Forge piece surface crack image recognition and automatic grinding system

Abstract

The utility model provides a forge piece surface crack image recognition and automatic grinding system which comprises a rack and a flaw detection terminal, the rack is provided with an operation table top, the operation table top is provided with a slide way, and the slide way is provided with a first clamping base and a second clamping base; a chuck is mounted on the clamping base; a magnetizing coil is arranged on the chuck; an ultraviolet lamp is further arranged at the upper part of the operation table board; an industrial camera is mounted at the side part of the rack; a grinding wheel is arranged on one side of the upper part of the operating table; the grinding wheel is connected with a grinding motor; and the single chip microcomputer controls the supply of the magnetic suspension. The industrial camera obtains image information in the flaw detection process and stores the image information in the memory; and the single chip microcomputer is connected with the flaw detection control key to obtain a flaw detection start-stop control instruction. According to the utility model, an image can be acquired, and the crack position in the flaw detection process can be identified by learning the crack image in the artificially marked crack area through the support vector machine. And automatic cleaning and grinding can be achieved, and the forge piece detection efficiency is improved.

Description

Technical Field

[0001] The utility model belongs to the field of magnetic particle flaw detection technology, in particular to a forging surface crack image recognition and automatic grinding cleaning system. Background Art

[0002] Forging products have the advantages of fine grains, dense organization, continuous streamlines, good strength, toughness and other mechanical properties. They are mostly used in automobiles that need to withstand complex full-torsion loads and certain impact loads, such as crankshafts, connecting rods, front axles, steering knuckles, gears, etc. In the forging process, due to the segregation of raw material components and imperfect forging technology, defects such as folding and cracking will inevitably occur, resulting in product failure during manufacturing and use, causing losses.

[0003] Generally, fluorescent magnetic particle inspection is used to inspect forgings. After the forgings are magnetized, the interaction between the leakage magnetic field at the defect and the fluorescent magnetic particles is used to identify and locate the defects. The current fluorescent magnetic particle inspection process cannot obtain images of forgings, nor can it perform the polishing function after inspection. If images need to be captured, the forgings need to be moved to other locations to take the images, which brings inconvenience to the inspection and reduces the inspection efficiency. Utility Model Content

[0004] The utility model provides a forging surface crack image recognition and automatic grinding cleaning system. The system can obtain forging image information during the flaw detection process and can automatically recognize the forging information, thereby facilitating operation and improving detection efficiency.

[0005] The forging surface crack image recognition and automatic grinding cleaning system comprises: a stand and a flaw detection terminal, the stand is provided with an operating table, the operating table is provided with a slideway, one end of the slideway is fixed with a first clamping base, and the slideway is slidably connected with a second clamping base;

[0006] The first clamping base and the second clamping base are respectively provided with chucks; the chucks are provided with magnetizing coils; and the chucks are connected with driving motors;

[0007] A support crossbar is installed on the stand, and the support crossbar is arranged on the upper part of the operating table. A plurality of nozzles are installed on the support crossbar, and the ends of the nozzles are connected to magnetic suspension nozzles.

[0008] An ultraviolet lamp is also installed on the top of the operating table; an industrial camera is installed on the side of the table;

[0009] A grinding wheel is arranged on one side of the upper part of the operating table; the grinding wheel is connected to a grinding motor;

[0010] The flaw detection terminal is internally provided with a single chip microcomputer, a motor control circuit, a grinding motor control circuit, a magnetic suspension liquid control circuit and a memory;

[0011] The flaw detection terminal is equipped with a display touch screen, a rotation control button, a grinding control button and a flaw detection control button;

[0012] The single chip microcomputer is connected to the driving motor through the motor control circuit, obtains the rotation control instruction through the rotation control button, and controls the operation of the driving motor;

[0013] The single chip microcomputer is connected to the grinding motor through the grinding motor control circuit, obtains the grinding control instruction through the grinding control button, and controls the operation of the grinding motor;

[0014] The single chip microcomputer is connected to the magnetic suspension driving motor through the suspension control circuit to control the operation of the magnetic suspension driving motor;

[0015] The single-chip microcomputer obtains the image information of the flaw detection process by connecting with the industrial camera and saves it to the memory. It also displays the flaw detection process information and the captured image information through the display touch screen.

