A membrane type tube panel fillet weld appearance quality detection system and its working method

Abstract

A visual quality inspection system and its working method for fillet welds of membrane panel tube screens, which belong to the technical field of product quality inspection. It includes a camera, a marking device, a bracket, a driving roller path, a driving motor and a control center; the driving roller path is connected to the driving motor; the bracket spans above the driving roller path, the camera and the marking device are arranged on the bracket, and the camera and the marking device are facing the driving roller path below; the camera, the marking device and the driving motor are respectively connected to the control center. The design of the present invention is convenient and reasonable, realizing the automatic inspection of the visual quality of fillet welds of membrane panel tube screens and the automatic marking of defects, greatly improving the efficiency and quality of the visual quality inspection of welds.

Description

Technical Field

[0001] The present invention belongs to the technical field of product quality inspection, and particularly relates to a detection system and a working method for the appearance quality of fillet welds of membrane tube screens. Background Art

[0002] Membrane tube screen components such as water walls of utility boilers are generally made by welding high-pressure seamless steel pipes and fins. The forming quality of the fillet welds between the pipes and the fins directly affects the quality of the tube screen.

[0003] Due to the large number and long length of the fillet welds of membrane tube screens, manual visual inspection is generally used. However, there are many factors affecting the quality of manual visual inspection, which are often affected by a series of factors such as the professional ability, responsibility of inspectors, and light in the inspection site. Moreover, manual inspection is time-consuming and laborious, and missed inspection accidents often occur. The fillet welds of membrane tube screens are often complained by customers due to poor forming quality, and a large amount of manpower and material resources are required for defect elimination treatment at the installation site, bringing great potential safety hazards. Summary of the Invention

[0004] In order to solve the above problems, the purpose of the present invention is to provide a detection system and a working method for the appearance quality of fillet welds of membrane tube screens, which realizes the automatic inspection of the appearance quality of fillet welds of boiler membrane tube screen components, and greatly improves the efficiency and quality of the inspection of the appearance quality of welds.

[0005] The present invention is realized through the following technical solutions:

[0006] The present invention discloses a detection system for the appearance quality of fillet welds of membrane tube screens, including a camera, a marking device, a bracket, a driving roller table, a driving motor, and a control center; the driving roller table is connected to the driving motor; the bracket spans above the driving roller table, the camera and the marking device are arranged on the bracket, and the camera and the marking device face the driving roller table below; the camera, the marking device, and the driving motor are respectively connected to the control center.

[0007] Preferably, one camera and one marking device form a set of detection modules. According to the number of fillet welds of the to-be-tested tube screen, several sets of detection modules are arranged on the bracket.

[0008] Further preferably, each set of detection modules further includes a lighting device.

[0009] Preferably, some of the driving rollers on the driving roller table are driving rollers, and the other part is driven rollers. The driving rollers are connected to the driving motor.

[0010] Preferably, the bracket includes a cross beam and two columns. The two columns are respectively arranged on both sides of the driving roller table and fixed to the ground. The two ends of the cross beam are respectively connected to the two columns; a slide rail is arranged on the cross beam, and several sliders are embedded in the slide rail. The camera and the marking device are fixed on the sliders.

[0011] Further preferably, a locking device is provided on the slider, which can be fixed at a specific position on the slide rail.

[0012] Further preferably, both the cross beam and the two columns are telescopic structures.

[0013] Preferably, the marking device includes an ink storage tank and a nozzle. The ink storage tank is connected to the nozzle through a connecting pipe. A ink delivery pump is provided on the connecting pipe, and the nozzle is arranged on the bracket.

[0014] The working method of the above-mentioned film type tube panel fillet weld appearance quality detection system disclosed by the present invention includes:

[0015] The film type tube panel to be tested is placed on the driving roller path. The cameras and marking devices on the bracket adjust their positions according to the positions of the fillet welds of the tube panel. The control center turns on the cameras, marking devices and driving motor. The film type tube panel is conveyed on the driving roller path and passes under the bracket. At the same time, the cameras collect the image information of the fillet welds of the tube panel and transmit it to the control center, and compare it with the defect photos in the control center. After finding the weld defects, the control center controls the driving motor to decelerate or stop. The marking device sprays identification color near the defect positions, which is convenient for subsequent defect confirmation and repair. The system continues to run for subsequent inspections until completion.

[0016] Preferably, a weld appearance defect database is provided in the control center. An evaluation model is established by using machine learning. A large number of weld appearance defect images are preprocessed and classified, and then the pixel-level information of the images is extracted to train and learn the recognition module, and an alarm threshold is set.

