CN113430679B - Production platform for identifying abnormality of carbon fiber in pre-oxidation furnace - Google Patents
Production platform for identifying abnormality of carbon fiber in pre-oxidation furnace Download PDFInfo
- Publication number
- CN113430679B CN113430679B CN202110985110.8A CN202110985110A CN113430679B CN 113430679 B CN113430679 B CN 113430679B CN 202110985110 A CN202110985110 A CN 202110985110A CN 113430679 B CN113430679 B CN 113430679B
- Authority
- CN
- China
- Prior art keywords
- carbon fiber
- image
- oxidation furnace
- light source
- abnormal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- D—TEXTILES; PAPER
- D01—NATURAL OR MAN-MADE THREADS OR FIBRES; SPINNING
- D01F—CHEMICAL FEATURES IN THE MANUFACTURE OF ARTIFICIAL FILAMENTS, THREADS, FIBRES, BRISTLES OR RIBBONS; APPARATUS SPECIALLY ADAPTED FOR THE MANUFACTURE OF CARBON FILAMENTS
- D01F9/00—Artificial filaments or the like of other substances; Manufacture thereof; Apparatus specially adapted for the manufacture of carbon filaments
- D01F9/08—Artificial filaments or the like of other substances; Manufacture thereof; Apparatus specially adapted for the manufacture of carbon filaments of inorganic material
- D01F9/12—Carbon filaments; Apparatus specially adapted for the manufacture thereof
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Abstract
The invention discloses a production platform for identifying abnormality of carbon fiber in a pre-oxidation furnace, which comprises: preoxidation stove, carbon fiber silk bundle, drafting arrangement, the platform still includes: gather carbon fiber image system, discernment carbon fiber unusual system and clear away the unusual device of carbon fiber, wherein: the system for acquiring carbon fiber images comprises: the system comprises an image acquisition mechanism, a light source supply mechanism, an image preprocessing module and an industrial personal computer; the carbon fiber abnormity identification system judges the abnormity of the carbon fiber image output by the carbon fiber image acquisition system and transmits the abnormal position information of the carbon fiber tow image to the carbon fiber abnormity clearing device; the carbon fiber abnormal clearing device carries out shearing treatment on the carbon fiber tows according to the position where the carbon fiber tow image is abnormal; the platform can accurately position the abnormal positions of broken filaments and jumping filaments of the carbon fiber tows in a pre-oxidation furnace with serious atomization; meanwhile, the broken and jumping carbon fiber tows can be quickly eliminated, and the production quality of the carbon fiber tows is ensured.
Description
Technical Field
The invention is mainly applied to the technical field of fiber detection, and particularly relates to a production platform for identifying abnormality of carbon fibers in a pre-oxidation furnace.
Background
Carbon fiber as an advanced PAN-based composite material has been widely applied to various fields such as aerospace, buildings, vehicles, wind power and the like. The carbon fiber has the advantages of high modulus, high strength, high temperature resistance, corrosion resistance, electric conduction, heat conduction and the like, is divided into two parts, has weak points, belongs to a brittle material, is easy to generate broken filaments or broken filaments due to fiber fracture in the production process, particularly in the pre-oxidation and high-low temperature carbonization processes, and can cause the reduction of quality and performance in the preparation process.
In the pre-oxidation production process, the fibers can be broken under the action of internal stress of the protofilaments, so that broken filaments are generated. During the pre-oxidation process, the fiber passes through various roller bodies such as a plurality of active drafting rollers, guide rollers and the like. Repeated contact with the roller body can cause damage to the surface of the fiber, and the produced broken filaments are easy to wind the roller if not found in time, so that the production stability and the quality of carbon fiber products are influenced. Particularly, in the pre-oxidation process of the carbon fiber, the reaction time is long, the drafting distance is long, and the broken filaments or filament breakage phenomenon of tows is easy to generate, so that the monitoring and timely treatment of the broken filaments of the carbon fiber are very important in the pre-oxidation process.
In the carbonization production of carbon fiber, the number of running fiber tows is large, the space between the tows is small, especially the fiber tows come and go in multiple layers in an oxidation furnace, if the fiber is broken in the running process, production personnel cannot handle the fiber timely, large-area broken fiber winding is easily caused, the stable production is seriously influenced, and in order to find out the broken fiber condition of production in time, a broken fiber detection device needs to be arranged right below the running outlet of each layer of fiber. The quality of the carbon fiber is guaranteed, broken filaments generated in the process are reduced and are rolled into a finished shaft, and the broken filaments generated in the process need to be monitored.