[0016] The single chip microcomputer is connected to the flaw detection control button to obtain the flaw detection start and stop control instructions.

[0017] It should be further explained that the chuck is provided with claws.

[0018] It should be further explained that vertical support rods are respectively installed at the four corners of the operating table, a top cover is installed on the top of the vertical support rods, and an ultraviolet lamp is installed at the bottom of the top cover and faces the detection position between the first clamping base and the second clamping base.

[0019] The UV lamp is fixed to the bottom of the top cover by a retractable cylinder.

[0020] It should be further explained that both ends of the supporting cross bar are respectively connected to the vertical supporting bars.

[0021] Both ends of the supporting crossbar are connected to the vertical supporting rod through three-way sleeves;

[0022] A through hole is arranged on the vertical support rod, and the three-way sleeve is fixed to the vertical support rod by bolts matching with the through hole.

[0023] It should be further explained that legs are provided at the bottom of the stand.

[0024] It should be further explained that at least two industrial cameras are arranged between the first clamping base and the second clamping base.

[0025] It should be further explained that a communication module is also provided inside the flaw detection terminal;

[0026] The single chip microcomputer is connected with the communication module to upload the flaw detection information and the image information of the flaw detection process to the host computer.

[0027] It should be further explained that the memory stores a support vector machine algorithm and a breadth-first search image recognition algorithm.

[0028] It can be seen from the above technical solutions that the utility model has the following advantages:

[0029] The utility model provides a forging surface crack image recognition and automatic grinding cleaning system. The system can obtain forging image information during the flaw detection process and can automatically recognize forging information, thereby facilitating operation and improving detection efficiency.

[0030] The forging surface crack image recognition and automatic grinding and cleaning system of the utility model can be used in the surface magnetic particle flaw detection process of forgings such as steering knuckles, differential covers, front axles, etc. It can automatically take images of forgings and realize automatic recognition. The system has a grinding wheel that can grind the forgings according to usage needs, thereby improving detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solution of the utility model, the drawings required for use in the description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the utility model. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 The schematic diagram of the forging surface crack image recognition and automatic grinding system;

[0033] Figure 2 This is a schematic diagram of an embodiment of a flaw detection terminal;

[0034] Figure 3 Schematic diagram of steering knuckle mesh division;

[0035] Figure 4 Schematic diagram of the crack in the steering knuckle;

[0036] Figure 5 This is a schematic diagram after the cracks on the steering knuckle have been cleaned and ground. DETAILED DESCRIPTION

[0037] In order to make the purpose, features and advantages of the utility model more obvious and easy to understand, the technical scheme of the utility model will be clearly and completely described below in combination with the drawings in the specific embodiment. Obviously, the embodiments described below are only part of the embodiments of the utility model, not all of them. Based on the embodiments in this patent, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this patent.

[0038] like Figure 1 and Figure 2As shown, the forging surface crack image recognition and automatic grinding cleaning system provided by the utility model includes: a stand 1 and a flaw detection terminal. The flaw detection terminal can be a handheld terminal device or an operating table terminal device installed on the side of the stand 1.

[0039] The stand 1 is provided with an operating table 2, and vertical support rods are respectively installed at the four corners of the operating table 2, and a top cover 3 is installed on the top of the vertical support rods. A support leg 4 is provided at the bottom of the stand 1.

[0040] The upper part of the operating table 2 of the utility model is provided with an ultraviolet lamp 5; the side of the stand 1 is provided with an industrial camera 6. In order to meet the requirements for image processing and realize multi-position image acquisition, at least two industrial cameras 6 are provided between the first clamping base 7 and the second clamping base 8.

[0041] A grinding wheel 9 is provided on one side of the upper part of the operating table 2; the grinding wheel 9 is connected to a grinding motor. The grinding wheel can be mounted on a grinding wheel bracket, and the grinding wheel bracket can adjust the height and horizontal position. In order to meet the needs of different forging shapes, users can use a manual grinding wheel for grinding.

[0042] The ultraviolet lamp 5 is installed at the bottom of the top cover 3 and faces the detection position between the first clamping base 7 and the second clamping base 8. The ultraviolet lamp 5 can also be fixed to the bottom of the top cover 3 through a retractable cylinder to achieve height adjustment.

