CN116461120A - Preparation method and system of phenolic composite material - Google Patents
Preparation method and system of phenolic composite material Download PDFInfo
- Publication number
- CN116461120A CN116461120A CN202310554023.6A CN202310554023A CN116461120A CN 116461120 A CN116461120 A CN 116461120A CN 202310554023 A CN202310554023 A CN 202310554023A CN 116461120 A CN116461120 A CN 116461120A
- Authority
- CN
- China
- Prior art keywords
- prepreg
- image
- feature vectors
- local area
- characteristic feature
- 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.)
- Granted
Links
- 239000002131 composite material Substances 0.000 title claims abstract description 77
- ISWSIDIOOBJBQZ-UHFFFAOYSA-N phenol group Chemical group C1(=CC=CC=C1)O ISWSIDIOOBJBQZ-UHFFFAOYSA-N 0.000 title claims abstract description 74
- 238000002360 preparation method Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims abstract description 39
- 238000001514 detection method Methods 0.000 claims abstract description 29
- 239000013598 vector Substances 0.000 claims description 202
- 238000000034 method Methods 0.000 claims description 50
- 238000013527 convolutional neural network Methods 0.000 claims description 44
- 230000000903 blocking effect Effects 0.000 claims description 23
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000010586 diagram Methods 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 11
- 238000005457 optimization Methods 0.000 claims description 8
- 238000000638 solvent extraction Methods 0.000 claims description 6
- 238000012546 transfer Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 5
- 238000011176 pooling Methods 0.000 claims description 5
- 238000013135 deep learning Methods 0.000 abstract description 6
- 238000005516 engineering process Methods 0.000 abstract description 6
- 238000005065 mining Methods 0.000 abstract description 6
- 238000009826 distribution Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 11
- 238000004590 computer program Methods 0.000 description 10
- 230000008569 process Effects 0.000 description 10
- 230000000694 effects Effects 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000003062 neural network model Methods 0.000 description 4
- 230000000306 recurrent effect Effects 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- KXGFMDJXCMQABM-UHFFFAOYSA-N 2-methoxy-6-methylphenol Chemical compound [CH]OC1=CC=CC([CH])=C1O KXGFMDJXCMQABM-UHFFFAOYSA-N 0.000 description 3
- WSFSSNUMVMOOMR-UHFFFAOYSA-N Formaldehyde Chemical compound O=C WSFSSNUMVMOOMR-UHFFFAOYSA-N 0.000 description 3
- 230000009471 action Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 239000003365 glass fiber Substances 0.000 description 3
- 239000005011 phenolic resin Substances 0.000 description 3
- 229920001568 phenolic resin Polymers 0.000 description 3
- 230000003252 repetitive effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000001149 cognitive effect Effects 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 239000000835 fiber Substances 0.000 description 2
- 238000010438 heat treatment Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000015654 memory Effects 0.000 description 2
- 230000005012 migration Effects 0.000 description 2
- 238000013508 migration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 229920000049 Carbon (fiber) Polymers 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 239000004917 carbon fiber Substances 0.000 description 1
- 239000003054 catalyst Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000004132 cross linking Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 229910002804 graphite Inorganic materials 0.000 description 1
- 239000010439 graphite Substances 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 239000012779 reinforcing material Substances 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000002791 soaking Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C70/00—Shaping composites, i.e. plastics material comprising reinforcements, fillers or preformed parts, e.g. inserts
- B29C70/04—Shaping composites, i.e. plastics material comprising reinforcements, fillers or preformed parts, e.g. inserts comprising reinforcements only, e.g. self-reinforcing plastics
- B29C70/28—Shaping operations therefor
- B29C70/40—Shaping or impregnating by compression not applied
- B29C70/42—Shaping or impregnating by compression not applied for producing articles of definite length, i.e. discrete articles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C70/00—Shaping composites, i.e. plastics material comprising reinforcements, fillers or preformed parts, e.g. inserts
- B29C70/04—Shaping composites, i.e. plastics material comprising reinforcements, fillers or preformed parts, e.g. inserts comprising reinforcements only, e.g. self-reinforcing plastics
- B29C70/28—Shaping operations therefor
- B29C70/54—Component parts, details or accessories; Auxiliary operations, e.g. feeding or storage of prepregs or SMC after impregnation or during ageing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29K—INDEXING SCHEME ASSOCIATED WITH SUBCLASSES B29B, B29C OR B29D, RELATING TO MOULDING MATERIALS OR TO MATERIALS FOR MOULDS, REINFORCEMENTS, FILLERS OR PREFORMED PARTS, e.g. INSERTS
- B29K2061/00—Use of condensation polymers of aldehydes or ketones or derivatives thereof, as moulding material
- B29K2061/04—Phenoplasts
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
A preparation method and a system of phenolic composite material, which acquire a detection image of prepreg; and the detection image of the prepreg is subjected to feature mining and capturing based on an image processing technology of deep learning, and the optimal curing temperature is automatically recommended through decoding processing, so that the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
Description
Technical Field
The application relates to the technical field of intelligent preparation, and in particular relates to a preparation method and a system of a phenolic composite material.
Background
Phenolic composite material is a high-performance composite material composed of phenolic resin and reinforcing materials (such as glass fiber, carbon fiber, graphite and the like), is generally used as structural members, friction members and heat insulation members, and is widely applied to the fields of aerospace, automobiles, electronics, buildings and the like.
Generally, the phenolic composite is prepared by the steps of: s1, mixing phenol, formaldehyde and a catalyst according to a certain proportion, and heating to react to obtain a phenolic resin solution; s2, soaking the glass fiber cloth in a phenolic resin solution to fully soak the glass fiber cloth, and then extruding the redundant solution to obtain a prepreg; and S3, heating and curing the prepreg at high temperature and high pressure to obtain the phenolic composite material.