Therefore, a tow monitoring system needs to be installed on the production equipment to monitor the state of the carbon fiber tows in real time.
At present, carbon fiber tow detection equipment is mainly provided with the following detection modes according to different principles:
(1) the sensor detection technology comprises the following steps: the light source irradiates on materials with different properties and can reflect different brightness, and the sensor can judge broken filaments or broken filament fibers in the filament bundle according to the characteristic, so that a command of removing the broken filaments is sent. This method requires a sensor to be installed on each carbon fiber tow layer in each pre-oxidation furnace, and thus the cost investment is high.
(2) Photoelectric detection technology: the color difference exists between the carbon fiber and the broken filament fiber, the color difference is reflected on the photodiode to become current difference, and then the current difference is subjected to signal amplification and processing to identify the broken filament fiber. The method is simple in principle and is mainly applied to early-stage broken filament fiber detection equipment. However, the service life of the photoelectric tube is short, the stability is poor, and the identification effect of the technology on the small foreign fiber is poor.
Carbon fiber as an advanced PAN-based composite material has been widely applied to various fields such as aerospace, buildings, vehicles, wind power and the like. The carbon fiber has the advantages of high modulus, high strength, high temperature resistance, corrosion resistance, electric conduction, heat conduction and the like, is divided into two parts, has weak points, belongs to a brittle material, is easy to generate broken filaments or broken filaments due to fiber fracture in the production process, particularly in the pre-oxidation and high-low temperature carbonization processes, and can cause the reduction of quality and performance in the preparation process.
In the pre-oxidation production process, the fibers can be broken under the action of internal stress of the protofilaments, so that broken filaments are generated. During the pre-oxidation process, the fiber passes through various roller bodies such as a plurality of active drafting rollers, guide rollers and the like. Repeated contact with the roller body can cause damage to the surface of the fiber, and the produced broken filaments are easy to wind the roller if not found in time, so that the production stability and the quality of carbon fiber products are influenced. Therefore, monitoring and timely processing of carbon fiber broken filaments are very important in the pre-oxidation process.
In the carbonization production of carbon fiber, the number of running fiber tows is large, the space between the tows is small, especially the fiber tows come and go in multiple layers in an oxidation furnace, if the fiber is broken in the running process, production personnel cannot handle the fiber timely, large-area broken fiber winding is easily caused, the stable production is seriously influenced, and in order to find out the broken fiber condition of production in time, a broken fiber detection device needs to be arranged right below the running outlet of each layer of fiber. The quality of the carbon fiber is guaranteed, broken filaments generated in the process are reduced and are rolled into a finished shaft, and the broken filaments generated in the process need to be monitored.
Disclosure of Invention
In order to solve the technical problems, the invention provides a production platform for identifying the abnormality of carbon fibers in a pre-oxidation furnace, which can accurately position the abnormal positions of broken and jumped carbon fiber tows in the pre-oxidation furnace with serious atomization, and can quickly eliminate the broken and jumped carbon fiber tows, thereby ensuring the production quality of the carbon fiber tows and meeting the requirements of the market on high-quality products.
The invention is implemented by adopting the following technical scheme:
a production platform for identifying anomalies in carbon fiber within a pre-oxidation furnace, the platform comprising: preoxidation stove, carbon fiber silk bundle, drafting arrangement, the platform still includes: gather carbon fiber image system, discernment carbon fiber unusual system and clear away the unusual device of carbon fiber, wherein:
the system for acquiring carbon fiber images comprises: the device comprises an image acquisition mechanism, a light source supply mechanism and an image preprocessing module; the light source supply mechanism is used for illuminating the tows which are drawn into a single-layer flat linear shape in the pre-oxidation furnace,
the image acquisition mechanism scans tows in the flat channel, and the image preprocessing module outputs the scanned image to a carbon fiber abnormality identification system by using a digital signal of 12 bits per pixel;
the carbon fiber abnormity identification system judges the abnormity of the carbon fiber image output by the carbon fiber image acquisition system and transmits the abnormal position information of the carbon fiber tow image to the carbon fiber abnormity clearing device;
the carbon fiber abnormal clearing device carries out shearing treatment on the carbon fiber tows according to the position where the carbon fiber tow image is abnormal; wherein:
an image acquisition mechanism is arranged on one side of the inner wall of the pre-oxidation furnace, and a light source supply mechanism is arranged on the inner wall of the other side of the pre-oxidation furnace; the image acquisition mechanism is connected with the inner wall of the pre-oxidation furnace through a reflecting plate at an included angle of 20-26 degrees; the light source supply mechanism is connected with the inner wall of the pre-oxidation furnace at an included angle of 21-24 degrees; the distance between the image acquisition mechanism and the carbon fiber tows is 16 mm-20 mm.