[0043] The operating table 2 is provided with a slideway, a first clamping base 7 is fixed at one end of the slideway, and a second clamping base 8 is slidably connected to the slideway. A chuck is respectively installed on the first clamping base 7 and the second clamping base 8; a claw is provided on the chuck. A magnetizing coil 10 is provided on the chuck; a driving motor is connected to the chuck. The sliding mode of the second clamping base 8 can be manual movement, or horizontal sliding can be achieved by a motor or a hydraulic push rod.

[0044] The first clamping base 7 can clamp the first end of the forging, and the second clamping base 8 can clamp the second end of the forging. A support crossbar is installed on the stand 1, and the support crossbar 12 is arranged on the upper part of the operating table 2. A plurality of nozzles are installed on the support crossbar 12, and a magnetic suspension nozzle 11 is connected to the end of the nozzle. The nozzle can be connected to the magnetic suspension reservoir through a magnetic suspension drive motor. The two ends of the support crossbar are respectively connected to the vertical support rod. In order to achieve height adjustment, the two ends of the support crossbar are respectively connected to the vertical support rod through a three-way sleeve; the vertical support rod is provided with a through hole, and the three-way sleeve is fixed to the vertical support rod by bolts and the through hole. In this way, the height of the support crossbar can be adjusted according to actual needs.

[0045] The flaw detection terminal is internally provided with a single chip microcomputer, a motor control circuit, a grinding motor control circuit, a magnetic suspension liquid control circuit and a memory.

[0046] The flaw detection terminal is provided with a display touch screen, a rotation control button, a grinding control button and a flaw detection control button; the single-chip microcomputer is connected to the drive motor through the motor control circuit, obtains the rotation control instruction through the rotation control button, and controls the operation of the drive motor; the single-chip microcomputer is connected to the grinding motor through the grinding motor control circuit, obtains the grinding control instruction through the grinding control button, and controls the operation of the grinding motor; the single-chip microcomputer is connected to the magnetic suspension drive motor through the suspension control circuit to control the operation of the magnetic suspension drive motor; the single-chip microcomputer is connected to the industrial camera 6 to obtain the image information of the flaw detection process and save it to the memory, and also displays the flaw detection process information and the captured image information through the display touch screen; the single-chip microcomputer is connected to the flaw detection control button to obtain the flaw detection start and stop control instructions.

[0047] In this way, the operator can control the flaw detection process of the forging surface through the display touch screen, rotation control button, grinding control button and flaw detection control button. In order to facilitate the monitoring personnel to understand the flaw detection process and data, a communication module is also set inside the flaw detection terminal; the single-chip microcomputer is connected to the communication module to upload the flaw detection information and the image information of the flaw detection process to the host computer.

[0048] In the utility model, a support vector machine algorithm and a breadth-first search image recognition algorithm are stored in the memory. The acquired forging image can be recognized and analyzed by the support vector machine algorithm and the breadth-first search image recognition algorithm to obtain the flaw detection result information.

[0049] For example, this embodiment takes the flaw detection and image analysis of a steering knuckle as an example. The steering knuckle is placed on a first clamping base and a second clamping base, and a flaw detection process is performed. At the same time, an industrial camera can acquire image information.

[0050] The MCU can be based on the machine learning part and process it through the training set in the memory. Figure 3 As shown, the knuckle surface can be divided into several uniform grids, in this case the grid side length is 3mm, then the grid positions are numbered, and the spatial coordinate value of each grid is established. The flaw detector identified the cracks on the knuckle surface after magnetization and spraying the suspension, and circled the cracks with white chalk.

[0051] After taking a photo of the knuckle surface with an industrial camera, the image within the white mark is automatically captured, and the grid position and number covered by the image are recorded to collect the image set for machine training. The image processing involves the following process:

[0052] 1.1 Feature Vector Learning

[0053] The image set data obtained by single chip processing. Firstly, the image is grayed, and the discrete points with drastic changes in gray value in the image are removed by weighted directed smoothing filter algorithm to achieve image noise reduction; and the gray-mean co-occurrence matrix is ​​used to achieve maximum entropy segmentation. The support vector machine algorithm is used to detect and analyze the cracks to form feature vectors.

