to obtain the filtered image;

[0017] Step 3: Multi-operator adaptive recognition: Preset operators are called according to priority to perform defect detection on the filtered image. The CLAHE algorithm is used to identify the filtered image to obtain a first recognition probability. If the first recognition probability is greater than or equal to a preset threshold, it is determined to be a crack defect. If the first recognition probability is less than the preset threshold, the gray-scale linear transformation algorithm is used to identify the filtered image to obtain a second recognition probability. If the second recognition probability is greater than or equal to the preset threshold, it is determined to be a porosity defect. If the second recognition probability is less than the preset threshold, the wavelet denoising algorithm is used to identify the filtered image to obtain a third recognition probability. If the third recognition probability is greater than or equal to the preset threshold, it is determined to be a slag inclusion defect.

[0018] Step 4, Comprehensive Scoring and Closed-Loop Control: The welding qualification is determined by calculating the average recognition probability of n images. When the average recognition probability is less than the preset threshold, the welding is deemed qualified and the modular welding equipment moves to the next welding area. When the average recognition probability is greater than or equal to the preset threshold, the welding is deemed unqualified, the defect type is output, and the vibratory hammer of the slag cleaning module is controlled to clean the defect area in a targeted manner. After cleaning, steps 1-4 are repeated until the welding is qualified.

[0019] The effects described in the invention are merely those of the embodiments, and not all the effects of the invention. The above technical solutions have the following advantages or beneficial effects:

[0020] 1. This welding equipment is multifunctional, integrating various modules in the welding process to meet diverse processing requirements; the main body of the equipment adopts a modular design, which can be adjusted according to processing needs, offering flexibility and variety; the welding quality is high, and the equipment mechanizes the processing by setting up vertical movement units, horizontal movement units, and swing units, avoiding the instability of manual labor; at the same time, it saves costs, offers diverse functions, and reduces the investment in manpower and other equipment;

[0021] 2. Start the equipment to perform welding operations. First, the temperature measuring gun measures the temperature of the base material to be welded. Then, control the air duct to blow air onto the welding position of the base material to ensure that its temperature meets the welding requirements. After the base material temperature reaches the standard, adjust the vehicle's travel speed and the size of the welding gun flame to perform welding operations on the base material. After welding is completed, use a vibrating hammer to remove the weld slag. This equipment is convenient and practical and can be adjusted according to processing needs.

[0022] 3. The welding quality evaluation method of this invention utilizes noise adaptive filtering technology, based on local variance estimation and weighted nonlocal mean denoising, to dynamically adjust the filtering intensity, effectively suppressing spatter noise and reflection interference while preserving weld texture details and avoiding excessive blurring of defect features. Combined with multi-operator adaptive recognition, it prioritizes and calls CLAHE, grayscale linear transformation, and wavelet denoising algorithms to systematically detect cracks, porosity, and slag inclusions, improving recognition accuracy and robustness. Furthermore, through closed-loop control, it judges passability based on average recognition probability, automatically triggering targeted cleaning and re-inspection, achieving real-time feedback and self-correction of welding quality. This significantly improves welding efficiency, reduces manual intervention, ensures the final weld surface quality meets standards, and reduces the scrap rate. Attached Figure Description

[0023] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0024] Figure 1 This is a top view of the present invention;

[0025] Figure 2 This is the left view of the present invention;

[0026] Figure 3 This is a schematic diagram of the welding module of the present invention.

[0027] Figure 4 This is a structural diagram of the temperature measurement module of the present invention;

[0028] Figure 5 This is a structural diagram of the temperature adjustment module of the present invention;

[0029] Figure 6 This is a structural diagram of the welding slag cleaning module of the present invention.

[0030] Figure 7 Original welding image;

[0031] Figure 8 This is the image after mean filtering.

[0032] Figure 9 The image after Gaussian filtering;

[0033] Figure 10 The image is processed by the noise adaptive filtering method of the present invention.