Further, the image acquisition mechanism consists of an air source spray head, a transparent assembly, a focusing component, a stretching mechanism, a camera, a protective sleeve and a flange; the protective sleeve is provided with a focusing component and is connected with the inner wall of the pre-oxidation furnace through a flange; the focusing component is connected with the camera through a stretching mechanism; the focusing component is provided with a transparent component with an air source nozzle; the stretching mechanism consists of a gear, a stepping motor and a driving gear; one end of the gear is connected with the transparent component; the other end of the stepping motor is connected with the stepping motor, and the stepping motor is connected with the driving gear through a transmission shaft.
Further, the light source supply mechanism adopts two combined light sources, namely a combination of a sunlight light source and an ultraviolet light source, and the ultraviolet light source is arranged below the sunlight light source; the sunlight light source is an LED lamp strip, and the luminous flux of the sunlight light source is 9000 lm; the ultraviolet light source is an ultraviolet lamp with the intensity of 80 mu W/cm2。
Further, the image acquisition mechanism is provided with a positive pressure air blowing component capable of cleaning the lens of the camera
Further, the system for identifying the carbon fiber abnormity sends the abnormal position of the carbon fiber tow image to a device for clearing the carbon fiber abnormity through the following steps of:
s101, carrying out defogging treatment on the carbon fiber tow image through the following formula to produce a first carbon fiber tow image;
I(x)＝J(x)t(x)+A(1-t(x))
wherein: i (x) -images to be dehazed; j (x) -fog free image; a-background light component; t (x) -refractive index
S102, generating a second fiber tow image for the first carbon fiber tow image through a T-SNE algorithm;
s103, extracting texture features of the second fiber tow image through a GLCM gray level co-occurrence matrix to obtain an abnormal carbon fiber tow image;
s104, judging broken filaments of abnormal parts in the abnormal carbon fiber tow image through an SVM algorithm, and if the broken filaments exist, sending broken filament position information to a carbon fiber abnormality clearing device; otherwise, returning to the step S101.
Further, the carbon fiber abnormity removing device comprises a walking guide rail, two actuating mechanisms, a foreign matter removing table, a fan and a waste fiber collecting box, wherein the fan and the waste fiber collecting box are respectively arranged on the foreign matter removing table; the actuating mechanism is even arranged on the walking guide rail; every actuating mechanism comprises positioning gear, switching-over part, step motor, cylinder, solenoid valve, cleaing away sword, speed reducer, wherein:
one end of the positioning gear is connected with the walking guide rail, and the other end of the positioning gear is connected with the speed reducer; the speed reducer is connected with the clearing knife through the stepping motor and the air cylinder in sequence; a reversing component is arranged between the cylinder and the stepping motor; the cylinder is also provided with an electromagnetic valve.
Furthermore, the carbon fiber image collecting system is respectively arranged on each layer of carbon fiber tows in the pre-oxidation furnace, and the front end of the image collecting mechanism is opposite to the moving direction of the carbon fiber tows.
Advantageous effects
1. According to the invention, the carbon fiber tow image in the pre-oxidation furnace with serious atomization is accurately captured by the carbon fiber image acquisition system, and especially the image acquisition mechanism, the light source supply mechanism and the image preprocessing module are organically matched, so that the influence of the deteriorated production environment in the pre-oxidation furnace on the image capture can be overcome, and the guarantee is provided for clearing the abnormal state on the carbon fiber tow in the later period.
2. The method has the advantages that the carbon fiber bundle image is subjected to feature extraction through the carbon fiber abnormality recognition system, the abnormal state of the carbon fiber operation is recognized in an online mode, and the abnormal state is accurately positioned and quickly responded; and finally, the quality of the carbon fiber tows is evaluated, and the quality of finished carbon fiber tows is guaranteed.
3. According to the device for clearing the abnormal carbon fiber tows, the abnormal carbon fiber tows can be cleared accurately through the carbon fiber exception clearing device and the executing mechanism, and the problems that procedures for clearing the abnormal carbon fiber tows are complex and efficiency is low in the prior art are solved.
