An intelligent inspection system and method for an automatic production line of a building PC component factory
By introducing an intelligent inspection system on the production line of PC component factory, and using high-resolution cameras and TOF cameras for automated inspection, the problem of lack of automated inspection in the existing technology is solved, the degree of automation of the production line and product qualification rate are improved, and the further development is achieved. Efficient production process.
Patent Information
- Application Number
- CN201911312583.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-12-18
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2039-12-18
AI Technical Summary
The existing PC component factory production lines lack automated intelligent inspection systems, resulting in production quality relying on manual visual and gage inspection, which has low efficiency and poor accuracy.
An intelligent inspection system including industrial control machine, HUB Gigabit Network Hub and high-resolution camera was designed. Through software analysis of image data, the automation of mold installation, junction box installation and integrated inspection before pouring was realized, and the L-shaped origin coordinate position was used. The calibration device and X-Y linear module are accurately detected, and the distance depth data is obtained by combining the TOF camera to realize automatic position and dimension inspection of the mold, junction box and steel mesh.
It realizes automated inspection of the production line of PC components factory, improves the automation level of the production line and product qualification rate, reduces manual errors, improves production efficiency and accuracy, and provides technical support for the automation and intelligent development of the factory.
Smart Images

Figure CN110838189B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of building manufacturing, and more specifically, to an intelligent inspection system and method for an automatic production line of a building PC component factory. Background Art
[0002] In recent years, prefabricated buildings have developed rapidly in my country, and vigorously developing prefabricated buildings has risen to a national strategy. As an important part of prefabricated buildings, the factory production of PC (Precast Concrete) components has also developed rapidly. The automation, intelligence and informatization of PC prefabricated component factory production will be an inevitable trend in the future. At present, the current status of the production line of PC component factories in China is still in the mechanization and semi-automation stage. Although production equipment manufacturers have developed some automatic and semi-automatic mechanical equipment, such as mixing stations, feeders, marking machines, material distribution machines, stacking machines, curing kilns, mobile circulation equipment, etc., the automation of the quality inspection link of each station on the automatic production line is still in a completely blank state. At this stage, it is basically based on manual visual inspection and measuring tool inspection against drawings. To realize the automation and intelligence of PC factory production, the automatic inspection system on the production line is an indispensable link, and there is currently a lack of corresponding intelligent inspection systems. Summary of the invention
[0003] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide an intelligent inspection system and method for an automatic production line of a building PC component factory.
[0004] To achieve the above purpose, the technical solution provided by the present invention is as follows:
[0005] An intelligent inspection system for an automatic production line of a building PC component factory includes a connected industrial computer, a HUB gigabit network hub, and a number of high-resolution cameras installed at each inspection station; the inspection stations include at least a mold installation inspection station, a junction box installation inspection station, and a pre-casting integrated inspection station in the order of process flow;
[0006] The mold installation inspection station has an L-shaped origin coordinate position calibrator for calibrating the coordinate origin placed on its plane; the high-resolution camera is installed above it; the image taken by the high-resolution camera is input into the industrial computer through the HUB gigabit network hub, and the qualified inspection result is displayed by the external warning light through software inspection and analysis;
[0007] The junction box is installed at the inspection station, and the high-resolution camera is installed above it; the image taken by the high-resolution camera is input into the industrial computer through the HUB gigabit network hub, and the qualified inspection result is displayed by the external warning light through software inspection and analysis;
[0008] The pre-casting integrated inspection station has an XY linear module that can move vertically and horizontally set above it; the high-resolution camera and the TOF camera are installed together on the X beam of the XY linear module and can move left and right along the X beam; the X beam can move forward and backward along the Y beam of the XY linear module; the high-resolution camera is used to capture the image position of the object, and the TOF camera is used to capture the distance depth of the object; the image data captured by the two cameras are input into the industrial computer through the HUB gigabit network hub, and the qualified inspection results are displayed by the external warning light through software inspection and analysis.