[0034] In the diagram, 1. Vehicle body; 2. Temperature measuring gun; 3. Air duct; 4. Welding gun; 5. Vibratory hammer; 6. Bolt; 7. Motor 1; 8. Drive wheel; 9. Track wheel; 10. Connecting shaft; 11. Mounting bracket; 12. Hydraulic cylinder 1; 13. Mounting plate; 14. Hydraulic cylinder 2; 15. Push-out plate; 16. Connecting plate; 17. Drive motor 2; 18. Swing shaft; 181. Horizontal shaft; 182. Inclined shaft; 19. Clamp; 20. Support plate 1; 21. Track. Detailed Implementation

[0035] To clearly illustrate the technical features of this solution, the invention will be described in detail below through specific embodiments and in conjunction with the accompanying drawings. The following disclosure provides many different embodiments or examples for implementing different structures of the invention. To simplify the disclosure of the invention, components and arrangements of specific examples are described below. Furthermore, reference numerals and / or letters may be repeated in different examples. This repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. It should be noted that the components illustrated in the drawings are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and processes are omitted to avoid unnecessarily limiting the invention. Terms such as "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0036] Example 1

[0037] like Figures 1 to 6As shown, a modular welding device includes a temperature measuring module, a temperature adjustment module, a welding module, and a slag cleaning module arranged sequentially. It also includes a track 21 arranged along the length of the weld. Each of the temperature measuring module, temperature adjustment module, welding module, and slag cleaning module includes a walking unit, a vehicle body 1, a vertical moving unit, a horizontal moving unit, and a swinging unit. Two adjacent vehicle bodies 1 are connected by a connector. The walking unit is located at the bottom of the vehicle body 1 and moves along the track 21. A vertical moving unit is provided on the vehicle body 1. A horizontal moving unit is provided at the output end of the vertical moving unit. A swinging unit is provided at the output end of the horizontal moving unit. A temperature measuring gun 2 for temperature measurement, a wind duct 3 for temperature adjustment, a welding gun 4 for welding, or a vibratory hammer 5 for slag cleaning is provided at the end of the swinging unit. This welding equipment is multifunctional, integrating various modules in the welding process to meet diverse processing requirements. The main body of the equipment adopts a modular design, which can be adjusted according to processing needs, offering flexibility and variety. It produces high-quality welds. By setting up vertical movement units, horizontal movement units, and swing units, the equipment mechanizes the processing process, avoiding the instability of manual labor. At the same time, it saves costs, offers diverse functions, and reduces the investment in manpower and other equipment.

[0038] In this embodiment, the walking unit includes a drive motor 7, a drive wheel 8, a track wheel 9, and a connecting shaft 10. The drive motor 7 is located at the bottom of the vehicle body 1, and the drive wheel 8 is located at the output end of the drive motor 7. The drive wheel 8 and the track wheel 9 are connected by the connecting shaft 10, which is rotatably located at the bottom of the vehicle body 1. The transmission mechanism between the drive motor 7 and the drive wheel 8 is an existing structure and will not be described in detail here. When the drive motor 7 is started, it drives the drive wheel 8 to rotate. At the same time, the track wheel 9 also rotates, thereby moving along the track 21 and driving each module forward or backward.

[0039] In this embodiment, the vertical moving unit includes a mounting frame 11, a hydraulic cylinder 12, and a mounting plate 13. The mounting frame 11 is disposed on the top surface of the vehicle body 1. The hydraulic cylinder 12 is vertically mounted on the mounting frame 11, and the mounting plate 13 is disposed at the output end of the hydraulic cylinder 12. A horizontal moving unit is disposed on the mounting plate 13. When the hydraulic cylinder 12 extends, it pushes the mounting plate 13 to move, thereby adjusting the height of the swing unit in the vertical direction.

[0040] In this embodiment, the horizontal movement unit includes a second hydraulic cylinder 14, an ejector plate 15, and a connecting plate 16. The second hydraulic cylinder 14 is horizontally mounted on the mounting plate 13 along the length direction perpendicular to the vehicle body 1. The ejector plate 15 is vertically mounted on the output end of the second hydraulic cylinder 14. The side of the ejector plate 15 away from the first hydraulic cylinder 12 is connected to the connecting plate 16. A swing unit is mounted on the end of the connecting plate 16 away from the ejector plate 15. When the second hydraulic cylinder 14 is activated, it drives the ejector plate 15 to extend horizontally, thereby driving the connecting plate 16 to extend as well. This adjusts the position of the swing unit in the horizontal direction to adapt to the working position of each module.

