A tow laying defect detection method based on active infrared technology

Abstract

The present application relates to a kind of based on active infrared technology's tow laying defect detection method, it is heated by heat source and is irradiated tow layer, infrared thermal imager obtains heated tow layer continuous infrared image, the infrared image collected is sent to host computer, extract pixel level tow temperature data history data, obtain and tow laying quality related information by analysis algorithm;That is, using the heat transfer characteristics of adjacent tow gap and tow body is different, calculate tow boundary by temperature distribution;The data obtained are clustered and analyzed to obtain defect temperature field, the center pixel point completed by clustering is traversed temperature data value according to time sequence, utilize the difference of the heat transfer characteristics of existing defect and normal tow and its regularity to temperature field distribution, the detection of tow laying defect is carried out.The present application utilizes infrared thermal imaging technology, image processing technology, realizes the defect detection of various types in the process of tow laying by comparing and analyzing the temperature field difference between defect area and normal area.

Description

[0001] The application relates to a towpreg laying defect detection method, in particular to a towpreg laying defect detection method based on active infrared technology, and belongs to the technical field of nondestructive testing.

[0002] Fiber reinforced composites are important engineering materials. With the rapid increase of domestic carbon fiber production lines and output in recent years, the manufacturing technology has developed rapidly. Among them, the fiber laying defects of composites are important factors affecting the manufacturing performance of composite components. The laying defects can be divided into orientation deviation (i.e. difference between actual fiber direction and design direction) and fiber defects (including overlap, gap, wrinkle and the like). Due to the characteristics of these defects, conventional detection methods are difficult to meet the requirements.

[0003] The laser-assisted detection system of the prior art is usually used to calibrate the laying boundary and cooperate with manual naked eye detection. However, this detection method has high requirements for relative position, the projection is prone to deviation, and the efficiency is low. For the orientation deviation, since the fiber towpreg has strong light reflection characteristics, the surface photo imaging effect is poor, and therefore the result accuracy of the common optical imaging method is not high. For the overlap, wrinkle and other defects, in most cases, the defects are located inside the composite component layer, and internal flaw detection methods must be used for testing. The current internal curve detection of composites usually adopts ultrasonic scanning. However, ultrasonic scanning needs medium to have better detection effect, such as water and the like. The existence of the medium will affect the quality of the composite surface.

[0004] Therefore, in order to solve the above technical problems, it is necessary to provide an innovative towpreg laying defect detection method based on active infrared technology.

[0005] The application aims to provide a towpreg laying defect detection method based on active infrared technology, which can realize rapid detection of defects in the towpreg laying process and can identify the defects.

[0006] To achieve the above-mentioned purpose, the technical scheme adopted by the application is as follows: a towpreg laying defect detection method based on active infrared technology, which comprises the following steps:

[0007] 1) Turn on the infrared thermal imager, and start the driving compression roller on the towpreg head to tightly lay the towpreg on the mold;

[0008] 2) Turn on the integrated pulse heat source lamp, and transfer the heat generated by the heat source lamp from one side of the towpreg to the other side through heat transfer;

[0009] ​​​3), the infrared thermal imager continuously records the dynamic image of the temperature change of the detected fiber bundle surface, and the host computer traverses the pixel-level temperature data of each infrared image in time sequence, and the laying track of the fiber bundle is segmented according to the isotherm;

[0010] 4), the extracted temperature data is subjected to cluster analysis, the pixel temperature of the center point of each class obtained is traversed by the host computer in time sequence, and the relationship between the center pixel point temperature and time is obtained;

[0011] 5), the temperature change information extracted by the host computer is compared with the standard temperature change information without defects, and the type of defects is judged through further analysis of the data.

[0012] The fiber laying defect detection method based on active infrared technology of the application further comprises that the infrared thermal imager is fixed on one side of the fiber laying head and located above and in front of the laying mold, and can continuously shoot the temperature information of the laying fiber bundle surface.