In the process of preparing the phenolic composite material, the too high or too low curing temperature can influence the crosslinking density and molecular structure of the composite material, reduce the strength and toughness of the composite material and influence the mechanical property, heat resistance and stability of the composite material. The control of the curing temperature is often not accurate enough, and different operators have different experiences and lack consistency in the control of the curing temperature, so that the consistency of the quality of the phenolic composite material is affected. Thus, a solution is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a preparation method and a system of a phenolic composite material, wherein a detection image of a prepreg is obtained; and the detection image of the prepreg is subjected to feature mining and capturing based on an image processing technology of deep learning, and the optimal curing temperature is automatically recommended through decoding processing, so that the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
In a first aspect, a method of preparing a phenolic composite is provided, comprising: obtaining a detection image of the prepreg; performing image blocking processing on the detection image of the prepreg to obtain a sequence of image blocks; passing each image block in the sequence of image blocks through a convolutional neural network model serving as a filter to obtain a plurality of prepreg local area characteristic feature vectors; passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being indicative of a recommended cure temperature value.
In the above method for preparing a phenolic composite material, performing image blocking processing on the detected image of the prepreg to obtain a sequence of image blocks, including: and uniformly partitioning the detected image of the prepreg to obtain the sequence of image blocks, wherein each image block in the sequence of image blocks has the same size.
In the above method for preparing a phenolic composite material, passing each image block in the sequence of image blocks through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors, including: each layer of the convolutional neural network model used as the filter performs the following steps on input data in forward transfer of the layer: carrying out convolution processing on the input data to obtain a convolution characteristic diagram; carrying out mean pooling treatment based on a feature matrix on the convolution feature map to obtain a pooled feature map; performing nonlinear activation on the pooled feature map to obtain an activated feature map; wherein the output of the last layer of the convolutional neural network model as a filter is the local area characteristic feature vector of the plurality of prepregs, and the input of the first layer of the convolutional neural network model as a filter is each image block in the sequence of image blocks.
In the above method for preparing a phenolic composite material, passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain a decoded feature vector, comprising: passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain a plurality of context prepreg local area characteristic feature vectors; respectively calculating Gaussian regression uncertainty factors of characteristic feature vectors of the local areas of the plurality of context prepregs to obtain a plurality of Gaussian regression uncertainty factors; weighting the plurality of context prepreg local area characteristic feature vectors based on the plurality of gaussian regression uncertainty factors respectively to obtain a plurality of weighted context prepreg local area characteristic feature vectors; and concatenating the plurality of weighted context prepreg local region characteristic feature vectors to obtain the decoded feature vector.
In the above method for preparing a phenolic composite material, passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain a plurality of context prepreg local area characteristic feature vectors, comprising: performing one-dimensional arrangement on the characteristic feature vectors of the local areas of the prepregs to obtain global feature vectors of the prepregs; calculating the product between the global prepreg characteristic vector and the transpose vector of each local prepreg regional characteristic vector in the local prepreg regional characteristic vectors to obtain a plurality of self-attention correlation matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and weighting each prepreg local region characteristic feature vector in the prepreg local region characteristic feature vectors by taking each probability value in the probability values as a weight so as to obtain the context prepreg local region characteristic feature vectors.
In the above method for preparing a phenolic composite material, the step of calculating the gaussian regression uncertainty factors of the characteristic feature vectors of the local areas of the plurality of contextual prepregs to obtain a plurality of gaussian regression uncertainty factors includes: calculating the Gaussian regression uncertainty factors of the characteristic feature vectors of the local areas of the plurality of the contextual prepregs according to the following optimization formula to obtain a plurality of Gaussian regression uncertainty factors; wherein, the optimization formula is:, wherein ,/>Is the length of the line feature vector, < >> and />Each of the plurality of context prepreg local area characteristic feature vectorsMean and variance of feature set of locations, wherein +.>Is the +.o.of the local area characteristic feature vector of the contextual prepreg>Characteristic value of individual position->Is the multiple Gaussian regression uncertainty factor, and +.>Is a logarithmic function with a base of 2.
In the above method for preparing a phenolic composite material, cascading the plurality of weighted local area characteristic feature vectors of the context prepreg to obtain the decoded feature vector includes: cascading the plurality of weighted context prepreg local area characteristic feature vectors with the following cascading formula to obtain the decoding feature vector; wherein, the cascade formula is: wherein , />Representing the local regional characteristic feature vector of the plurality of weighted contextual prepregs, +.>Representing a cascade function->Representing the decoded feature vector.
In the above method for preparing a phenolic composite material, passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being used to represent a recommended cure temperature value, comprising: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein the decoding algorithmThe formula is:，/>representing said decoded feature vector,/->Representing the decoded value->Representing a weight matrix, +.>Representing the bias vector +_>Representing a matrix multiplication.
In a second aspect, there is provided a phenolic composite preparation system comprising: the image acquisition module is used for acquiring a detection image of the prepreg; the image blocking module is used for carrying out image blocking processing on the detection image of the prepreg to obtain a sequence of image blocks; the feature extraction module is used for enabling each image block in the sequence of the image blocks to pass through a convolutional neural network model serving as a filter so as to obtain a plurality of prepreg local area characteristic feature vectors; a context encoding unit for passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and a decoding module for passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being indicative of a recommended cure temperature value.
In the above phenolic composite preparation system, the image blocking module is configured to: and uniformly partitioning the detected image of the prepreg to obtain the sequence of image blocks, wherein each image block in the sequence of image blocks has the same size.
Compared with the prior art, the preparation method and the system of the phenolic composite material provided by the application acquire the detection image of the prepreg; and the detection image of the prepreg is subjected to feature mining and capturing based on an image processing technology of deep learning, and the optimal curing temperature is automatically recommended through decoding processing, so that the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments or the description of the prior art will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a schematic view of a scenario of a method of preparing a phenolic composite according to an embodiment of the present application.