[0009] As a further improvement of the present invention, the L-shaped origin coordinate position calibrator includes a right-angled single-sided X-axis scale and a Y-axis scale; each axis scale is provided with a plurality of luminous bodies for photographing and identification; the luminous bodies are arranged at a spacing of 150 mm.
[0010] As a further improvement of the present invention, the Y straight beams of the XY linear module are two parallel ones, which are spanned on the gantry on both sides of the pre-casting integrated inspection station.
[0011] As a further improvement of the present invention, the Y straight beam and the X horizontal beam are both provided with stepper motors for controlling operation; the operation of the stepper motors is directed and controlled by the industrial computer.
[0012] As a further improvement of the present invention, the industrial computer can also communicate with the host computer through Ethernet and connect with the production execution management MES system.
[0013] The present invention also discloses an intelligent inspection method for an automatic production line of a building PC component factory, comprising the following steps:
[0014] (1) Convert the user's CAD design file into a standard DXF file;
[0015] (2) According to the production line execution plan sequence, obtain from the corresponding DXF files:
[0016] a. Parameter data of the shape, size and dimensions of the composite plate;
[0017] b. Parameter data of the size and location of the junction box;
[0018] c. Parameter data of the horizontal and vertical number, length, spacing between steel bars, location of bars, length of bars, and location of truss beam bars of the tied steel bars;
[0019] (3) The parameter data obtained above is input into the corresponding workstation inspection program module to prepare for installation inspection;
[0020] (4) Check the mold installation station;
[0021] After the peripheral molds are installed, the L-shaped origin coordinate position calibrator is placed on the plane to set the origin coordinates, and then the high-resolution camera above is used to take pictures to form a mold station image that is input into the computer and analyzed and processed by intelligent image recognition and detection software. The content includes image denoising, feature extraction, mold installation origin and XY coordinate direction determination, and mold installation position size and drawing parameter data consistency check; if the inspection result is inconsistent, a warning signal is issued to prompt correction; and then check again until it is qualified;
[0022] (5) Check the junction box installation station:
[0023] After the junction box is installed, it is photographed by a high-resolution camera above, and the resulting junction box workstation image is input into the computer and analyzed and processed by intelligent image recognition and detection software. The content includes image noise removal, feature extraction of the junction box, determination of the mold origin and XY coordinate direction, and consistency check of the junction box installation position size and drawing parameter data. If there is any discrepancy, a warning signal is issued and the operator makes corrections. Then the inspection is repeated again until it is qualified.
[0024] (6) Place the tied steel mesh and embedded components into the mold frame for installation;
[0025] (7) Carry out inspection at the integrated inspection station before pouring:
[0026] The stepper motor of the XY linear module is controlled by a computer, so that the high-resolution camera and the TOF camera are simultaneously moved along the X-axis and Y-axis of the XY linear module to the designated measuring point for shooting; the high-resolution camera obtains the image position of the object, and the TOF camera obtains the distance depth of the object, and the image data taken by the two cameras are transmitted to the computer for storage;
[0027] Continue to move and control the high-resolution camera and TOF camera on the XY linear module until the image position map and distance depth map of all designated position measurement points are obtained; then the computer's intelligent image recognition and detection software will analyze and process; its content includes image denoising, relevant feature quantity extraction, mold installation origin and XY coordinate direction determination, vertical and horizontal steel bar number and spacing distance of steel mesh installation, steel bar position length, truss beam steel bar position, embedded parts position and corresponding parameter data consistency check of the drawings; if there is any discrepancy, a warning signal will be issued, and the operator will make corrections and check again until all installations are correct and all inspections are qualified;
[0028] (8) Enter the casting station and complete the casting of PC components.