Drawings
FIG. 1 is a schematic structural view of a production platform for identifying carbon fiber anomalies in a pre-oxidation furnace according to the present invention;
FIG. 2 is a schematic structural diagram of a system for collecting images of carbon fibers in a production platform for identifying anomalies in carbon fibers in a pre-oxidation furnace according to the present invention;
FIG. 3 is a schematic structural diagram of an image acquisition mechanism in a production platform for identifying abnormality of carbon fiber in a pre-oxidation furnace according to the present invention;
FIG. 4 is a schematic diagram of the distribution of a system for collecting images of carbon fibers in a production platform for identifying anomalies in carbon fibers in a pre-oxidation furnace according to the present invention;
FIG. 5 is a flow chart of a system for identifying carbon fiber anomalies in a production platform for identifying carbon fiber anomalies in a pre-oxidation furnace according to the present invention;
FIG. 6 is a schematic structural diagram of a device for removing carbon fiber anomalies in a production platform for identifying carbon fiber anomalies in a pre-oxidation furnace according to the invention.
Reference numerals
801-preoxidation furnace 802-carbon fiber bundle 803-drafting device 100-carbon fiber image acquisition system 200-carbon fiber abnormity identification system 300-carbon fiber abnormity clearing device
101-image acquisition mechanism 102-light source supply mechanism 103-image preprocessing module
104-reflector 105-LED lamp strip 106-ultraviolet lamp 107-air source spray head
108-transparent component 109-rack 110-motor drive gear 111-stepping motor
112-focusing member 113-camera 114-protective sleeve 115-flange
116-Positive pressure blow-up 301-guide, (302a,302b) -actuator
303-foreign matter removing table 304-fan 305-waste fiber collecting box 306-positioning gear
307-reversing component 308-stepping motor 309-air cylinder 310-electromagnetic valve 311-clearing knife
312-speed reducer.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the following detailed discussion of the present invention will be made with reference to the accompanying drawings and examples, which are only illustrative and not limiting, and the scope of the present invention is not limited thereby.
As shown in fig. 1 and 2, the present invention provides a production platform for identifying abnormality of carbon fiber in a pre-oxidation furnace, the platform comprising: a pre-oxidation furnace 801, a carbon fiber tow 802, a drawing device 803; the carbon fiber tow 802 passes through a pre-oxidation furnace 801, and drafting devices 803 are respectively connected to both ends of the carbon fiber tow 802. Under the drive of the drafting device 803, the fully expanded tows enter a pre-oxidation furnace at a speed of 10m/s, a central flat pre-oxidation channel formed by the carbon fiber tows 802 has a width of about 2.5m generally, and the detection precision requirement is 0.5 mm. As shown in fig. 1, the platform further comprises: gather carbon fiber image system 100, discern unusual system 200 of carbon fiber and clear away unusual device 300 of carbon fiber, wherein:
the system 100 for acquiring carbon fiber images includes: the system comprises an image acquisition mechanism 101, a light source replenishment mechanism 102 and an image preprocessing module 103; the light source supplying mechanism 102 is used for illuminating the filament bundle which is drawn into a single-layer flat linear shape in the pre-oxidation furnace; the image acquisition mechanism 101 scans tows in the flat channel, and the image preprocessing module 103 outputs the scanned image to a carbon fiber abnormality identification system by using a digital signal of 12 bits per pixel.
Identifying carbon fiber anomalies system 200: the carbon fiber abnormity identification system is applied to an industrial personal computer and used for transmitting position information of carbon fiber tow defects to the carbon fiber abnormity clearing device. The system for identifying the carbon fiber abnormity analyzes the image to judge whether broken filaments and broken filaments exist in the carbon fiber tows. And if the foreign fiber exists, calculating the position of the foreign fiber in the image.
Carbon fiber anomaly removal device 300: the carbon fiber abnormity clearing device converts position information calculated in the carbon fiber abnormity identifying system into the running linear speed and the line process of the drafting mechanism, when a defective tow runs to the position interval range of the clearing device, the clearing cutter runs along the running track, the stepping motor drives the commutator and the reducer to drive the positioning gear to run, and the position interval of shearing is positioned.