Fig. 2 is a flow chart of a method of preparing a phenolic composite according to an embodiment of the present application.
Fig. 3 is a schematic diagram of a method of preparing a phenolic composite according to an embodiment of the present application.
Fig. 4 is a flow chart of the sub-steps of step 140 in a method of preparing a phenolic composite in accordance with an embodiment of the present application.
Fig. 5 is a flow chart of the sub-steps of step 141 in a method of preparing a phenolic composite in accordance with an embodiment of the present application.
Fig. 6 is a block diagram of a phenolic composite preparation system according to an embodiment of the present application.
Description of the embodiments
The following description of the technical solutions in the embodiments of the present application will be made with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Unless defined otherwise, all technical and scientific terms used in the examples of this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In the description of the embodiments of the present application, unless otherwise indicated and defined, the term "connected" should be construed broadly, and for example, may be an electrical connection, may be a communication between two elements, may be a direct connection, or may be an indirect connection via an intermediary, and it will be understood by those skilled in the art that the specific meaning of the term may be understood according to the specific circumstances.
It should be noted that, the term "first\second\third" in the embodiments of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that the embodiments of the present application described herein may be implemented in sequences other than those illustrated or described herein.
Aiming at the technical problems, the technical conception of the method is that the image processing technology based on deep learning is utilized to perform feature mining and capturing on the detection image of the prepreg, and the optimal curing temperature is automatically recommended through decoding processing, so that the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
Specifically, in the technical solution of the present application, first, a detection image of a prepreg is acquired. Here, the detected image of the prepreg may reflect information of thickness, density, etc. of the prepreg, which reflect the state of the current prepreg, thereby providing a basis for control of the curing temperature.
Then, the detected image of the prepreg is subjected to image blocking processing to obtain a sequence of image blocks. Here, the image blocking process divides the detected image into several small image blocks, and the local area information of the prepreg can be highlighted to some extent.
Then, each image block in the sequence of image blocks is passed through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors. The convolutional neural network model (CNN) is a deep learning model, is good at processing high-dimensional data such as images, and can extract local features in the images. Specifically, by using the convolutional neural network model as a filter, characteristics of a local region of the prepreg, such as fiber distribution, density, saturation, etc., which are closely related to control of the curing temperature, can be effectively extracted from the detected image of the prepreg.
The local area characteristic feature vectors of the prepregs represent high-dimensional implicit feature information of each image block, but relevance among the image blocks is not considered. Here, the transducer-based context encoder is a neural network model that is capable of efficiently capturing long-range dependencies between different locations in a sequence, and that uses a self-attention mechanism to calculate the correlation of each location in the sequence with other locations, thereby generating an encoded representation that contains global context information. The context encoder based on the converter has higher parallelism, lower computational complexity and better generalization capability than the conventional recurrent or convolutional neural network.
The decoded feature vector is then passed through a decoder to obtain a decoded value, which is used to represent the recommended cure temperature value. The decoder is a neural network model capable of generating output data according to input data, and can automatically learn related rules from the input data and realize data output. In practical application, the curing temperature can be adjusted according to the decoding value, so that the cost and error of manual intervention are reduced.
Here, in consideration of source image noise introduced in the image acquisition process of the detected image of the prepreg, after image blocking is performed on the detected image of the prepreg, source image noise exists in each image block, after image semantic feature extraction is performed through a convolutional neural network model serving as a filter and context-dependent encoding of image semantic features is performed through a context encoder based on a converter, gaussian distribution error uncertainty of respective feature distribution is further introduced into a plurality of context prepreg local area feature vectors obtained through the context encoder based on the converter, and thus, in consideration of the decoded feature vectors obtained by directly cascading the plurality of context prepreg local area feature vectors, direct superposition of the gaussian distribution error uncertainty also causes decoding regression errors of the decoded feature vectors, and accuracy of decoding values obtained by the decoder of the decoded feature vectors is affected.
Based on this, in the technical solution of the present application, each of the plurality of contextual prepreg local area characteristic feature vectors is calculated separately, for example, as Is expressed as: /> Is the length of the line feature vector, < >> and />Feature set +.>Mean and variance of (1), wherein>Is a feature vector +.>Is>Characteristic value of individual position, and->The base 2 logarithm.
Here, for the unknown regression of the decoded feature vector, which may be caused by the distribution uncertainty information of the integrated feature set of each of the local regional feature vectors of the plurality of local regional feature vectors of the contextual prepreg, the scalar measurement of the statistical characteristics of the feature set is performed by using the mean value and the variance as the statistical quantization parameters, so that the normal distribution cognitive mode represented by the feature error is expanded to an unknown distribution regression mode, and the migration learning based on the natural distribution transfer on the feature set scale is realized, so that the decoded feature vector is obtained by weighting each local regional feature vector of the contextual prepreg by the gaussian regression uncertainty factors and cascading the weighted local regional feature vector of the contextual prepreg, and the uncertainty correction based on the self calibration of each local regional feature vector of the contextual prepreg when the decoded feature vector is formed is realized, so that the decoding regression error existing in the decoded feature vector is corrected, and the accuracy of the decoding value obtained by the decoder of the decoded feature vector is improved.