[0029] The present invention uses computer image recognition and detection technology to carry out automatic inspection of each station before concrete pouring for the automatic production line of composite panels in the PC component factory, including mold installation station inspection, junction box installation station inspection and integrated inspection of steel bar binding embedded parts installation station. Compared with the existing technology, the intelligent automatic inspection system of the present invention fills the gap in production inspection automation technology on the automatic production line of PC components in the construction industry. It is an industry-first technology that can greatly improve the automation level, operation rhythm and product qualification rate of the production line, and also provides a more advanced method technology for the development of automation and intelligence in the production of PC prefabricated components factories in my country. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] Figure 1 It is a structural schematic diagram of the present invention;
[0031] Figure 2 for Figure 1 A schematic diagram of the structure of the L-shaped origin coordinate position calibrator;
[0032] Figure 3 for Figure 1 The structural diagram of the XY linear module in;
[0033] Figure 4 for Figure 3 A top view of
[0034] Figure 5 is a flow chart of the method of the present invention;
[0035] Figure 6 Schematic diagram of various software modules of the system of the present invention.
[0036] In the figure: 1 industrial computer, 2 HUB Gigabit network hub, 3 mold installation inspection station, 4 junction box installation inspection station, 5 pre-casting integrated inspection station, 6 steel bar-embedded parts delivery station, 7 TOF camera, 8 high-resolution camera, 9 warning light, 10 L-type origin coordinate position calibrator, 11 luminous body, 12 X-Y linear module, 13Y straight beam, 14X beam, 15 stepper motor, 16 gantry. DETAILED DESCRIPTION
[0037] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0038] like Figure 1 As shown, the intelligent inspection system of the automatic production line of the building PC component factory of the present invention is arranged on the assembly line of the composite board production workshop of the PC prefabrication factory, and is mainly used for automatic inspection of each station before concrete pouring, the most important of which are: mold installation inspection station 3, junction box installation inspection station 4, and pre-pouring integrated inspection station 5 after the installation of the steel bar binding embedded parts.
[0039] The main components of the present invention are also configured as follows Figure 1 As shown: an industrial computer 1 (industrial control computer) for inspection, a HUB Gigabit network hub 2 equipped with a Gigabit network card and a digital input and output DIO card, a warning light 9 showing the inspection result, and a number of high-resolution cameras 8 installed at each inspection station. In this embodiment, Gigabit network cameras are used. Intelligent software for photo control, image recognition and size detection of each station is also installed on the industrial computer 1; the HUB Gigabit network hub 2 is used to connect the industrial computer 1 and each high-resolution camera 8.
[0040] The mold installation inspection station 3 in the present invention has a Gigabit network high-resolution camera 8 set just above it, which is controlled by the industrial computer 1 to take pictures; after taking pictures, the image is input to the industrial computer 1 through the HUB Gigabit network hub 2, and the intelligent software on the industrial computer 1 checks whether the mold installation position is correct, and the warning light 9 displays whether the inspection result is qualified. In addition, an L-shaped origin coordinate position calibrator 10 is pre-placed on the plane of the mold installation inspection station 3 for calibrating the coordinate origin.
[0041] The structure of the L-type origin coordinate position calibrator 10 is as follows Figure 2 As shown, it is composed of a right-angled single-sided X-axis ruler and a Y-axis ruler; each axis ruler is provided with a number of luminous bodies 11 for photographing and identifying; in this embodiment, the X-axis ruler is the short axis, on which there are 3 luminous bodies 11, and the Y-axis ruler is the long axis, on which there are 4 luminous bodies 11 (where the luminous body 11 at the origin is shared and overlapped by the X-axis ruler and the Y-axis ruler), and each luminous body 11 is arranged at a spacing of 150 mm from the origin. The intelligent image recognition and detection software of the present invention can find the luminous body 11 of the L-shaped mold installation origin calibrator 10 from the image to calibrate the image origin and XY coordinate direction and image scale, and provide the origin and coordinate positioning and scale data for the inspection and measurement at each station, thereby ensuring the accuracy of the detection and positioning.
[0042] The junction box installation inspection station 4 in the present invention also has a Gigabit network high-resolution camera 8 set just above it, which is controlled by the industrial computer 1 to take pictures. After taking pictures, the image is input into the industrial computer 1 through the HUB Gigabit network hub 2, and the intelligent software checks whether the installation position of the junction box is correct, and the warning light 9 displays whether the inspection result is qualified.