[0041] In this embodiment, the swing unit includes a second drive motor 17, a swing shaft 18, and a clamp 19. The second drive motor 17 is vertically fixed at the end of the connecting plate 16 away from the push-out plate 15. The swing shaft 18 is located at the output end of the second drive motor 17. The swing shaft 18 includes a horizontal shaft 181 and an inclined shaft 182. One end of the horizontal shaft 181 is connected to the second drive motor 17, and the other end is connected to the inclined shaft 182. The inclined shaft 182 is inclined downwards. The clamp 19 is located at the end of the inclined shaft 182 away from the horizontal shaft 181. The clamp 19 is used to hold the temperature measuring gun 2, the air duct 3, the welding gun 4, or the vibratory hammer 5. The air duct 3 is used in conjunction with a device that uses cooling gas such as nitrogen or argon to blow into the welding area to reduce the temperature. This type of device is prior art and will not be described in detail here. The vibratory hammer 5 is a structure that can vibrate and hammer the weld slag. It will not be described in detail here. Drive motor 17 starts, causing the swing shaft 18 to rotate horizontally, adjusting the position of the heat gun 2, the air duct 3, the welding gun 4, or the vibratory hammer 5.

[0042] The clamp 19 is a snap-fit, and the connecting part is a bolt 6. The snap-fit ​​can be adjusted according to the different structures of the temperature measuring gun 2, the air duct 3, the welding gun 4, or the vibratory hammer 5 being clamped.

[0043] In this embodiment, a guide component one is vertically arranged on the mounting frame 11. The output end of the guide component one is connected to a support plate one 20, and the top of the support plate one 20 is connected to a mounting plate 13. A guide component two is horizontally arranged on the side of the support plate one 20 away from the hydraulic cylinder one 12, and the output end of the guide component two is connected to a connecting plate 16. By setting the guide component one and the guide component two, the rod ends of the hydraulic cylinder one 12 and the hydraulic cylinder two 14 can be made more stable during the extension or retraction process, while supporting the overall structure and making the movement more stable.

[0044] In this embodiment, both guide component one and guide component two are guide rail slider pairs.

[0045] The specific working process is as follows: A track 21 is erected according to the length of the weld seam, and the functional modules are assembled. According to the welding requirements, the assembled module body 1 is assembled using bolts 6 and installed on the track 21. Hydraulic cylinder 12, hydraulic cylinder 14, and drive motor 17 are adjusted to adjust the swing shaft 18, ensuring that the temperature measuring gun 2, air duct 3, welding gun 4, or vibrating hammer 5 are aligned with the welding position. After adjustment, the equipment is started for welding. The temperature measuring gun 2 first measures the temperature of the base material, then controls the air duct 3 to blow air onto the welding position of the base material to ensure its temperature meets the welding requirements. After the base material temperature reaches the standard, the traveling speed of the body 1 and the flame size of the welding gun 4 are adjusted to perform welding operations on the base material. After welding is completed, the weld slag is removed using the vibrating hammer 5.

[0046] Example 2

[0047] This embodiment provides a method for evaluating welded surfaces, applied to the modular welding equipment described in Embodiment 1, characterized in that it includes:

[0048] Step 1: Weld Surface Image Acquisition: The industrial camera is driven to be directly above the weld seam by the swing unit of the weld slag cleaning module; the drive motor 2 adjusts the fixture angle via the swing axis to make the optical axis of the industrial camera perpendicular to the weld seam surface; the height is adjusted by the vertical movement unit, and the formula for calculating the height is:

[0049] ,

[0050] in, Indicates the weld height; Indicates the camera's focal length; Indicates the size of the camera sensor; Indicates the weld width; fine-tunes the lateral offset using the horizontal movement unit. To center the weld seam in the image; take n images at equal intervals along the track according to the weld seam length, where n≥3, with a step size of... , Indicates the weld length;

[0051] Step 2: To address common issues in weld surface images such as spatter noise, glare interference, and blurred defect features, an adaptive noise filtering method is used to process the images.

[0052] Adaptive noise filtering: Calculating image noise intensity using local variance estimation. ,in, For pixel coordinates, the formula is as follows:

[0053] ,

[0054] in, The sliding window size is 3×3 or 5×5; The pixel grayscale value; The mean gray level within the window is used; weighted nonlocal mean denoising is applied based on image noise intensity to obtain the filtered image; the formula for calculating the filtered pixel value is:

[0055] ,

[0056] in, For search window; These are the pixel coordinates in the search window; As weight, ,in, To prevent constants with a denominator of 0, this invention uses image noise intensity. The filtering intensity is dynamically adjusted; the stronger the noise, the greater the filtering weight, thus preserving the details of defects while suppressing noise.