The technical effects of the application are as follows: 1. provides a preparation scheme of an intelligent phenolic composite material, and more particularly relates to a curing temperature recommended scheme of the phenolic composite material. 2. According to the scheme, the complex relation between the prepreg detection image and the curing temperature can be automatically learned, the optimal curing temperature is automatically recommended, the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
Fig. 1 is a schematic view of a scenario of a method of preparing a phenolic composite according to an embodiment of the present application. As shown in fig. 1, in this application scenario, first, a detection image (e.g., C as illustrated in fig. 1) of a prepreg (e.g., M as illustrated in fig. 1) is acquired; the acquired detection image is then input into a server (e.g., S as illustrated in fig. 1) deployed with a phenolic composite preparation algorithm, wherein the server is capable of processing the detection image based on the phenolic composite preparation algorithm to generate a decoded value representative of the recommended cure temperature value.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
In one embodiment of the present application, fig. 2 is a flow chart of a method of preparing a phenolic composite according to an embodiment of the present application. As shown in fig. 2, a method 100 for preparing a phenolic composite according to an embodiment of the present application includes: 110, obtaining a detection image of the prepreg; 120, performing image blocking processing on the detected image of the prepreg to obtain a sequence of image blocks; 130, passing each image block in the sequence of image blocks through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors; 140 passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and, 150, passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being used to represent a recommended cure temperature value.
Fig. 3 is a schematic diagram of a method of preparing a phenolic composite according to an embodiment of the present application. As shown in fig. 3, in the network architecture, first, a detection image of a prepreg is acquired; then, performing image blocking processing on the detection image of the prepreg to obtain a sequence of image blocks; then, each image block in the sequence of image blocks is passed through a convolutional neural network model serving as a filter to obtain a plurality of prepreg local area characteristic feature vectors; then, passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and finally, passing the decoded feature vector through a decoder to obtain a decoded value, wherein the decoded value is used for representing the recommended curing temperature value.
Specifically, in step 110, a detection image of the prepreg is acquired. Aiming at the technical problems, the technical conception of the method is that the image processing technology based on deep learning is utilized to perform feature mining and capturing on the detection image of the prepreg, and the optimal curing temperature is automatically recommended through decoding processing, so that the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
Specifically, in the technical solution of the present application, first, a detection image of a prepreg is acquired. Here, the detected image of the prepreg may reflect information of thickness, density, etc. of the prepreg, which reflect the state of the current prepreg, thereby providing a basis for control of the curing temperature.
Specifically, in step 120, the detected image of the prepreg is subjected to image blocking processing to obtain a sequence of image blocks. Then, the detected image of the prepreg is subjected to image blocking processing to obtain a sequence of image blocks. Here, the image blocking process divides the detected image into several small image blocks, and the local area information of the prepreg can be highlighted to some extent.
The method for performing image blocking processing on the detection image of the prepreg to obtain a sequence of image blocks comprises the following steps: and uniformly partitioning the detected image of the prepreg to obtain the sequence of image blocks, wherein each image block in the sequence of image blocks has the same size.
Specifically, in step 130, each image block in the sequence of image blocks is passed through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors. Then, each image block in the sequence of image blocks is passed through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors. The convolutional neural network model (CNN) is a deep learning model, is good at processing high-dimensional data such as images, and can extract local features in the images. Specifically, by using the convolutional neural network model as a filter, characteristics of a local region of the prepreg, such as fiber distribution, density, saturation, etc., which are closely related to control of the curing temperature, can be effectively extracted from the detected image of the prepreg.
Passing each image block in the sequence of image blocks through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors, comprising: each layer of the convolutional neural network model used as the filter performs the following steps on input data in forward transfer of the layer: carrying out convolution processing on the input data to obtain a convolution characteristic diagram; carrying out mean pooling treatment based on a feature matrix on the convolution feature map to obtain a pooled feature map; performing nonlinear activation on the pooled feature map to obtain an activated feature map; wherein the output of the last layer of the convolutional neural network model as a filter is the local area characteristic feature vector of the plurality of prepregs, and the input of the first layer of the convolutional neural network model as a filter is each image block in the sequence of image blocks.
The convolutional neural network (Convolutional Neural Network, CNN) is an artificial neural network and has wide application in the fields of image recognition and the like. The convolutional neural network may include an input layer, a hidden layer, and an output layer, where the hidden layer may include a convolutional layer, a pooling layer, an activation layer, a full connection layer, etc., where the previous layer performs a corresponding operation according to input data, outputs an operation result to the next layer, and obtains a final result after the input initial data is subjected to a multi-layer operation.
The convolutional neural network model has excellent performance in the aspect of image local feature extraction by taking a convolutional kernel as a feature filtering factor, and has stronger feature extraction generalization capability and fitting capability compared with the traditional image feature extraction algorithm based on statistics or feature engineering.
Specifically, in step 140, the plurality of prepreg local area characteristic feature vectors are passed through a converter-based context encoder to obtain a decoded feature vector. The local area characteristic feature vectors of the prepregs represent high-dimensional implicit feature information of each image block, but relevance among the image blocks is not considered.
Here, the transducer-based context encoder is a neural network model that is capable of efficiently capturing long-range dependencies between different locations in a sequence, and that uses a self-attention mechanism to calculate the correlation of each location in the sequence with other locations, thereby generating an encoded representation that contains global context information. The context encoder based on the converter has higher parallelism, lower computational complexity and better generalization capability than the conventional recurrent or convolutional neural network.
Fig. 4 is a flowchart of a sub-step of step 140 in a method of preparing a phenolic composite material according to an embodiment of the present application, as shown in fig. 4, the step of passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain decoded feature vectors, including: 141 passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a plurality of context prepreg local area characteristic feature vectors; 142, respectively calculating the Gaussian regression uncertainty factors of the characteristic feature vectors of the local areas of the plurality of the contextual prepregs to obtain a plurality of Gaussian regression uncertainty factors; 143, weighting the plurality of context prepreg local area characteristic feature vectors based on the plurality of gaussian regression uncertainty factors respectively to obtain a plurality of weighted context prepreg local area characteristic feature vectors; and, 144 concatenating the plurality of weighted contextual prepreg local region characteristic feature vectors to obtain the decoded feature vector.