[0054] 1.2 Typical defect location matching

[0055] The single chip microcomputer matches the feature vectors at different positions with the feature vector positions recorded in 1.1 to form a defect recognition matrix, and gives priority to recognition; for images outside the acquisition positions of the training image set, supplementary recognition is performed.

[0056] 1.3 Image acquisition and processing

[0057] Add printing and identification of forging serial number (QR code) on the surface of forging.

[0058] Use a robotic arm to place forgings in three fixed positions and take photos. After completion, the photos are processed. First, the image is grayed out, and the discrete points with drastic changes in grayscale values ​​in the image are removed through a weighted directed smoothing filter algorithm to achieve image noise reduction; and the gray-mean co-occurrence matrix is ​​used to achieve maximum entropy segmentation.

[0059] 1.4 Crack Identification

[0060] Based on the eigenvector of 1.2, crack identification is performed on the defect marked area and the unmarked area. If there is a crack, the grid number of the area where the crack exists is recorded.

[0061] The automatic grinding part of the utility model can also perform path planning based on point positions, summarize the grid numbers of cracked grids and their corresponding spatial coordinate values, and connect the coordinates of two adjacent points with a distance of no more than 3 mm to form a crack line segment.

[0062] According to the provisions of GB / T 12361 and GB / T 12362 on forging surface defects, the grinding width should not be less than 6 times the depth, and the length should exceed the defect length by more than 3mm at both ends. The grinding depth of the machined surface is allowed to be half of the machining allowance, and the grinding depth of the non-machined surface is allowed to be 1 / 3 of the thickness tolerance. The machining allowance of this steering knuckle is 3mm, and the thickness tolerance is 3.2mm (+2.1mm, -1.1mm).

[0063] The grinding parameters are:

[0064]

[0065] The grinding starting point is: along the crack extension direction, extending about 3mm from the starting point;

[0066] The grinding end point is: along the crack extension direction, extending about 3mm from the end point;

[0067] by Figure 4 Take the knuckle crack in the figure as an example. Assume that the crack length is 16 mm and involves 7 grid elements with coordinates (x 1 ,y 1 ,z 1 ), (x 2 ,y 2 ,z 2 ) ... (x 7 ,y 7 ,z 7 ). Then the crack trajectory is to connect these 7 points.

[0068] In this embodiment, two grinding wheels with a length of 5 mm and a diameter of 6 mm are spliced ​​together for grinding. Before grinding, the outer dimensions of the grinding wheels are measured, and the grinding wheel track is compensated according to the results.

[0069] The grinding wheel moves along the crack track. After the cylindrical surface is tangent to the crack surface, the grinding amount of the machined surface / non-machined surface is 1.5 / 1mm.

[0070] like Figure 5 As shown in the figure, after the automatic grinding is completed, the forging will be identified for cracks again. This time, the crack identification will focus on the crack area for photography and identification based on the forging number and the grid number with cracks recorded in the previous period. If there are still cracks, they will be isolated for review and scrapped.

[0071] It should be understood that when an element or layer is referred to as being "on", "connected" or "coupled" to another element or layer, it may be directly on, directly connected or coupled to the other element or layer, or there may be intermediate elements or layers. Conversely, when an element is referred to as being "directly on", "directly connected" or "directly coupled" to another element or layer, there are no intermediate elements or layers. Similar numbers in all figures indicate similar elements. As used herein, the term "and / or" includes any and all combinations of one or more of the relevant listed items.

[0072] The technology described in the forging surface crack image recognition and automatic grinding cleaning system provided by the utility model can be implemented in hardware, software, firmware or any combination thereof. The various features described as modules, units or components can be implemented together in an integrated logic device or separated as discrete but interoperable logic devices or other hardware devices. In some cases, various features of the electronic circuit can be implemented as one or more integrated circuit devices, such as integrated circuit chips or chipsets.

[0073] If implemented in hardware, the present invention relates to a device, such as a processor or an integrated circuit device, such as an integrated circuit chip or chipset. Alternatively or additionally, if implemented in software or firmware, the technology may be implemented at least in part by a computer-readable data storage medium, including instructions that, when executed, cause a processor to perform one or more of the above methods. For example, a computer-readable data storage medium may store instructions such as those executed by a processor.