As shown in fig. 2, an image collecting mechanism 101 is disposed on one side of the inner wall of the pre-oxidation furnace, and a light source replenishing mechanism 102 is disposed on the inner wall of the other side of the inner wall of the pre-oxidation furnace; the image acquisition mechanism 101 is connected with the inner wall of the pre-oxidation furnace 801 at an included angle of 25 degrees through a reflector 104; the distance between the image acquisition mechanism 101 and the incident carbon fiber tows 802 is 16 mm-20 mm. The light source supply mechanism 102 adopts two combined light sourcesI.e. a combination of a sunlight source and an ultraviolet light source, the light source replenishment mechanism 102 comprises a combination of an LED strip 105 and an ultraviolet lamp 106; the LED light strip 105 illuminates the carbon fiber tow 802 at a 22 ° oblique angle, with a LED luminous flux of 9000 lm. An ultraviolet lamp 106 is arranged below the LED lamp, and the ultraviolet intensity is 80 mu W/cm2. The light source supply mechanism 102 is used to specifically identify the precursor carbon fiber bundle 802 that has a fluorescence effect under ultraviolet light. The daylight light source LED lamp strip 105 is used to detect broken filament fibers having a large color difference with the carbon fiber bundle 802, and the ultraviolet light source 106 is used to detect white PAN-type precursor fibers having a color similar to that of the precursor fibers. The image acquisition mechanism 101 is obliquely installed at 25 ︒ and is installed at the center point of the entire width of the carbon fiber tow. The reflector 104 is arranged behind the image acquisition mechanism and plays a role in enhancing the illumination intensity in the hearth.
As shown in fig. 3, the image capturing mechanism 101 is composed of an air source nozzle 107, a transparent component 108, a rack 109, a motor drive gear 110, a stepping motor 111, a focusing component 112, a camera 113, a protective sleeve 114 and a flange 115. The camera 113 is arranged in the protective sleeve 114, and the outer end of the protective sleeve 114 is fixed on the inner wall of the pre-oxidation furnace through a flange 115. The transparent sealing component 108 is sleeved on the inner side of the front section of the protective sleeve 114, and the head of the transparent component 108 is an air source spray head 107. The front end inside the protective sleeve 114 is provided with a focusing component 112, the focusing component 112 drags a lens 113 of the camera to move by a rack of a gear 109, a stepping motor 111 passes through a motor transmission gear 110, the motor transmission gear 110 is meshed with the rack 109, the camera 113 forms a clear and complete image by the focusing component 112, and the camera 113 is composed of the lens of the camera and a camera sensitive piece. The image acquisition mechanism 101 is obliquely installed at 25 ︒ and is installed at the center point of the entire width of the carbon fiber tow. The reflector 104 is arranged behind the image acquisition mechanism and plays a role in enhancing the illumination intensity in the hearth. In the present invention, the central axes of the transparent components 108 in the image capturing mechanism 101 are coincident, and the distance between the head of the transparent component 108 and the (carbon fiber) filament bundle is preferably 17 mm. The front end of the transparent component 108 is made of glass with high temperature resistance of 350 ℃. The reflector 104 is fixedly installed at the rear end of the image acquisition mechanism, the size of the reflector 104 is 40cm x 40cm, and the central point of the reflector 104 is superposed with the wide central line of the carbon fiber tow 802 and is superposed with the horizontal position of the image acquisition mechanism 101. The image capturing mechanism 101 is provided with a positive pressure air blowing member 116 that can clean the lens of the camera.
As shown in fig. 3 and 4, the collected carbon fiber image system 102 is provided for each layer of carbon fiber tows in the pre-oxidation furnace 801, wherein the end of the camera 113 in the image collecting mechanism 101 is opposite to the moving direction of the carbon fiber tows 802.
The image acquisition module 103 sends image information to the carbon fiber abnormality identification system 102 by using a data acquisition card of a PCI bus.
As shown in fig. 5, the system 102 for identifying carbon fiber anomalies analyzes the image to determine whether a broken fiber or a broken fiber exists in the carbon fiber tow image by the following steps:
firstly, reading a black and white image from an image acquisition card, preprocessing the image, reducing dimensions, extracting texture features, and classifying by a support vector machine to finally obtain an image identification result.
S1, image defogging and dimension reduction processing:
1.1 defogging method
The collected carbon fiber tow image may have the defect that the smoke shielding image is not clear. Defogging algorithm steps:
1. calculating dark image channel and calculating dark image channel
In the fog-free image, there is a high possibility that each local area will be shaded, either in pure color or black. Therefore, each local area will have a very low value. Firstly, the minimum value in each pixel is calculated, and is stored in a gray scale image with the same size as the original image, and then the minimum value filtering is carried out on the gray scale image.
Dark channel definition
Jc denotes a black-and-white image; Ω (X) represents a window centered on pixel X.
2. And (3) calculating the existing distance and depth of partial fog due to the space perspective phenomenon/shade distance and the refractive index by utilizing a dark channel, and correcting the weighted value:
3. Estimation of background light using dark channels
1. And selecting the brightest 0.1% pixel of the dark channel image.
2. And taking the brightest value of the pixel corresponding to the pixel in the input image as background light.
Defogging by substituting fog return diagram formula
Go drawing model
I(x)＝J(x)t(x)+A(1-t(x))
Wherein: i (x) -images to be dehazed; j (x) -fog free image; a-background light component t (x) -refractive index;
the invention relates to dimension reduction through a T-SNE algorithm.