Wherein, fig. 5 is a flowchart of a sub-step of step 141 in the method for preparing a phenolic composite material according to an embodiment of the present application, and as shown in fig. 5, the step of passing the plurality of prepreg local area characteristic feature vectors through a context encoder based on a converter to obtain a plurality of context prepreg local area characteristic feature vectors includes: 1411, performing one-dimensional arrangement on the characteristic feature vectors of the local areas of the prepregs to obtain global feature vectors of the prepregs; 1412, calculating a product between the global prepreg feature vector and a transpose vector of each of the local prepreg region feature vectors to obtain a plurality of self-attention correlation matrices; 1413, respectively performing standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; 1414, obtaining a plurality of probability values by using a Softmax classification function for each normalized self-attention correlation matrix in the plurality of normalized self-attention correlation matrices; and 1415 weighting each of the plurality of prepreg local area characteristic feature vectors with each of the plurality of probability values as a weight to obtain the plurality of context prepreg local area characteristic feature vectors.
The context encoder aims to mine for hidden patterns between contexts in the word sequence, optionally the encoder comprises: CNN (Convolutional Neural Network ), recurrent NN (RecursiveNeural Network, recurrent neural network), language Model (Language Model), and the like. The CNN-based method has a better extraction effect on local features, but has a poor effect on Long-Term Dependency (Long-Term Dependency) problems in sentences, so Bi-LSTM (Long Short-Term Memory) based encoders are widely used. The repetitive NN processes sentences as a tree structure rather than a sequence, has stronger representation capability in theory, but has the weaknesses of high sample marking difficulty, deep gradient disappearance, difficulty in parallel calculation and the like, so that the repetitive NN is less in practical application. The transducer has a network structure with wide application, has the characteristics of CNN and RNN, has a better extraction effect on global characteristics, and has a certain advantage in parallel calculation compared with RNN (RecurrentNeural Network ).
Here, in consideration of source image noise introduced in the image acquisition process of the detected image of the prepreg, after image blocking is performed on the detected image of the prepreg, source image noise exists in each image block, after image semantic feature extraction is performed through a convolutional neural network model serving as a filter and context-dependent encoding of image semantic features is performed through a context encoder based on a converter, gaussian distribution error uncertainty of respective feature distribution is further introduced into a plurality of context prepreg local area feature vectors obtained through the context encoder based on the converter, and thus, in consideration of the decoded feature vectors obtained by directly cascading the plurality of context prepreg local area feature vectors, direct superposition of the gaussian distribution error uncertainty also causes decoding regression errors of the decoded feature vectors, and accuracy of decoding values obtained by the decoder of the decoded feature vectors is affected.
Based on this, in the technical solution of the present application, each of the plurality of contextual prepreg local area characteristic feature vectors is calculated separately, for example, asIs expressed as: calculating the Gaussian regression uncertainty factors of the characteristic feature vectors of the local areas of the plurality of the contextual prepregs according to the following optimization formula to obtain a plurality of Gaussian regression uncertainty factors; wherein, the optimization formula is:, wherein ,/>Is the length of the line feature vector, < >> and />The mean and variance of feature sets at each location in the local area characteristic feature vectors of the plurality of contextual prepregs, respectively, wherein ∈ ->Is the +.o.of the local area characteristic feature vector of the contextual prepreg>Characteristic value of individual position->Is the multiple Gaussian regression uncertainty factor, and +.>Is a logarithmic function with a base of 2.
Here, for the unknown regression of the decoded feature vector, which may be caused by the distribution uncertainty information of the integrated feature set of each of the local regional feature vectors of the plurality of local regional feature vectors of the contextual prepreg, the scalar measurement of the statistical characteristics of the feature set is performed by using the mean value and the variance as the statistical quantization parameters, so that the normal distribution cognitive mode represented by the feature error is expanded to an unknown distribution regression mode, and the migration learning based on the natural distribution transfer on the feature set scale is realized, so that the decoded feature vector is obtained by weighting each local regional feature vector of the contextual prepreg by the gaussian regression uncertainty factors and cascading the weighted local regional feature vector of the contextual prepreg, and the uncertainty correction based on the self calibration of each local regional feature vector of the contextual prepreg when the decoded feature vector is formed is realized, so that the decoding regression error existing in the decoded feature vector is corrected, and the accuracy of the decoding value obtained by the decoder of the decoded feature vector is improved.
Further, concatenating the plurality of weighted contextual prepreg local region characteristic feature vectors to obtain the decoded feature vector, comprising: pre-weighting the plurality of weighted contexts in a cascading formulaCascading characteristic feature vectors of the local areas of the immersed material to obtain the decoding feature vectors; wherein, the cascade formula is: wherein , />Representing the local regional characteristic feature vector of the plurality of weighted contextual prepregs, +.>Representing a cascade function->Representing the decoded feature vector.
Specifically, in step 150, the decoded feature vector is passed through a decoder to obtain a decoded value, which is used to represent a recommended cure temperature value. The decoded feature vector is then passed through a decoder to obtain a decoded value, which is used to represent the recommended cure temperature value. The decoder is a neural network model capable of generating output data according to input data, and can automatically learn related rules from the input data and realize data output. In practical application, the curing temperature can be adjusted according to the decoding value, so that the cost and error of manual intervention are reduced.
Wherein the decoding feature vector is passed through a decoder to obtain a decoded value, the decoded value being used to represent a recommended cure temperature value, comprising: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein, the decoding formula is:，/>representing said decoded feature vector,/->Representation ofDecoding value->Representing a weight matrix, +.>Representing the bias vector +_>Representing a matrix multiplication.
In summary, a method 100 for preparing a phenolic composite material in accordance with embodiments of the present application is illustrated that acquires a test image of a prepreg; and the detection image of the prepreg is subjected to feature mining and capturing based on an image processing technology of deep learning, and the optimal curing temperature is automatically recommended through decoding processing, so that the quality and consistency of the phenolic composite material are improved, and the cost and error of manual intervention are reduced.