[0013] The fiber laying defect detection method based on active infrared technology of the application further comprises that the pulse heat source lamp is fixed on the other side of the fiber laying head, which is used to provide a heat source during the laying of the fiber bundle, and the change of the laying fiber bundle surface temperature is recorded by the infrared thermal imager through the principle of infrared radiation.

[0014] The fiber laying defect detection method based on active infrared technology of the application further comprises that the step 3) is specifically that the host computer traverses the pixel-level temperature data of each infrared image in time sequence: the temperature data of the fiber bundle surface corresponding to each image pixel recorded by the infrared thermal imager is obtained, the number of pixel points of each infrared image is N x ×N y , x rows and y columns of temperature data are obtained; by using the heat transfer characteristics between the gaps between the fiber bundles and the fiber bundle bodies, the gap between the fiber bundles is selected as the reference temperature, and the isotherm is drawn based on the reference temperature, and the obtained area is the boundary of the laying fiber bundle.

[0015] The fiber laying defect detection method based on active infrared technology of the application further comprises that the step 4) is specifically that each infrared temperature data is clustered by K-means clustering, the pixel point temperature data of the infrared image is divided into k classes C={c1, c2,..., ck}, and the Euclidean distance is selected as the similarity clustering judgment, and the clustering formula is calculated as:

[0016]

[0017] Wherein, J(C) represents the minimum total distance square sum of each class, k represents the number of classes, x i represents the data of the sample, and u kThe centroid of each class is represented; after the clustering is completed, the host computer traverses each pixel point temperature of the class center of the completed clustering to obtain data of the temperature of each class center changing with time.

[0018] The active infrared technology-based tow laying defect detection method of the application further comprises the following specific judgment method of the defect type in step 5):

[0019] Overlap: the fiber layer appears overlapping laying, the heat transfer energy is reduced, the surface temperature is reduced, the time required for the temperature of the defect area to reach the peak value is longer than that of the normal laying, the energy absorbed by the single tow is reduced, and the peak value is lower;

[0020] Gap: the fiber layer appears a gap, the gap in the tow laying makes the single tow absorb more heat, the surface temperature of the tow is higher, and the time required for the fiber laying to reach the peak value is shorter than that of the normal laying and the peak value is higher;

[0021] Resin deposition: the resin sinks during the laying process, so that the peak value is lower than that of the normal laying, the resin deposition is less, and the time required to reach the peak value is similar to that of the normal laying;

[0022] Wrinkle: the fiber appears a bulge during the laying process, the bulge also lags behind the peak value, and the gap between the bulges is large, so that the peak temperature is lower than that of the tow laying of the overlap defect, and the time required to reach the peak value is shorter.

[0023] Compared with the prior art, the application has the following beneficial effects:

[0024] 1. The active infrared technology-based tow laying defect detection method of the application utilizes the pulse heat source and the infrared thermal imaging method to compare and analyze the temperature field difference between the defect area and the normal area, realizes dynamic defect detection in the tow laying process, and can accurately and effectively judge the defect position.

[0025] 2. Compared with the traditional ultrasonic scanning detection, the active infrared technology-based tow laying defect detection method of the application does not need to contact the surface of the composite material, and reduces the pollution of the composite material layer.

[0026] 3. The active infrared technology-based tow laying defect detection method of the application can accurately judge the type of the laying defect according to the curve of the temperature of the tow surface changing with time, and improves the manufacturing process of the composite material layer. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1It is a structure schematic diagram of the towpreg defect detection device applied to the present application.

[0028] Figure 2 It is a towpreg orientation boundary map segmented by the isotherm in step 3) of the present application.

[0029] Figure 3 It is a temperature change over time map of the cluster centroid of the infrared image traversed by the host computer in step 4) of the present application.

[0030] Figure 4 It is a towpreg layer different defect type and corresponding temperature time image of the present application.