[0017] Compared with the prior art, the present invention has the following beneficial technical effects:

[0018] A film type tube panel fillet weld appearance quality detection system disclosed by the present invention arranges a plurality of cameras on the bracket above the film type tube panel, and moves the film type tube panel on the driving roller path under the cameras. Each camera checks a fillet weld until the whole weld passes under the cameras for inspection. For the defects found during the inspection, the marking device beside the camera can mark by inkjet, which is convenient for workers to repair later. The system is designed conveniently and reasonably, realizes the automatic inspection of the film type tube panel fillet weld appearance quality and the automatic marking of defects, and greatly improves the efficiency and quality of the weld appearance quality inspection.

[0019] Further, one camera and one marking device form a set of detection modules, which are configured according to the number of fillet welds of the tube panel to be tested, improving the detection quality and saving the operation cost.

[0020] Furthermore, each group of detection modules further includes a lighting device, which can automatically adjust the light brightness required for imaging and photographing according to the ambient light, thereby improving the clarity of the captured images.

[0021] Further, some of the drive rollers on the drive roller path are driving rollers, and the other part are driven rollers. The driving rollers are connected to the drive motor, which can reduce the number of motors and lower the energy consumption.

[0022] Further, the slide rails on the bracket can enable the cameras and marking devices on the sliders to be adjusted according to the structure and size of the membrane type tube screen.

[0023] Furthermore, locking devices are provided on the sliders, which can be fixed at specific positions on the slide rails, avoiding displacement and vibration during operation and improving stability.

[0024] Furthermore, the cross beam and the two columns are both telescopic structures, which can adjust the height and span according to the actual situation and adapt to various workpiece sizes.

[0025] The working method of the above-mentioned membrane type tube screen fillet weld appearance quality detection system disclosed by the present invention has a high degree of automation, can efficiently identify the defects of the tube screen fillet welds, saves manpower and material resources, and does not depend on the skill level and experience of the operator, and has a good application prospect. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0027] In the figure: 1 is the camera, 2 is the lighting device, 3 is the marking device, 4 is the bracket, 5 is the drive roller path, 6 is the tube screen fillet weld, 7 is the drive motor, and 8 is the control center. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] The following further describes the present invention in detail with reference to the drawings. The content is an explanation of the present invention rather than a limitation:

[0029] As Figure 1 , it is the membrane type tube screen fillet weld appearance quality detection system of the present invention, including the camera 1, the marking device 3, the bracket 4, the drive roller path 5, the drive motor 7 and the control center 8; the drive roller path 5 is connected to the drive motor 7; the bracket 4 spans above the drive roller path 5, the camera 1 and the marking device 3 are arranged on the bracket 4, and the camera 1 and the marking device 3 are facing the drive roller path 5 below; the camera 1, the marking device 3 and the drive motor 7 are respectively connected to the control center 8.

[0030] In a preferred embodiment of the present invention, a camera 1 and a marking device 3 form a set of detection modules. According to the number of fillet welds 6 of the tube screen to be measured, several sets of detection modules are provided on the bracket 4. Preferably, each set of detection modules further includes a lighting device 2. Preferably, the camera 1, the lighting device 2 and the marking device 3 are connected to the control center 8 wirelessly to reduce wiring. Preferably, the camera 1 can rotate 360 degrees. Preferably, the brightness of the lighting device 2 can be automatically adjusted according to the ambient light to meet the requirements of the camera 1.

[0031] In a preferred embodiment of the present invention, some of the driving rollers on the driving roller path 5 are driving rollers and the other part are driven rollers. The driving rollers are connected to the driving motor 7; the driving roller path 5 is controlled by the driving motor 7 and the speed is adjustable.

[0032] In a preferred embodiment of the present invention, the bracket 4 includes a cross beam and two columns. The two columns are respectively arranged on both sides of the driving roller path 5 and fixed to the ground. The two ends of the cross beam are respectively connected to the two columns; a slide rail is provided on the cross beam, and several sliders are embedded in the slide rail. The camera 1 and the marking device 3 are fixed on the sliders. Preferably, a locking device is provided on the sliders, which can be fixed at a specific position on the slide rail. Preferably, both the cross beam and the two columns are telescopic structures.

[0033] In a preferred embodiment of the present invention, the marking device 3 includes an ink storage tank and a nozzle. The ink storage tank is connected to the nozzle through a connecting pipe, and an ink delivery pump is provided on the connecting pipe. The nozzle is arranged on the bracket 4. The ink storage tank is detachable and different colors can be replaced according to the color of the workpiece to be measured.