In one embodiment of the present application, fig. 6 is a block diagram of a phenolic composite preparation system according to an embodiment of the present application. As shown in fig. 6, a phenolic composite preparation system 200 according to an embodiment of the present application includes: an image acquisition module 210 for acquiring a detection image of the prepreg; an image blocking module 220, configured to perform image blocking processing on the detected image of the prepreg to obtain a sequence of image blocks; a feature extraction module 230, configured to pass each image block in the sequence of image blocks through a convolutional neural network model serving as a filter to obtain a plurality of prepreg local area characteristic feature vectors; a context encoding module 240 for passing the plurality of prepreg local region characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and a decoding module 250 for passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being indicative of a recommended cure temperature value.
In a specific example, in the above phenolic composite preparation system, the image blocking module is configured to: and uniformly partitioning the detected image of the prepreg to obtain the sequence of image blocks, wherein each image block in the sequence of image blocks has the same size.
In a specific example, in the above phenolic composite preparation system, the feature extraction module is configured to: each layer of the convolutional neural network model used as the filter performs the following steps on input data in forward transfer of the layer: carrying out convolution processing on the input data to obtain a convolution characteristic diagram; carrying out mean pooling treatment based on a feature matrix on the convolution feature map to obtain a pooled feature map; performing nonlinear activation on the pooled feature map to obtain an activated feature map; wherein the output of the last layer of the convolutional neural network model as a filter is the local area characteristic feature vector of the plurality of prepregs, and the input of the first layer of the convolutional neural network model as a filter is each image block in the sequence of image blocks.
In a specific example, in the above phenolic composite preparation system, the context encoding module includes: an encoding unit for passing the plurality of prepreg local area characteristic feature vectors through a context encoder based on a converter to obtain a plurality of context prepreg local area characteristic feature vectors; an optimizing unit, configured to calculate gaussian regression uncertainty factors of the local area characteristic feature vectors of the plurality of context prepregs respectively to obtain a plurality of gaussian regression uncertainty factors; a weighting unit, configured to weight the local area characteristic feature vectors of the context prepreg based on the gaussian regression uncertainty factors to obtain weighted local area characteristic feature vectors of the context prepreg; and a concatenation unit, configured to concatenate the plurality of weighted context prepreg local area characteristic feature vectors to obtain the decoded feature vector.
In a specific example, in the above phenolic composite preparation system, the coding unit includes: a vector construction subunit, configured to perform one-dimensional arrangement on the characteristic feature vectors of the local areas of the plurality of prepregs to obtain a global feature vector of the prepreg; a self-attention subunit, configured to calculate a product between the global prepreg feature vector and a transpose vector of each prepreg partial region feature vector in the plurality of prepreg partial region feature vectors to obtain a plurality of self-attention correlation matrices; the normalization subunit is used for respectively performing normalization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of normalized self-attention correlation matrices; the attention calculating subunit is used for obtaining a plurality of probability values through a Softmax classification function by each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and an attention applying subunit configured to weight each of the plurality of prepreg local area characteristic feature vectors with each of the plurality of probability values as a weight to obtain the plurality of context prepreg local area characteristic feature vectors.
In a specific example, in the above phenolic composite preparation system, the optimizing unit is configured to: calculating the Gaussian regression uncertainty factors of the characteristic feature vectors of the local areas of the plurality of the contextual prepregs according to the following optimization formula to obtain a plurality of Gaussian regression uncertainty factors; wherein, the optimization formula is:, wherein ,/>Is the length of the line feature vector, < >> and />The mean and variance of feature sets at each location in the local area characteristic feature vectors of the plurality of contextual prepregs, respectively, wherein ∈ ->Is the +.o.of the local area characteristic feature vector of the contextual prepreg>Characteristic value of individual position->Is the multiple Gaussian regression uncertainty factor, and +.>Is a logarithmic function with a base of 2.
In a specific example, in the above phenolic composite preparation system, the cascade unit is configured to: cascading the plurality of weighted context prepreg local area characteristic feature vectors with the following cascading formula to obtain the decoding feature vector; wherein, the cascade formula is: wherein , />Representing the local regional characteristic feature vector of the plurality of weighted contextual prepregs, +. >Representing a cascade function->Representing the decoded feature vector.
In a specific example, in the above phenolic composite preparation system, the decoding module is configured to: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein, the decoding formula is:，/>representing said decoded feature vector,/->Representing the decoded value->Representing a weight matrix, +.>Representing the bias vector +_>Representing a matrix multiplication.
Here, it will be understood by those skilled in the art that the specific functions and operations of the respective units and modules in the above-described phenolic composite material preparation system have been described in detail in the above description of the phenolic composite material preparation method with reference to fig. 1 to 5, and thus, repetitive descriptions thereof will be omitted.
As described above, the phenolic composite preparation system 200 according to the embodiments of the present application may be implemented in various terminal devices, such as a server or the like for the preparation of phenolic composite. In one example, the phenolic composite preparation system 200 according to embodiments of the present application may be integrated into the terminal device as one software module and/or hardware module. For example, the phenolic composite preparation system 200 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the phenolic composite preparation system 200 could equally be one of the numerous hardware modules of the end device.
Alternatively, in another example, the phenolic composite preparation system 200 and the end device may be separate devices, and the phenolic composite preparation system 200 may be connected to the end device via a wired and/or wireless network and communicate the interactive information in accordance with a agreed data format.
The present application also provides a computer program product comprising instructions which, when executed, cause an apparatus to perform operations corresponding to the above-described methods.
In one embodiment of the present application, there is also provided a computer readable storage medium storing a computer program for executing the above-described method.