CONCRETE EMBODIMENT

[0031] Please refer to the drawings in the description Figure 1 to the drawings Figure 4 The present application relates to a towpreg defect detection method based on active infrared technology, which adopts a towpreg defect detection device, which is composed of a driving pressure roller 1, an infrared thermal imager 2, a towpreg head 4, a pulse heat source lamp 5 and a mold 6. Among them, the towpreg head 4 is used to lay the fiber towpreg 3 on the mold 6, which is a prior art and will not be described here. The driving pressure roller 1 is installed at the bottom of the towpreg head 4, which is pressed on the fiber towpreg 3.

[0032] The infrared thermal imager 2 is fixed on one side of the towpreg head 4 and located above and in front of the laying mold 6, which can continuously shoot the temperature information of the surface of the laying towpreg 3. The pulse heat source lamp 5 is fixed on the other side of the towpreg head 4, which is used to provide heat source during the towpreg laying process, and the change of the surface temperature of the laying towpreg is recorded by the infrared thermal imager 2 through the principle of infrared radiation.

[0033] The towpreg defect detection method based on active infrared technology of the present application includes the following steps:

[0034] 1), turn on the infrared thermal imager 2, and start the driving pressure roller 1 on the towpreg head 4 to lay the fiber towpreg 3 closely on the mold 6.

[0035] 2), turn on the integrated pulse heat source lamp 5, and transfer the heat generated by the heat source lamp from one side of the towpreg 3 to the other side through heat transfer.

[0036] 3), the infrared thermal imager 2 continuously records the temperature change dynamic image of the surface of the detected towpreg 3, and the host computer traverses the pixel-level temperature data of each infrared image in time sequence, and the laying track of the towpreg is segmented according to the isotherm.

[0037] Specifically, the host computer traverses the pixel-level temperature data of each infrared image in time sequence: the temperature data of the surface of the tows corresponding to each pixel point of each image recorded by the infrared thermal imager in sequence, the number of pixel points of each infrared image is N x ×N y , obtaining temperature data of x rows and y columns; using the heat transfer characteristics between the gaps between the tows and the tow bodies, selecting the gaps between the tows as the reference temperature, and taking the reference temperature as the isotherm, the obtained area is the boundary of the laid tows.

[0038] 4), the extracted temperature data is subjected to cluster analysis, and the pixel temperature of the center point of each class obtained is traversed by the host computer in time sequence to obtain the relationship between the center pixel temperature and time.

[0039] In this step, each infrared temperature data is clustered by K-means clustering, and the pixel temperature data of the infrared image is divided into k classes, C={Pc1, c2,... ck}, C represents the set of sample clusters, and the Euclidean distance is selected as the similarity clustering judgment, and the clustering formula is calculated as:

[0040]

[0041] Wherein, J(C) represents the minimum total distance square sum of each class, k represents the number of classes, x i represents the data of the sample, u k represents the centroid of each class; after clustering, the host computer traverses the pixel temperature of each class center, and obtains the data of the temperature change of each cluster center with time.

[0042] 5), the temperature change information extracted by the host computer is compared with the standard temperature change information without defects, and the type of defects is judged by further analyzing the data.

[0043] Wherein, the defect types include but are not limited to overlap 7, gap 9, resin deposition 8, wrinkle 10, etc., and the specific judgment method is:

[0044] Overlap 7: the fiber lay-up appears overlapping lay-up, the heat transfer energy decreases, the surface temperature decreases, the overlapping lay-up makes the temperature of the defect area reach the peak value, and the time required for the temperature of the normal 11 lay-up to reach the peak value is longer, the energy absorbed by the single tow decreases, and the peak value is lower.

[0045] Gap 9: the fiber lay-up appears gap, the gap in the tow lay-up makes the single tow absorb more heat, the temperature of the tow surface is higher, and the fiber lay-up reaches the peak value in a shorter time and with a higher peak value compared with the normal 11 lay-up.

[0046] Resin deposition 8: the fiber sinks in the resin during the laying process, so the peak size is lower, the resin deposition is less, and it does not affect the time to reach the peak, so the time to reach the peak is similar to that of normal laying.