[0043] The pre-casting integrated inspection station 5 of the present invention has an XY linear module 12 arranged directly above it, and a Gigabit network high-resolution camera 8 and a TOF camera 7 are installed on the module frame; the TOF camera 7 obtains the distance depth map of the mold, steel mesh, embedded parts and other objects installed on the mold platform, and the high-resolution camera 8 is used to take pictures for inspection;
[0044] In this embodiment, the structure of the XY linear module 12 is as follows: Figure 3 and Figure 4 As shown, it is composed of mutually perpendicular X beams 14 and Y beams 13. The Y beams 13 can be two parallel beams, which are erected across the gantry 16 on both sides of the pre-casting integrated inspection station 5. The X beam 14 can move forward and backward along the Y beam 13; the high-resolution camera 8 and the TOF camera 7 are installed together on the X beam 14 and can move left and right along the X beam 14. The stepper motor 15 for controlling the operation is provided on both the Y beam 13 and the X beam 14, and the operation of the stepper motor 15 is controlled by the industrial computer 1. Figure 4 As shown, the industrial computer 1 controls the stepper motor 15 to drive the XY linear module 12 to move in the XY direction, moves the Gigabit network high-resolution camera 8 and the TOF camera 7 to the specified position to take pictures, and then transmits the distance depth map and the high-resolution image to the industrial computer 1 through the HUB Gigabit network hub 2, and is analyzed and detected by the intelligent recognition and measurement software.
[0045] During the inspection, the inspection software obtains the inspection image and distance depth map from the camera set at each workstation, extracts the corresponding features after performing the front-end image denoising processing on the image, finds the position of the corresponding feature, calculates the size of the object at the corresponding position, and matches it with the corresponding parameters of the design drawing, and compares and determines whether its position and size are qualified. If there is any unqualified part, a warning signal is issued to the corresponding workstation to prompt the operator of the workstation to make corrections. After the correction work, check again until it is qualified, and the qualified result is displayed by the warning light 9.
[0046] The industrial computer 1 used for inspection and control in the present invention can also communicate with a host computer via Ethernet and connect to a production execution management MES system.
[0047] like Figure 5 As shown, the present invention also discloses an intelligent inspection method for an automatic production line of a building PC component factory, comprising the following steps:
[0048] 1. Obtain inspection-related parameter data from PC component design drawing files, that is, convert the user's CAD design file into the standard composite plate DXF format file required by this system (DXF is widely used in various CAD software and has become a de facto standard CAD format file).
[0049] 2. According to the production line execution plan sequence, obtain the parameter data of the shape, size and dimensions of the composite plate from the corresponding DXF file; obtain the parameter data of the size, size and position of the junction box; obtain the various parameter data of the horizontal and vertical number of the binding steel bars, length, steel bar spacing, reinforcement position, reinforcement length, truss beam reinforcement position, etc.
[0050] 3. Input the parameter data obtained above into the corresponding station inspection program module to prepare for the inspection of the mold installation station;
[0051] 4. Check the mold installation station 3:
[0052] After the mold is installed, the operator places the L-shaped origin coordinate position calibrator 10 at the origin of the mold installation, and then presses the completion button; then the high-resolution camera 8 takes the mold station image and inputs it into the industrial computer 1, allowing the intelligent image recognition and detection software to analyze and process it, including image denoising, feature extraction, mold installation origin and XY coordinate direction determination, and mold installation position size and drawing parameter data consistency check. If there is any discrepancy during the inspection, a warning signal is issued and the operator makes corrections; then press the button to check again until the mold is installed correctly and the inspection is qualified, which is displayed by the warning light 9.
[0053] 5. Check the junction box installation station 4:
[0054] After the junction box is installed, the operator presses the completion button, and the high-resolution camera 8 takes a picture of the junction box installation station image, which is input into the industrial computer 1, and the intelligent image recognition and detection software performs analysis and detection processing, including noise removal of the station image, extraction of junction box feature quantity, determination of mold origin and XY coordinate direction, and consistency check of junction box installation position size and drawing parameter data. If there is any discrepancy during the inspection, a warning signal is issued, and the operator makes corrections; then press the button to check again until the junction box is installed correctly and the inspection is qualified, which is displayed by the warning light 9.