It should be appreciated that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the forms of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects may be utilized. Furthermore, the computer program product may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
Methods, systems, and computer program products of embodiments of the present application are described in terms of flow diagrams and/or block diagrams. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.
The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (10)
1. A method of preparing a phenolic composite material, comprising: obtaining a detection image of the prepreg; performing image blocking processing on the detection image of the prepreg to obtain a sequence of image blocks; passing each image block in the sequence of image blocks through a convolutional neural network model serving as a filter to obtain a plurality of prepreg local area characteristic feature vectors; passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being indicative of a recommended cure temperature value.
2. The method of preparing a phenolic composite material according to claim 1, wherein image blocking the detected image of the prepreg to obtain a sequence of image blocks comprises: and uniformly partitioning the detected image of the prepreg to obtain the sequence of image blocks, wherein each image block in the sequence of image blocks has the same size.
3. The method of preparing a phenolic composite material of claim 2, wherein passing each image block in the sequence of image blocks through a convolutional neural network model as a filter to obtain a plurality of prepreg local area characteristic feature vectors, comprising: each layer of the convolutional neural network model used as the filter performs the following steps on input data in forward transfer of the layer: carrying out convolution processing on the input data to obtain a convolution characteristic diagram; carrying out mean pooling treatment based on a feature matrix on the convolution feature map to obtain a pooled feature map; non-linear activation is carried out on the pooled feature map so as to obtain an activated feature map; wherein the output of the last layer of the convolutional neural network model as a filter is the local area characteristic feature vector of the plurality of prepregs, and the input of the first layer of the convolutional neural network model as a filter is each image block in the sequence of image blocks.
4. A method of preparing a phenolic composite material in accordance with claim 3, wherein passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain a decoded feature vector comprises: passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain a plurality of context prepreg local area characteristic feature vectors; respectively calculating Gaussian regression uncertainty factors of characteristic feature vectors of the local areas of the plurality of context prepregs to obtain a plurality of Gaussian regression uncertainty factors; weighting the plurality of context prepreg local area characteristic feature vectors based on the plurality of gaussian regression uncertainty factors respectively to obtain a plurality of weighted context prepreg local area characteristic feature vectors; and concatenating the plurality of weighted context prepreg local region characteristic feature vectors to obtain the decoded feature vector.
5. The method of preparing a phenolic composite of claim 4, wherein passing the plurality of prepreg local area characteristic feature vectors through a transducer-based context encoder to obtain a plurality of context prepreg local area characteristic feature vectors comprises: performing one-dimensional arrangement on the characteristic feature vectors of the local areas of the prepregs to obtain global feature vectors of the prepregs; calculating the product between the global prepreg characteristic vector and the transpose vector of each local prepreg regional characteristic vector in the local prepreg regional characteristic vectors to obtain a plurality of self-attention correlation matrices; respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices; obtaining a plurality of probability values by using a Softmax classification function through each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and weighting each prepreg local area characteristic feature vector in the prepreg local area characteristic feature vectors by taking each probability value in the probability values as a weight so as to obtain the context prepreg local area characteristic feature vectors.
6. The method of preparing a phenolic composite of claim 5, wherein separately calculating gaussian regression uncertainty factors for the plurality of contextual prepreg local region characteristic feature vectors to obtain a plurality of gaussian regression uncertainty factors comprises: calculating the Gaussian regression uncertainty factors of the characteristic feature vectors of the local areas of the plurality of the contextual prepregs according to the following optimization formula to obtain a plurality of Gaussian regression uncertainty factors; wherein, the optimization formula is:, wherein ,/>Is the length of the line feature vectorDegree (f)> and />The mean and variance of feature sets at each location in the local area characteristic feature vectors of the plurality of contextual prepregs, respectively, wherein ∈ ->Is the +.o.of the local area characteristic feature vector of the contextual prepreg>Characteristic value of individual position->Is the multiple Gaussian regression uncertainty factor, and +.>Is a logarithmic function with a base of 2.
7. The method of preparing a phenolic composite material of claim 6, wherein concatenating the plurality of weighted contextual prepreg local region characteristic feature vectors to obtain the decoded feature vector comprises: cascading the plurality of weighted context prepreg local area characteristic feature vectors with the following cascading formula to obtain the decoding feature vector; wherein, the cascade formula is: wherein , />Representing the local regional characteristic feature vector of the plurality of weighted contextual prepregs, +.>Representing a cascade function->Representing the decoded feature vector.
8. The method of preparing a phenolic composite material in accordance with claim 7, wherein passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being indicative of a recommended cure temperature value, comprises: performing decoding regression on the decoding eigenvector with a decoding formula using the decoder to obtain the decoded value; wherein, the decoding formula is:，/>representing said decoded feature vector,/->Representing the decoded value->The weight matrix is represented by a matrix of weights,representing the bias vector +_>Representing a matrix multiplication.
9. A phenolic composite preparation system, comprising: the image acquisition module is used for acquiring a detection image of the prepreg; the image blocking module is used for carrying out image blocking processing on the detection image of the prepreg to obtain a sequence of image blocks; the feature extraction module is used for enabling each image block in the sequence of the image blocks to pass through a convolutional neural network model serving as a filter so as to obtain a plurality of prepreg local area characteristic feature vectors; a context encoding unit for passing the plurality of prepreg local area characteristic feature vectors through a converter-based context encoder to obtain a decoded feature vector; and a decoding module for passing the decoded feature vector through a decoder to obtain a decoded value, the decoded value being indicative of a recommended cure temperature value.
10. The phenolic composite preparation system of claim 9, wherein the image blocking module is configured to: and uniformly partitioning the detected image of the prepreg to obtain the sequence of image blocks, wherein each image block in the sequence of image blocks has the same size.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310554023.6A CN116461120B (en) | 2023-05-17 | 2023-05-17 | Preparation method and system of phenolic composite material |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310554023.6A CN116461120B (en) | 2023-05-17 | 2023-05-17 | Preparation method and system of phenolic composite material |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116461120A true CN116461120A (en) | 2023-07-21 |
| CN116461120B CN116461120B (en) | 2023-10-03 |
Family
ID=87182641
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310554023.6A Active CN116461120B (en) | 2023-05-17 | 2023-05-17 | Preparation method and system of phenolic composite material |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116461120B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109348708A (en) * | 2015-08-04 | 2019-02-15 | 三菱化学株式会社 | Fibre reinforced plastics and its manufacturing method |
| WO2019170286A1 (en) * | 2018-03-07 | 2019-09-12 | MTU Aero Engines AG | Method and device for self-optimising, additive production of components |
| US20200324494A1 (en) * | 2019-04-10 | 2020-10-15 | The Boeing Company | Fabrication optimization for composite parts |
| KR102272100B1 (en) * | 2020-07-28 | 2021-07-02 | 현대제철 주식회사 | Method and system for surface crack detection of continuous casting using deep learning images |
| CN113205181A (en) * | 2021-05-20 | 2021-08-03 | 上海电力大学 | Graph combination optimization problem solving method based on deep graph learning |
| CN114347505A (en) * | 2021-11-24 | 2022-04-15 | 航天海鹰(镇江)特种材料有限公司 | Method for separately controlling curing temperature of ultra-thickness ratio composite material workpiece |
| CN115860271A (en) * | 2023-02-21 | 2023-03-28 | 杭州唛扑网络科技有限公司 | System and method for managing art design scheme |
| CN116003126A (en) * | 2023-03-22 | 2023-04-25 | 珠海和泽科技有限公司 | Preparation method and system of electrostatic chuck surface ceramic material |
| CN116119589A (en) * | 2023-02-21 | 2023-05-16 | 德玛克(浙江)精工科技有限公司 | Liquid PET bottled drink purging, filling and rotating integrated machine and method thereof |
-
2023
- 2023-05-17 CN CN202310554023.6A patent/CN116461120B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109348708A (en) * | 2015-08-04 | 2019-02-15 | 三菱化学株式会社 | Fibre reinforced plastics and its manufacturing method |
| WO2019170286A1 (en) * | 2018-03-07 | 2019-09-12 | MTU Aero Engines AG | Method and device for self-optimising, additive production of components |
| US20200324494A1 (en) * | 2019-04-10 | 2020-10-15 | The Boeing Company | Fabrication optimization for composite parts |
| KR102272100B1 (en) * | 2020-07-28 | 2021-07-02 | 현대제철 주식회사 | Method and system for surface crack detection of continuous casting using deep learning images |
| CN113205181A (en) * | 2021-05-20 | 2021-08-03 | 上海电力大学 | Graph combination optimization problem solving method based on deep graph learning |
| CN114347505A (en) * | 2021-11-24 | 2022-04-15 | 航天海鹰(镇江)特种材料有限公司 | Method for separately controlling curing temperature of ultra-thickness ratio composite material workpiece |
| CN115860271A (en) * | 2023-02-21 | 2023-03-28 | 杭州唛扑网络科技有限公司 | System and method for managing art design scheme |
| CN116119589A (en) * | 2023-02-21 | 2023-05-16 | 德玛克(浙江)精工科技有限公司 | Liquid PET bottled drink purging, filling and rotating integrated machine and method thereof |
| CN116003126A (en) * | 2023-03-22 | 2023-04-25 | 珠海和泽科技有限公司 | Preparation method and system of electrostatic chuck surface ceramic material |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116461120B (en) | 2023-10-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wei et al. | PRI modulation recognition based on squeeze-and-excitation networks | |
| CN108197087A (en) | Character code recognition methods and device | |
| CN113988449B (en) | Wind power prediction method based on transducer model | |
| CN107291836A (en) | A kind of Chinese text summary acquisition methods based on semantic relevancy model | |
| CN116336400B (en) | Baseline detection method for oil and gas gathering and transportation pipeline | |
| CN116046810A (en) | Nondestructive testing method based on RPC cover plate damage load | |
| CN115951883B (en) | Service component management system of distributed micro-service architecture and method thereof | |
| CN116245513A (en) | Automatic operation and maintenance system and method based on rule base | |
| CN116577464A (en) | Intelligent monitoring system and method for atmospheric pollution | |
| CN117270082A (en) | Precipitation prediction system and method based on multi-sensor data fusion | |
| CN116461120B (en) | Preparation method and system of phenolic composite material | |
| CN116736703A (en) | Intelligent monitoring system and method for sewage treatment | |
| CN116301166B (en) | Non-band gap reference voltage source circuit | |
| CN116108143B (en) | Digital economic monitoring method and system based on block chain technology | |
| CN116129251A (en) | Intelligent manufacturing method and system for office desk and chair | |
| CN117158923A (en) | Remote home-care monitoring method based on meta universe | |
| CN113656066B (en) | Clone code detection method based on feature alignment | |
| CN116393188B (en) | Microfluidic chip and method suitable for capturing circulating tumor cells | |
| CN114817562A (en) | Knowledge graph construction method, knowledge graph training method, information recommendation method and information recommendation device | |
| Kim et al. | SIGNAL FAULT IDENTIFICATION IN NUCLEAR POWER PLANTS BASED ON DEEP NEURAL NETWORKS. | |
| CN113886520A (en) | Code retrieval method and system based on graph neural network and computer readable storage medium | |
| CN116718894B (en) | Circuit stability test method and system for corn lamp | |
| CN116304603A (en) | Pattern recognition method based on double-contrast architecture | |
| CN116663004B (en) | Binary function similarity detection method and system based on graph transformations | |
| CN116678004A (en) | RCO waste gas treatment system and method thereof |
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 |