S2, extracting image texture features
And (3) extracting image characteristics by adopting a mode GLCM gray level co-occurrence matrix according to the difference of texture characteristics of broken filaments and the normally straightened carbon fiber tows.
The gray level co-occurrence matrix provides information of image gray level direction, interval and variation amplitude, and statistical attributes used for quantitatively describing texture features are extracted on the basis of the gray level co-occurrence matrix.
Exploiting dependencies
Wherein cov is the element row and column coordinates of the similarity (i, j); (i, j) overall correlation coefficient; e (i) is the row element mean; e (j) is the mean value of the column elements
Is the standard deviation;
the similarity between the row or column elements in the GLCM reflects the extension length of the gray value along a certain direction, and the longer the extension is, the greater the correlation is.
Calling a graycotatrix () function in MATLAB, taking gray level co-occurrence matrixes in different directions (0, 45, 90 and 135 degrees), circularly calculating the gray level co-occurrence matrixes in all directions, and then taking an average value and a variance as finally extracted features.
S3 image classification
And (3) judging: no yarn breakage exists, and yarn breakage exists.
And the classification adopts a Support Vector Machine (SVM) algorithm. The support vector machine is a supervised learning model based on a statistical learning dimension theory and a structural risk minimization principle. And (3) mapping the input space to a high-dimensional space through a kernel function to construct an optimal classification plane, so that different classes are accurately separated.
And classifying the images by adopting a training svm function fitcecac carried by MATLAB. The final image identifies broken and normal tows. The method realizes the identification and classification of broken carbon fibers, calculates the positions of the broken carbon fibers by identifying the carbon fiber abnormality system, and cuts and removes the broken carbon fibers by removing the carbon fiber abnormality device.
As shown in fig. 1, 2 and 6, the carbon fiber abnormality removing device 300 includes a traveling guide rail 301, two actuators (302a,302b), a foreign substance removing table 303, a fan 304 and a waste fiber collecting box 305, wherein the fan 304 and the waste fiber collecting box 305 are respectively disposed on the foreign substance removing table 303; the actuators (302a,302b) are arranged on the walking guide rail 301; each of the actuators (302a,302b) is composed of a positioning gear 306, a reversing component 307, a stepping motor 308, an air cylinder 309, an electromagnetic valve 310, a clearing knife 311 and a speed reducer 312, wherein:
one end of the positioning gear 306 is connected with the traveling guide rail 301, and the other end thereof is connected with the speed reducer 312; the speed reducer 312 is connected with the cylinder 309 through a reversing component 307, one side of the reversing component 307 is connected with the stepping motor 308, and the cylinder bottom 309 is connected with the clearing knife 311; a reversing component 307 is arranged between the cylinder 309 and the stepping motor 308; the cylinder 308 is also provided with an electromagnetic valve 310.
In the device 300 for clearing carbon fiber abnormality of the present invention, both actuators (302a,302b) reach the positioning area. And identifying the gas source spray head of the carbon fiber abnormity system 200, spraying high-pressure gas, rapidly pushing out the tool bit 311 through the cylinder 309, and cutting the defective tows. The carbon fiber abnormality removing apparatus 300 drives a fan 304 to blow the cut tow on the foreign matter table 303 into the waste fiber collecting box 305. The carbon fiber abnormity clearing device 300 calculates the time for sending the control command according to the conveying speed of the tows, and drives corresponding control execution mechanisms (302a,302b) to clear when broken filaments and broken filaments reach a clearing area, so that the on-line automatic clearing of the broken filaments and the broken filaments is realized.