[0074] The single-chip microcomputer may include one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other integrated circuits, or discrete logic circuits. Therefore, the term "processor" as used herein may refer to any of the aforementioned structures or any other structures that are more suitable for implementing the technology described herein. Specifically, the single-chip microcomputer uses an STM32F051C8T6 single-chip microcomputer, or an ARM microprocessor.

[0075] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present utility model and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the utility model described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions.

[0076] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A forging surface crack image recognition and automatic grinding system, characterized in that: include: The stand and the flaw detection terminal are provided with an operating table on which a slide is provided, a first clamping base is fixed at one end of the slide, and a second clamping base is slidably connected to the slide; The first clamping base and the second clamping base are respectively provided with chucks; the chucks are provided with magnetizing coils; and the chucks are connected with driving motors; A support crossbar is installed on the stand, and the support crossbar is arranged on the upper part of the operating table. A plurality of nozzles are installed on the support crossbar, and the ends of the nozzles are connected to magnetic suspension nozzles. An ultraviolet lamp is also installed on the top of the operating table; an industrial camera is installed on the side of the table; A grinding wheel is arranged on one side of the upper part of the operating table; the grinding wheel is connected to a grinding motor; The flaw detection terminal is internally provided with a single chip microcomputer, a motor control circuit, a grinding motor control circuit, a magnetic suspension liquid control circuit and a memory; The flaw detection terminal is equipped with a display touch screen, a rotation control button, a grinding control button and a flaw detection control button; The single chip microcomputer is connected to the driving motor through the motor control circuit, obtains the rotation control instruction through the rotation control button, and controls the operation of the driving motor; The single chip microcomputer is connected to the grinding motor through the grinding motor control circuit, obtains the grinding control instruction through the grinding control button, and controls the operation of the grinding motor; The single chip microcomputer is connected to the magnetic suspension driving motor through the suspension control circuit to control the operation of the magnetic suspension driving motor; The single-chip microcomputer obtains the image information of the flaw detection process by connecting to the industrial camera and saves it to the memory. It also displays the flaw detection process information and the captured image information through the display touch screen; The single chip microcomputer is connected to the flaw detection control button to obtain the flaw detection start and stop control instructions.

2. The forging surface crack image recognition and automatic grinding system according to claim 1 is characterized in that: The chuck is provided with claws.

3. The forging surface crack image recognition and automatic grinding system according to claim 1, characterized in that: The four corners of the operating table are respectively equipped with vertical support rods, the top of the vertical support rods is equipped with a top cover, and the ultraviolet lamp is installed at the bottom of the top cover and faces the detection position between the first clamping base and the second clamping base.

4. The forging surface crack image recognition and automatic grinding cleaning system according to claim 3 is characterized in that: The UV lamp is fixed to the bottom of the top cover by a retractable cylinder.

5. The forging surface crack image recognition and automatic grinding system according to claim 3 is characterized in that: Both ends of the supporting cross bar are respectively connected to the vertical supporting bars.

6. The forging surface crack image recognition and automatic grinding cleaning system according to claim 5 is characterized in that: Both ends of the supporting crossbar are connected to the vertical supporting rod through three-way sleeves; A through hole is arranged on the vertical support rod, and the three-way sleeve is fixed to the vertical support rod by bolts matching with the through hole.

7. The forging surface crack image recognition and automatic grinding cleaning system according to claim 3 is characterized in that: The bottom of the stand is provided with supporting legs.

8. The forging surface crack image recognition and automatic grinding cleaning system according to claim 1 or 2, characterized in that: At least two industrial cameras are arranged between the first clamping base and the second clamping base.

9. The forging surface crack image recognition and automatic grinding cleaning system according to claim 1 or 2, characterized in that: A communication module is also provided inside the flaw detection terminal; The single chip microcomputer is connected with the communication module to upload the flaw detection information and the image information of the flaw detection process to the host computer.

10. The forging surface crack image recognition and automatic grinding cleaning system according to claim 1 or 2, characterized in that: The memory stores a support vector machine algorithm and a breadth-first search image recognition algorithm.

Images