[0055] 6. Enter the steel bar-embedded parts placement station 6, and place the bound steel mesh and embedded parts into the mold frame for installation.
[0056] 7. Carry out the inspection of the integrated inspection station 5 before pouring:
[0057] The industrial computer 1 controls the XY linear module 12 to move in the XY direction, moves the high-resolution camera 8 and the TOF camera 7 to the designated measuring point, and uses the TOF camera 7 to obtain the distance depth map of the object at the position on the mold platform and input it into the industrial computer 1; uses the high-resolution camera 8 to take the image of the position and input it into the industrial computer 1; continues to move the XY linear module 12 to move the high-resolution camera 8 and the TOF camera 7 to the next designated measuring point, and uses the TOF camera 7 to obtain the distance depth map of the object at the position on the mold platform and input it into the industrial computer 1; uses the high-resolution camera 8 to take the image of the position and input it into the industrial computer 1; Enter into industrial computer 1; move repeatedly until the distance depth map and image of all the specified position measurement points are obtained, and then the intelligent image recognition and detection software performs analysis and processing, including image denoising, relevant feature quantity extraction, mold installation origin and XY coordinate direction determination, installation of steel mesh vertical and horizontal steel bar number and spacing distance, reinforcement position length, truss beam reinforcement position, embedded parts position and other parameters and data consistency check with the drawings; if there is any discrepancy, a warning signal is issued, and the operator makes corrections, and then presses the button to check again until the installation is correct and all inspections are qualified.
[0058] 8. Enter the casting station and complete the casting of PC components.
[0059] In addition, it should be noted that Figure 6 As shown, the main contents of the intelligent image recognition and detection software of the present invention include:
[0060] (1) Obtain various inspection parameter data modules from the DXF file of the CAD design drawing. Its function is to find the relevant parameter area segment from the DXF format file, find the corresponding key parameters in the relevant area segment, read the corresponding parameters and save them in the output list.
[0061] (2) Mould installation position inspection module.
[0062] (3) Junction box position inspection module.
[0063] (4) Integrated inspection module before pouring.
[0064] (5) Camera shooting control image acquisition, XY execution group module movement control, warning light signal control and input and output control module.
[0065] (6) Mould installation origin and coordinate XY direction identification module.
[0066] Finally, it should be noted that the above is only a specific embodiment of the present invention. Obviously, the present invention can also have many variations, and all equivalent or simple changes made according to the structure, features and principles described in the patent concept of the present invention are included in the protection scope of the patent of the present invention. All variations that can be directly derived or associated with the content disclosed by ordinary technicians in this field should be considered as the protection scope of the present invention.
Claims
1. An intelligent inspection system for an automatic production line of a building PC component factory, comprising an industrial computer (1), a HUB gigabit network hub (2) and a plurality of high-resolution cameras (8) installed at each inspection station; characterized in that: The inspection stations at least include, in the order of process flow: a mold installation inspection station (3), a junction box installation inspection station (4), and a pre-casting integrated inspection station (5); The mold installation inspection station (3) has an L-shaped origin coordinate position calibrator (10) for calibrating the coordinate origin placed on its plane; the high-resolution camera (8) is installed above it; the image captured by the high-resolution camera (8) is input into the industrial computer (1) through the HUB gigabit network hub (2), and the qualified inspection result is displayed by the external warning light (9) through software inspection and analysis; The junction box installation inspection station (4) has the high-resolution camera (8) installed above it; the image captured by the high-resolution camera (8) is input into the industrial computer (1) through the HUB gigabit network hub (2), and is checked and analyzed by software, and the qualified inspection result is displayed by an external warning light (9); The pre-casting integrated inspection station (5) has an XY linear module (12) that can move vertically and horizontally mounted above it; the high-resolution camera (8) and the TOF camera (7) are installed together on the X beam (14) of the XY linear module (12) and can move left and right along the X beam (14); the X beam (14) can move forward and backward along the Y beam (13) of the XY linear module (12); the high-resolution camera (8) is used to capture the image position of the object, and the TOF camera (7) is used to capture the distance depth of the object; the image data captured by the two cameras are input into the industrial computer (1) through the HUB gigabit network hub (2), and are checked and analyzed by software, and the qualified inspection results are displayed by the external warning light (9); The L-shaped origin coordinate position calibrator (10) comprises a right-angled single-sided X-axis scale and a Y-axis scale; each axis scale is provided with a plurality of luminous bodies (11) for photographing and identifying; the luminous bodies (11) are arranged at a spacing of 150 mm; The Y straight beams (13) of the XY linear module (12) are two parallel beams, which are erected across the gantry (16) on both sides of the pre-casting integrated inspection station (5).