[0034] The working method of the above-mentioned film tube screen fillet weld appearance quality detection system:

[0035] The film tube screen to be measured is placed on the driving roller path 5. The camera 1 and the marking device 3 on the bracket 4 adjust their positions according to the position of the fillet weld 6 of the tube screen. The control center 8 turns on the camera 1, the marking device 3 and the driving motor 7. The film tube screen is conveyed on the driving roller path 5 and passes under the bracket 4. At the same time, the camera 1 collects the image information of the fillet weld 6 of the tube screen and transmits it to the control center 8, and compares it with the defect photos in the control center 8. After detecting a weld defect, the control center 8 controls the driving motor 7 to decelerate or stop. The marking device 3 sprays a marking color near the defect position to facilitate subsequent defect confirmation and repair. The system continues to run for subsequent inspections until completion.

[0036] In a preferred embodiment of the present invention, a weld appearance defect database is provided in the control center 8. An evaluation model is established by using machine learning. A large number of weld appearance defect images are preprocessed and classified, and then the pixel-level information of the images is extracted to train and learn the recognition module, and an alarm threshold is set.

[0037] As described above, it is only a part of the embodiments of the present invention. Although some terms are used in the present invention, the possibility of using other terms is not excluded. The use of these terms is only for the convenience of describing and explaining the essence of the present invention, and interpreting them as any additional limitation is contrary to the spirit of the present invention. The above description only uses embodiments to further illustrate the content of the present invention for easier understanding, but it does not mean that the embodiments of the present invention are limited to this. Any technical extension or re-creation based on the present invention is protected by the present invention.

Claims

1. A visual quality inspection system for fillet welds of membrane tube screens, characterized in that it includes a camera (1), a marking device (3), a bracket (4), a driving roller path (5), a driving motor (7) and a control center (8); the driving roller path (5) is connected to the driving motor (7); the bracket (4) spans above the driving roller path (5), the camera (1) and the marking device (3) are arranged on the bracket (4), and the camera (1) and the marking device (3) face the driving roller path (5) below; the camera (1), the marking device (3) and the driving motor (7) are respectively connected to the control center (8); one camera (1) and one marking device (3) form a set of detection modules, and according to the number of fillet welds (6) of the tube screen to be measured, several sets of detection modules are arranged on the bracket (4); the bracket (4) includes a cross beam and two columns, the two columns are respectively arranged on both sides of the driving roller path (5) and fixed to the ground, and the two ends of the cross beam are respectively connected to the two columns; a slide rail is arranged on the cross beam, and several sliders are embedded in the slide rail, and the camera (1) and the marking device (3) are fixed on the sliders.

2. The appearance quality detection system for the fillet welds of the membrane type tube platen according to claim 1, wherein Each set of detection modules further includes a lighting device (2).

3. The appearance quality detection system for the fillet weld of the membrane tube panel according to claim 1, characterized in that, Some of the driving rollers on the driving roller path (5) are driving rollers, and the other part is driven rollers, and the driving rollers are connected to the driving motor (7).

4. The appearance quality detection system for the fillet weld of the membrane type tube panel according to claim 1, characterized in that The slider is provided with a locking device, which can be fixed at a specific position on the slide rail.

5. The appearance quality detection system for the fillet weld of the membrane tube panel according to claim 1, characterized in that Both the cross beam and the two columns are telescopic structures.

6. The visual quality inspection system for the fillet welds of the membrane tube panel according to claim 1, characterized in that, The marking device (3) includes an ink storage tank and a nozzle, the ink storage tank is connected to the nozzle through a connecting pipe, an ink delivery pump is arranged on the connecting pipe, and the nozzle is arranged on the bracket (4).

7. The working method of the membrane tube panel fillet weld appearance quality inspection system according to any one of claims 1 to 6, characterized in that: It includes: The membrane tube screen to be measured is placed on the driving roller path (5), the camera (1) and the marking device (3) on the bracket (4) adjust their positions according to the position of the fillet weld (6) of the tube screen, the control center (8) turns on the camera (1), the marking device (3) and the driving motor (7), the membrane tube screen is conveyed under the bracket (4) through the driving roller path (5), and at the same time the camera (1) collects the image information of the fillet weld (6) of the tube screen and transmits it to the control center (8), and compares it with the defect photos in the control center (8). After finding a weld defect, the control center (8) controls the driving motor (7) to decelerate or stop, and the marking device (3) sprays a marking color near the defect position to facilitate subsequent defect confirmation and repair. The system continues to run for subsequent inspections until completion.

8. The visual quality inspection system for fillet welds of membrane tube screens according to claim 7, characterized in that a weld appearance defect database is provided in the control center (8), an evaluation model is established by using machine learning, a large number of weld appearance defect images are preprocessed and classified, then the pixel-level information of the images is extracted to train and learn the recognition module, and an alarm threshold is set.

Images