The present invention is not limited to the above-described embodiments. The foregoing description of the specific embodiments is intended to describe and illustrate the technical solutions of the present invention, and the above specific embodiments are merely illustrative and not restrictive. Those skilled in the art can make many changes and modifications to the invention without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (4)
1. A production platform for identifying anomalies in carbon fiber within a pre-oxidation furnace, the platform comprising: preoxidation stove, carbon fiber silk bundle, drafting device, its characterized in that: the platform further comprises: gather carbon fiber image system, discernment carbon fiber unusual system and clear away the unusual device of carbon fiber, wherein:
the system for acquiring carbon fiber images comprises: the system comprises an image acquisition mechanism, a light source supply mechanism, an image preprocessing module and an industrial personal computer; the light source supply mechanism is used for illuminating the tows which are drawn into a single-layer flat linear shape in the pre-oxidation furnace;
the image acquisition mechanism scans tows in the flat channel, and the image preprocessing module outputs the scanned image to a carbon fiber abnormality identification system by using a digital signal of 12 bits per pixel; the image acquisition mechanism consists of an air source spray head, a transparent component, a focusing component, a stretching mechanism, a camera, a protective sleeve and a flange; the protective sleeve is provided with a focusing component and is connected with the inner wall of the pre-oxidation furnace body through a flange; the focusing component is connected with the camera through a stretching mechanism; the focusing component is provided with a transparent component with an air source nozzle; the stretching mechanism consists of a gear, a stepping motor and a driving gear; one end of the gear is connected with the transparent component; the other end of the gear is connected with the stepping motor, and the stepping motor is connected with the gear through a transmission shaft;
the carbon fiber abnormity identification system judges the abnormity of the carbon fiber image output by the carbon fiber image acquisition system and transmits the abnormal position information of the carbon fiber tow image to the carbon fiber abnormity clearing device;
the carbon fiber abnormity removing device comprises a walking guide rail, two actuating mechanisms, a foreign matter removing table, a fan and a waste fiber collecting box, wherein the fan and the waste fiber collecting box are respectively arranged on the foreign matter removing table; the actuating mechanism is even arranged on the walking guide rail; every actuating mechanism comprises positioning gear, switching-over part, step motor, cylinder, solenoid valve, cleaing away sword, speed reducer, wherein:
one end of the positioning gear is connected with the walking guide rail, and the other end of the positioning gear is connected with the speed reducer; the speed reducer is connected with the clearing knife through the stepping motor and the air cylinder in sequence; a reversing component is arranged between the cylinder and the stepping motor; the cylinder is also provided with an electromagnetic valve;
the carbon fiber abnormal clearing device carries out shearing treatment on the carbon fiber tows according to the position where the carbon fiber tow image is abnormal; wherein:
an image acquisition mechanism is arranged on one side of the inner wall of the pre-oxidation furnace, and a light source supply mechanism is arranged on the inner wall of the other side of the pre-oxidation furnace; the image acquisition mechanism is connected with the inner wall of the pre-oxidation furnace in an included angle of 20-26 degrees through a reflector; the light source supply mechanism is connected with the inner wall of the pre-oxidation furnace at an included angle of 21-24 degrees; the distance between the carbon fiber tows incident to the image acquisition mechanism is 16-20 mm;
the light source supply mechanism adopts two combined light sources, namely a combination of a sunlight light source and an ultraviolet light source, and the ultraviolet light source is arranged below the sunlight light source; the sunlight source is an LED lamp strip, and the luminous flux of the sunlight source is 90001 m; the ultraviolet light source is an ultraviolet lamp with the intensity of 80 mu W/cm2。
2. The production platform for identifying the abnormality of the carbon fiber in the pre-oxidation furnace as claimed in claim 1, wherein: the image acquisition mechanism is provided with a positive pressure blowing component capable of cleaning a camera lens.
3. The production platform for identifying the abnormality of the carbon fiber in the pre-oxidation furnace as claimed in claim 1, wherein: the carbon fiber beam image abnormal position identification system sends the carbon fiber beam image abnormal position to the carbon fiber abnormality clearing device through the following steps of:
s101, carrying out defogging treatment on the carbon fiber tow image through the following formula to produce a first carbon fiber tow image;
I(x)＝J(x)t(x)+A(1-t(x))
wherein: i (x) -images to be dehazed; j (x) -fog free image; a-background light component t (x) -refractive index;
s102, generating a second fiber tow image for the first carbon fiber tow image through a T-SNE algorithm;
s103, extracting texture features of the second fiber tow image through a GLCM gray level co-occurrence matrix to obtain an abnormal carbon fiber tow image;
s104, judging broken filaments of abnormal parts in the abnormal carbon fiber tow image through an SVM algorithm, and if the broken filaments exist, sending broken filament position information to a carbon fiber abnormality clearing device; otherwise, returning to the step S101.