2. The intelligent inspection system for the automatic production line of the building PC component factory according to claim 1 is characterized in that: The Y straight beam (13) and the X cross beam (14) are both provided with a stepping motor (15) for controlling operation; the operation of the stepping motor (15) is directed and controlled by the industrial control computer (1).
3. The intelligent inspection system for the automatic production line of building PC component factory according to claim 1 is characterized in that: The industrial control computer (1) can also communicate with a host computer via Ethernet and connect to a production execution management MES system.
4. An inspection method based on the intelligent inspection system of the automatic production line of the building PC component factory according to claim 1, characterized in that: The following steps are involved: (1) Convert the user's CAD design file into a standard DXF file; (2) According to the production line execution plan sequence, obtain from the corresponding DXF files: a. Parameter data of the shape, size and dimensions of the composite plate; b. Parameter data of the size and location of the junction box; c. Parameter data of the horizontal and vertical number, length, spacing between steel bars, location of bars, length of bars, and location of truss beam bars of the tied steel bars; (3) All parameter data obtained in step (2) are input into the corresponding workstation inspection program modules to prepare for installation inspection; (4) Check the mold installation station; After the peripheral molds are installed, the L-shaped origin coordinate position calibrator is placed on the plane to set the origin coordinates, and then the high-resolution camera above is used to take pictures to form a mold station image that is input into the computer and analyzed and processed by intelligent image recognition and detection software. The content includes image denoising, feature extraction, mold installation origin and XY coordinate direction determination, and mold installation position size and drawing parameter data consistency check; if the inspection result is inconsistent, a warning signal is issued to prompt correction; and then check again until it is qualified; (5) Inspection of the junction box installation station: After the junction box is installed, it is photographed by a high-resolution camera above, and the resulting junction box workstation image is input into the computer and analyzed and processed by intelligent image recognition and detection software. The content includes image noise removal, feature extraction of the junction box, determination of the mold origin and XY coordinate direction, and consistency check of the junction box installation position size and drawing parameter data. If there is any discrepancy, a warning signal is issued and the operator makes corrections. Then the inspection is repeated again until it is qualified. (6) Place the tied steel mesh and embedded components into the mold frame for installation; (7) Inspection at the integrated inspection station before pouring: The stepper motor of the XY linear module is controlled by a computer, so that the high-resolution camera and the TOF camera are simultaneously moved along the X-axis and Y-axis of the XY linear module to the designated measuring point for shooting; the high-resolution camera obtains the image position of the object, and the TOF camera obtains the distance depth of the object, and the image data taken by the two cameras are transmitted to the computer for storage; Continue to move and control the high-resolution camera and TOF camera on the XY linear module until the image position map and distance depth map of all designated position measurement points are obtained; then the computer's intelligent image recognition and detection software will analyze and process; its content includes image denoising, relevant feature quantity extraction, mold installation origin and XY coordinate direction determination, vertical and horizontal steel bar number and spacing distance of steel mesh installation, steel bar position length, truss beam steel bar position, embedded parts position and corresponding parameter data consistency check of the drawings; if there is any discrepancy, a warning signal will be issued, and the operator will make corrections and check again until all installations are correct and all inspections are qualified; (8) Enter the casting station and complete the casting of PC components.