[0057] In one embodiment, the noise adaptive filtering method of the present invention is compared with existing methods. As shown in Figures 7, 8, 9, and 10, the kernel size of the mean filter is 7×7, and the kernel size of the Gaussian filter is 9×9. As can be seen from the figures, the method of the present invention accurately calculates the image noise intensity through local variance estimation, and can specifically identify and process noise. Compared with mean filtering (which easily blurs images and loses details) and Gaussian filtering (which has limited adaptability to complex noise), the noise adaptive filtering method can efficiently remove noise. Figure 10 Noise in and around the weld seam is reduced, resulting in a cleaner image. Weighted nonlocal mean denoising, with its reasonable weight allocation, better preserves weld seam texture details while reducing noise. In contrast, mean filtering blurs the weld seam outline, and Gaussian filtering is insufficient for improving subtle textures. The image after noise adaptive filtering, however, shows clear weld seam details and good layering. This invention avoids the blurring of defect features caused by over-filtering, preserving key details for subsequent defect identification and improving the accuracy of evaluation.

[0058] Step 3: Multi-operator adaptive recognition: Preset operators are invoked according to priority to perform defect detection on the filtered image. The CLAHE algorithm is used to identify the filtered image, obtaining a first recognition probability. If the first recognition probability p1 is greater than or equal to a preset threshold, it is determined to be a crack defect. If the first recognition probability p1 is less than the preset threshold, a gray-scale linear transformation algorithm is used to identify the filtered image, obtaining a second recognition probability. If the second recognition probability p2 is greater than or equal to a preset threshold, it is determined to be a porosity defect. If the second recognition probability p2 is less than the preset threshold, a wavelet denoising algorithm is used to identify the filtered image, obtaining a third recognition probability. If the third recognition probability p3 is greater than or equal to a preset threshold, it is determined to be a slag inclusion defect. The preset threshold is set to 0.8.

[0059] Step 4, Comprehensive Scoring and Closed-Loop Control: The welding qualification is determined by calculating the average recognition probability of n images. When the average recognition probability p is less than the preset threshold, the welding is deemed qualified and the modular welding equipment moves to the next welding area. When the average recognition probability p is greater than or equal to the preset threshold, the welding is deemed unqualified, the defect type is output, and the vibratory hammer of the slag cleaning module is controlled to clean the defect area in a targeted manner. After cleaning, steps 1-4 are repeated until the welding is qualified.

[0060] Although the specific embodiments of the invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the invention. Based on the technical solutions of the invention, various modifications or variations that can be made by those skilled in the art without creative effort are still within the scope of protection of the invention.

Claims

1. A welding quality evaluation method, based on a modular welding device, comprising a temperature measuring module, a temperature adjustment module, a welding module, and a slag cleaning module arranged sequentially, and a track (21) arranged along the length of the weld. The temperature measuring module, temperature adjustment module, welding module, and slag cleaning module each include a walking unit, a vehicle body (1), a vertical moving unit, a horizontal moving unit, and a swinging unit. Two adjacent vehicle bodies (1) are connected by a connector. The walking unit is located at the bottom of the vehicle body (1) and moves along the track (21). A vertical moving unit is provided on the vehicle body (1). A horizontal moving unit is provided at the output end of the vertical moving unit. A swinging unit is provided at the output end of the horizontal moving unit. A temperature measuring gun (2) for temperature measurement, a wind duct (3) for temperature adjustment, a welding gun (4) for welding, or a vibrating hammer (5) for slag cleaning is provided at the end of the swinging unit. The method is characterized by including: Step 1: Image acquisition of the weld surface: The industrial camera is driven to be directly above the weld seam by the swing unit of the slag cleaning module; The second drive motor adjusts the angle of the clamp via a swing shaft, making the optical axis of the industrial camera perpendicular to the weld surface; the height is adjusted via a vertical moving unit; the lateral offset is finely adjusted via a horizontal moving unit, so that the weld is located at the center of the image; n images are taken at equal intervals along the track according to the weld length; Step 2, Adaptive Noise Filtering: The image noise intensity is calculated using local variance estimation; weighted nonlocal mean denoising is then applied based on the image noise intensity to obtain the filtered image; specifically, The formula for calculating image noise intensity is as follows: ， in, These are pixel coordinates; The sliding window size is 3×3 or 5×5; The pixel grayscale value; This represents the average grayscale value within the window. The formula for weighted nonlocal means denoising is expressed as follows: ， in, These are the filtered pixel values; For search window; These are the pixel coordinates in the search window; As weight, ,in, To prevent constants with a denominator of 0; Step 3: Multi-operator adaptive recognition: Preset operators are called according to priority to perform defect detection on the filtered image. The CLAHE algorithm is used to identify the filtered image to obtain a first recognition probability. If the first recognition probability is greater than or equal to a preset threshold, it is determined to be a crack defect. If the first recognition probability is less than the preset threshold, the gray-scale linear transformation algorithm is used to identify the filtered image to obtain a second recognition probability. If the second recognition probability is greater than or equal to the preset threshold, it is determined to be a porosity defect. If the second recognition probability is less than the preset threshold, the wavelet denoising algorithm is used to identify the filtered image to obtain a third recognition probability. If the third recognition probability is greater than or equal to the preset threshold, it is determined to be a slag inclusion defect. Step 4, Comprehensive Scoring and Closed-Loop Control: The welding qualification is determined by calculating the average recognition probability of n images. When the average recognition probability is less than the preset threshold, the welding is deemed qualified and the modular welding equipment moves to the next welding area. When the average recognition probability is greater than or equal to the preset threshold, the welding is deemed unqualified, the defect type is output, and the vibratory hammer of the slag cleaning module is controlled to clean the defect area in a targeted manner. After cleaning, steps 1-4 are repeated until the welding is qualified.

2. The welding quality evaluation method according to claim 1, characterized in that: The walking unit includes a drive motor (7), a power wheel (8), a track wheel (9) and a connecting shaft (10). The drive motor (7) is located at the bottom of the vehicle body (1), and the output end of the drive motor (7) is provided with a power wheel (8). The power wheel (8) and the track wheel (9) are connected by the connecting shaft (10), which is rotatably located at the bottom of the vehicle body (1).

3. The welding quality evaluation method according to claim 1 or 2, characterized in that: vertical The moving unit includes a mounting frame (11), a hydraulic cylinder (12) and a mounting plate (13). The mounting frame (11) is set on the top surface of the vehicle body (1). The hydraulic cylinder (12) is vertically mounted on the mounting frame (11). The output end of the hydraulic cylinder (12) is mounted on the mounting plate (13). The horizontal moving unit is mounted on the mounting plate (13).

4. The welding quality evaluation method according to claim 3, characterized in that: The horizontal moving unit includes a second hydraulic cylinder (14), an ejector plate (15), and a connecting plate (16). The second hydraulic cylinder (14) is horizontally mounted on the mounting plate (13) along the length direction perpendicular to the vehicle body (1). The ejector plate (15) is vertically mounted at the output end of the second hydraulic cylinder (14). The side of the ejector plate (15) away from the first hydraulic cylinder (12) is connected to the connecting plate (16). The end of the connecting plate (16) away from the ejector plate (15) is equipped with a swing unit.

5. The welding quality evaluation method according to claim 4, characterized in that: The swing unit includes a second drive motor (17), a swing shaft (18), and a clamp (19). The second drive motor (17) is vertically fixed at one end of the connecting plate (16) away from the push-out plate (15). The output end of the second drive motor (17) is provided with a swing shaft (18). The swing shaft (18) includes a horizontal shaft (181) and an inclined shaft (182). One end of the horizontal shaft (181) is connected to the second drive motor (17), and the other end is connected to the inclined shaft (182). The inclined shaft (182) is inclined downward. The end of the inclined shaft (182) away from the horizontal shaft (181) is provided with a clamp (19). The clamp (19) is used to hold a temperature measuring gun (2), a blower (3), a welding gun (4), or a vibratory hammer (5).

6. The welding quality evaluation method according to claim 5, characterized in that: The clamp (19) is a snap fastener; the connector is a bolt (6).

7. The welding quality evaluation method according to claim 4, characterized in that: A guide component 1 is set on the mounting bracket (11) in the vertical direction. The output end of the guide component 1 is connected to the support plate 1 (20). The top of the support plate 1 (20) is connected to the mounting plate (13). A guide component 2 is set on the side of the support plate 1 (20) away from the hydraulic cylinder 1 (12) in the horizontal direction. The output end of the guide component 2 is connected to the connecting plate (16).

8. The welding quality evaluation method according to claim 7, characterized in that: Both guide component one and guide component two are guide rail slider pairs.

Images