[0047] Wrinkles 10: the fiber appears to be raised during the laying process, the raised peak also lags, and the raised gap is large, the peak temperature is lower, and the time to reach the peak is shorter, compared to the temperature change rule of the overlapping defect tows.

[0048] In summary, the tow laying defect detection method based on active infrared technology of the present application applies infrared thermal imaging technology to the defect detection of composite materials, can obtain the temperature change information of the surface of the composite material, and through further analysis can judge which type of defect it belongs to, so as to judge the good or bad of the tow laying process, and ensure the process quality of the production product.

[0049] The above specific embodiments are only the preferred embodiments of the present application, and are not intended to limit the present application, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A tow laying defect detection method based on active infrared technology, characterized in that: It comprises the following steps: 1) Turn on the infrared thermal imager, and start the driving pressure roller on the filament head to tightly lay the fiber tows on the mold; 2) Turn on the integrated pulse heat source lamp, and transfer the heat generated by the heat source lamp from one side of the tows to the other side through heat transfer; 3), the infrared thermal imager continuously records the dynamic image of the temperature change of the detected fiber bundle surface, the host computer traverses the pixel-level temperature data of each infrared image in time sequence, and the laying track of the fiber bundle is segmented according to the isotherm, and each frame of image recorded by the infrared thermal imager corresponds to the temperature data of the fiber bundle surface, and the number of pixel points of each infrared image is , x rows of y columns of temperature data are obtained; by using the heat transfer characteristics between the gaps between the fiber bundles and the fiber bundle bodies, the gap between the fiber bundles is selected as the reference temperature, and the obtained region is the boundary of the laid fiber bundle; 4) Cluster analysis is performed on the extracted temperature data, and the pixel temperature of the center point of each class obtained is traversed in time sequence by the upper computer to obtain the relationship between the center pixel point temperature and time; 5) The temperature change information extracted by the upper computer is compared with the standard temperature change information without defects, and the defect type is judged through further analysis of the data; The defect types include overlap, gap, resin deposition and wrinkle, and the specific judgment method is: Overlap: refers to the overlap of fiber laying, the heat transfer energy decreases, the surface temperature decreases, the overlap of laying makes the temperature of the defect area reach the peak value for a longer time than the normal laying, the energy absorbed by the single tow decreases, and the peak value is lower; Gap: refers to the gap in the fiber laying, the gap in the tow laying makes the single tow absorb more heat, the surface temperature of the tow is higher, and the peak value is higher than that of the normal laying; Resin deposition: refers to the sinking of resin during the laying process, so the peak value is lower compared with normal laying, the resin deposition is less and will not affect the time to reach the peak value, so the time to reach the peak value is similar to that of normal laying; Wrinkle: refers to the bulge of fiber during the laying process, the bulge also lags behind the peak value, and the gap of the bulge is large, so the peak temperature is lower and the time to reach the peak value is shorter than that of the overlap defect.

2. The tow placement defect detection method based on active infrared technology according to claim 1, characterized in that: The infrared thermal imager is fixed on one side of the filament head and located above and in front of the laying mold, which can continuously shoot the temperature information of the laid tow surface.

3. The tow placement defect detection method based on active infrared technology according to claim 2, characterized in that: The integrated pulse heat source lamp is fixed on the other side of the filament head, which is used to provide heat source during the laying of the tows, and the change of the surface temperature of the laid tows is recorded by the infrared thermal imager through the principle of infrared radiation.

4. The method of tow placement defect detection based on active infrared technology as claimed in claim 1, wherein: The step 4) is specifically: clustering each infrared temperature data by K-means clustering, and dividing the pixel temperature data of the infrared image into k classes , selecting the Euclidean distance as the similarity clustering judgment, and the clustering formula is: wherein, represents the minimum of the total distance square sum of each class, k represents the number of classes, represents the data of the sample, represents the centroid of each class; after clustering is completed, the host computer traverses the pixel point temperature of each clustering completed class center to obtain the data of the temperature change of each clustering center with time.

Images