4. A production platform for identifying abnormality of carbon fiber in a pre-oxidation furnace according to any one of claims 1 to 3, characterized in that: the carbon fiber image acquisition system is respectively arranged on each layer of carbon fiber tows in the pre-oxidation furnace, and the front end of the image acquisition mechanism is opposite to the movement direction of the carbon fiber tows.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110985110.8A CN113430679B (en) | 2021-08-26 | 2021-08-26 | Production platform for identifying abnormality of carbon fiber in pre-oxidation furnace |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110985110.8A CN113430679B (en) | 2021-08-26 | 2021-08-26 | Production platform for identifying abnormality of carbon fiber in pre-oxidation furnace |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113430679A CN113430679A (en) | 2021-09-24 |
| CN113430679B true CN113430679B (en) | 2021-11-05 |
Family
ID=77797958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110985110.8A Active CN113430679B (en) | 2021-08-26 | 2021-08-26 | Production platform for identifying abnormality of carbon fiber in pre-oxidation furnace |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113430679B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114581404B (en) * | 2022-03-03 | 2022-08-30 | 常州市宏发纵横新材料科技股份有限公司 | Broken yarn detection method for interweaving binding yarns |
| CN115314618A (en) * | 2022-08-04 | 2022-11-08 | 南京耘瞳科技有限公司 | Carbon fiber carbon filament image on-line acquisition device |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7807002B2 (en) * | 2007-12-17 | 2010-10-05 | The Boeing Company | Verification of tow cut for automatic fiber placement |
| WO2009128541A1 (en) * | 2008-04-18 | 2009-10-22 | 三菱レイヨン株式会社 | Production system and production method of carbon fiber thread |
| CN101629336B (en) * | 2009-06-01 | 2012-12-26 | 上海中方宝达纺织智能仪器有限公司 | Method and device for online detection of foreign fiber in cotton |
| CN102564972A (en) * | 2012-01-06 | 2012-07-11 | 江南大学 | Evaluation method for soaping resistance color fastness of yarn-dyed fabrics based on machine vision |
| CN103163150B (en) * | 2013-03-18 | 2015-07-15 | 长飞光纤光缆股份有限公司 | Online cable surface defect detection device and online cable surface defect detection method |
| CN204162841U (en) * | 2014-10-15 | 2015-02-18 | 大英县天骄纺织有限公司 | The device of the dual on-line monitoring of a kind of foreign cotton fiber and removing |
| CN110349093B (en) * | 2019-06-11 | 2022-11-29 | 西安电子科技大学 | Single image defogging model construction and defogging method based on multi-stage hourglass structure |
| CN112697800B (en) * | 2020-12-09 | 2022-08-26 | 上海御微半导体技术有限公司 | Defect detection device and method |
-
2021
- 2021-08-26 CN CN202110985110.8A patent/CN113430679B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN113430679A (en) | 2021-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113430679B (en) | Production platform for identifying abnormality of carbon fiber in pre-oxidation furnace | |
| WO2021248554A1 (en) | High-speed and high-precision burr detection method and detection system for lithium ion battery electrode sheet | |
| WO2015055060A1 (en) | Online detecting method for continuous casting slab surface quality | |
| CN208313802U (en) | A kind of chip pin defect detecting system based on machine vision | |
| CN101696945A (en) | On-line detection method of machine vision system to photovoltaic glass flaws | |
| CN110702699B (en) | Rolled glass defect detection device and method | |
| CN101957326A (en) | Method and device for performing multi-spectral detection on surface quality of fabrics | |
| CN104165653A (en) | Device for collecting end face image and chamfer image of bearing roller and collecting method thereof | |
| CN102608125B (en) | Appearance quality detection system of tows | |
| CN105158273A (en) | Optical cable surface defect detecting system based on planar array CMOS cameras | |
| CN104132945A (en) | On-line surface quality visual inspection device for bar based on optical fiber conduction | |
| CN101696943A (en) | On-line detection method of machine vision system to medical surgical knife flaws | |
| CN113592852A (en) | Carbon fiber carbon wire quality online detection system and method | |
| CN202351182U (en) | Online high-speed detection system for defects on surface of tinplate | |
| CN208172420U (en) | A kind of liquid crystal module open defect detection system | |
| CN108490002B (en) | Method for detecting quality of microstructure prism lens | |
| CN214211370U (en) | Detection system and sorter based on time-sharing coaxial illumination imaging | |
| Wang et al. | Identification and detection of surface defects of outer package printed matter based on machine vision | |
| CN201345056Y (en) | Rubber production online monitoring device | |
| JP5696582B2 (en) | Running yarn inspection method and yarn manufacturing method | |
| CN105973910A (en) | Structure and texture characteristic-based lamp tube quality detection and flaw classification method and system thereof | |
| CN115170493A (en) | Identification method and system for micro-defect of sterile package and storage medium | |
| CN114878585A (en) | Large-breadth silk screen defect detection device | |
| CN102721703A (en) | Illuminating device and imaging method for online detection | |
| Li et al. | Research and design of inspection of LR6 battery negative surface scratches online defects based on computer vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |