Thick copper high-density circuit board and method based on fast charging and battery life

Abstract

The present application discloses a thick copper high-density circuit board and method based on fast charging and battery life, which collects the circuit board drilling distribution image through a camera, extracts the circuit board drilling distribution design image from the database, and then introduces an image processing and analysis algorithm based on artificial intelligence and machine vision in the back end to analyze the circuit board drilling distribution image and the circuit board drilling distribution design image, so as to learn and characterize the two-way fine-grained global joint perception semantics between the detection drilling distribution characteristics and the design drilling distribution characteristics, so as to automatically identify the hole deviation. In this way, the intelligent identification and detection of hole deviation can be realized in the manufacturing process of thick copper high-density circuit boards, so as to timely discover and correct the hole position deviation problem in the circuit board manufacturing, reduce the scrap rate caused by the hole position deviation, and ensure the reliability of the circuit board preparation.

Description

Technical Field

[0001] The present application relates to the field of intelligent identification, and more specifically, to a thick copper high-density circuit board and method based on fast charging and battery life. Background Art

[0002] As electronic devices develop towards miniaturization and high performance, the design and manufacture of circuit boards face increasingly higher requirements. Especially in portable electronic devices, such as smartphones and tablets, circuit boards not only need to carry more functional components, but also need to have efficient energy management and longer use time. Therefore, thick copper high-density circuit boards have become one of the key technologies to achieve these goals.

[0003] Thick copper circuit boards can withstand greater currents than traditional circuit boards, and due to the increased thickness of the copper layer, heat dissipation performance is also improved, which is especially important for supporting fast charging functions. In addition, high-density wiring allows the circuit board to integrate more electronic components in a limited space, thereby reducing the size of the device and improving integration.

[0004] Chinese patent CN117395877A proposes a method for making a local high-density thick copper circuit board, which can start from a copper-clad laminate, produce a high-density circuit pattern, form a sub-board graphic board, and then press the insulating dielectric layer and cut it into a high-density sub-core board. In addition, starting from a thick copper-clad laminate, a thick copper circuit pattern is produced, which includes an embedded area that matches the high-density sub-core board. Then, the high-density sub-core board is embedded in the embedded area of ​​the thick copper graphic board, and a thick copper insulating dielectric layer and a surface copper foil layer are superimposed and pressed to form a laminate. Subsequently, holes are drilled in the thick copper circuit pattern area and the high-density circuit pattern area, and the hole metallization processing is performed to form a through hole, and a current shunt is set.

[0005] Although the above-mentioned method for making a local high-density thick copper circuit board can embed a high-density sub-core board into a thick copper board to form a multifunctional combined circuit board, it effectively solves the problems of complex and difficult processing in traditional designs, as well as low space utilization and difficulty in matching expansion and contraction caused by the layered design of the high-density interconnect layer and the thick copper layer of the entire board. However, this production method also has some potential defects or shortcomings. For example, in the process of drilling holes in the thick copper circuit graphic area and the high-density circuit graphic area, there is a lack of effective detection and identification means for hole position deviation, and hole position deviation will lead to poor electrical connection, affecting the functionality and reliability of the circuit board. In addition, the above-mentioned production method does not involve the battery status control function, and it is difficult to maintain the balance between fast charging and endurance of the thick copper high-density circuit board, affecting the service life of the circuit board and the battery.

[0006] Therefore, an optimized manufacturing solution for thick copper high-density circuit boards is desired. Summary of the invention

[0007] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a thick copper high-density circuit board and method based on fast charging and endurance, which collects the circuit board drilling distribution image through a camera, and extracts the circuit board drilling distribution design image from the database, and then introduces the image processing and analysis algorithm based on artificial intelligence and machine vision at the back end to analyze the circuit board drilling distribution image and the circuit board drilling distribution design image, so as to learn and characterize the bidirectional fine-grained global joint perception semantics between the detection drilling distribution characteristics and the design drilling distribution characteristics, so as to automatically identify the hole deviation. In this way, the intelligent identification and detection of hole deviation can be realized in the manufacturing process of thick copper high-density circuit boards, so as to timely discover and correct the hole position deviation problem in the circuit board manufacturing, reduce the scrap rate caused by the hole position deviation, and ensure the reliability of the circuit board preparation.

[0008] According to one aspect of the present application, a method for a thick copper high-density circuit board based on fast charging and battery life is provided, comprising:

[0009] Taking the copper clad board and making a high density circuit pattern to form a daughter board pattern board;

[0010] Laminating an insulating dielectric layer to a surface of the daughterboard graphic board to form a laminated daughterboard;

[0011] Cutting the laminated sub-boards to form high-density sub-core boards;

[0012] Taking a thick copper clad board and making a thick copper circuit pattern to form a thick copper pattern board;

[0013] Placing the high-density sub-core board in the embedding area of ​​the thick copper graphic board, and laminating it with the thick copper insulating medium layer and the surface copper foil layer to form a laminate;

[0014] Drilling holes in the area where the thick copper circuit pattern of the laminate is located, and drilling holes in the area where the high-density circuit pattern of the laminate is located, and then performing hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes;

[0015] A current shunt is provided between the thick copper circuit pattern via hole and the high-density circuit pattern via hole to form a high-density thick copper circuit board;

[0016] Installing a power management chip on the high-density thick copper circuit board, wherein the power management chip is used to monitor the battery status;

[0017] The method comprises drilling holes in the area where the thick copper circuit pattern of the laminate is located, drilling holes in the area where the high-density circuit pattern of the laminate is located, and then performing hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes, including:

[0018] Acquire a circuit board drilling distribution image captured by a camera;

[0019] Extracting a circuit board drilling distribution design image from a database;

[0020] Performing feature extraction and joint analysis based on fine-grained deconstruction on the circuit board drilling distribution image and the circuit board drilling distribution design image to obtain a drilling distribution-drilling design joint perception alignment feature map;

[0021] Based on the drilling distribution-drilling design joint perception alignment feature map, hole deviation identification is performed to determine whether drilling deviation occurs.

[0022] According to another aspect of the present application, a thick copper high-density circuit board based on fast charging and long battery life is provided, and the thick copper high-density circuit board based on fast charging and long battery life is manufactured by the method for a thick copper high-density circuit board based on fast charging and long battery life as described in any one of claims 1-8.

[0023] Compared with the prior art, the present application provides a thick copper high-density circuit board and method based on fast charging and battery life, which collects the circuit board drilling distribution image through a camera, extracts the circuit board drilling distribution design image from the database, and then introduces an image processing and analysis algorithm based on artificial intelligence and machine vision at the back end to analyze the circuit board drilling distribution image and the circuit board drilling distribution design image, so as to learn and characterize the bidirectional fine-grained global joint perception semantics between the detection drilling distribution characteristics and the design drilling distribution characteristics, so as to automatically identify the hole deviation. In this way, the intelligent identification and detection of hole deviation can be realized in the manufacturing process of thick copper high-density circuit boards, so as to timely discover and correct the hole position deviation problem in circuit board manufacturing, reduce the scrap rate caused by hole position deviation, and ensure the reliability of circuit board preparation. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0025] Figure 1 It is a flow chart of a method for a thick copper high-density circuit board based on fast charging and battery life according to an embodiment of the present application;

[0026] Figure 2 A data flow diagram of a method for a thick copper high-density circuit board based on fast charging and battery life according to an embodiment of the present application;

[0027] Figure 3 It is a flowchart of sub-step S6 of the method for a thick copper high-density circuit board based on fast charging and long battery life according to an embodiment of the present application;

[0028] Figure 4 This is a flowchart of sub-step S63 of the method for thick copper high-density circuit board based on fast charging and long battery life according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0030] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0031] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0032] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0033] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0034] In the technical solution of the present application, a method for a thick copper high-density circuit board based on fast charging and battery life is proposed. Figure 1This is a flow chart of a method for a thick copper high-density circuit board based on fast charging and long battery life according to an embodiment of the present application. Figure 2 Schematic diagram of data flow of the method for fast charging and long-lasting thick copper high-density circuit board according to the embodiment of the present application. Figure 1 and Figure 2 As shown, according to the embodiment of the present application, the method for a thick copper high-density circuit board based on fast charging and battery life includes the following steps: S1, taking a copper clad board and making a high-density circuit pattern to form a daughter board graphic board; S2, laminating an insulating dielectric layer to the surface of the daughter board graphic board to form a laminated daughter board; S3, cutting the laminated daughter board to form a high-density daughter core board; S4, taking a thick copper clad board and making a thick copper circuit pattern to form a thick copper graphic board; S5, placing the high-density daughter core board in the embedding area of ​​the thick copper graphic board, and bonding it with the thick copper insulating dielectric layer and the surface copper The foil layers are pressed to form a laminate; S6, drilling holes in the area where the thick copper circuit pattern of the laminate is located, and drilling holes in the area where the high-density circuit pattern of the laminate is located, and then performing hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes; S7, arranging a current shunt between the thick copper circuit pattern via holes and the high-density circuit pattern via holes to form a high-density thick copper circuit board; S8, installing a power management chip on the high-density thick copper circuit board, and the power management chip is used to monitor the battery status.

[0035] In particular, the S1, takes a copper clad laminate and makes a high-density line pattern to form a daughterboard pattern board. The daughterboard pattern board is an important concept in electronic manufacturing, which refers to the process of making high-density line patterns on a copper clad laminate (copper foil laminated on an insulating substrate). These line patterns form the circuit pattern of the daughterboard, defining the interconnection and function of the electronic components.

[0036] In particular, in S2, an insulating dielectric layer is laminated to the surface of the daughterboard graphic board to form a laminated daughterboard. Laminated daughterboard is a technology in electronic manufacturing that involves bonding multiple layers of materials together to form a single, multi-layer structure. In daughterboard manufacturing, lamination is generally used to bond a copper clad laminate to an insulating substrate. By laminating daughterboards, daughterboards with complex structures and high reliability can be created to achieve various functions of electronic devices.

[0037] In particular, the S3, cutting the laminated sub-boards to form high-density sub-core boards. By cutting the laminated sub-boards, a high-density sub-core board with a desired shape, size and properties can be created.

[0038] In particular, in S4, a thick copper clad copper board is taken and a thick copper circuit pattern is made to form a thick copper pattern board. The main feature of the thick copper pattern board is its thick copper layer, which provides higher current carrying capacity and heat dissipation capacity. In particular, the thick copper circuit pattern includes an embedded area, the size of the embedded area matches the size of the high-density sub-core board, and the depth of the embedded area matches the depth of the high-density sub-core board.

[0039] In particular, in S5, the high-density sub-core board is placed in the embedded area of ​​the thick copper graphic board, and is pressed together with the thick copper insulating medium layer and the surface copper foil layer to form a laminated board. The laminated board is a plate-like structure formed by bonding multiple layers of materials together to form a composite material.

[0040] In particular, the S6 is to drill holes in the area where the thick copper circuit pattern of the laminate is located, and to drill holes in the area where the high-density circuit pattern of the laminate is located, and then perform hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes. In the process of drilling holes in the area where the thick copper circuit pattern of the laminate is located, and drilling holes in the area where the high-density circuit pattern of the laminate is located, and then perform hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes, the traditional hole deviation identification method mostly relies on manual visual inspection, which is inefficient and prone to errors. In a specific example of the present application, Figure 3 As shown, the S6 includes: S61, acquiring a circuit board drilling distribution image captured by a camera; S62, extracting a circuit board drilling distribution design image from a database; S63, performing feature extraction and joint analysis based on fine-grained deconstruction on the circuit board drilling distribution image and the circuit board drilling distribution design image to obtain a drilling distribution-drilling design joint perception alignment feature map; S64, performing hole deviation identification based on the drilling distribution-drilling design joint perception alignment feature map to determine whether drilling deviation occurs.

[0041] Specifically, the S61 and S62 obtain the circuit board drilling distribution image collected by the camera; and extract the circuit board drilling distribution design image from the database. Among them, the drilling distribution image shows the positions of all the holes on the circuit board, can quantify the spacing between the holes, and can identify the hole deviation by comparing with the circuit board drilling distribution design image. The hole deviation refers to the deviation between the position of the hole and the design position. In the technical solution of the present application, the circuit board drilling distribution image and the circuit board drilling distribution design image are compared and analyzed to automatically identify the hole deviation. In this way, the intelligent identification and detection of hole deviation can be realized in the manufacturing process of thick copper and high-density circuit boards, so as to timely discover and correct the hole position deviation problem in circuit board manufacturing, reduce the scrap rate caused by hole position deviation, and ensure the reliability of circuit board preparation.

[0042] Specifically, the S63 performs feature extraction and joint analysis based on fine-grained deconstruction on the circuit board drilling distribution image and the circuit board drilling distribution design image to obtain a drilling distribution-drilling design joint perception alignment feature map. In a specific example of the present application, Figure 4 As shown, the S63 includes: S631, scaling the circuit board drilling distribution image to obtain a scaled circuit board drilling distribution image having the same size as the circuit board drilling distribution design image; S632, respectively inputting the scaled circuit board drilling distribution image and the circuit board drilling distribution design image into a drilling distribution feature extractor based on a deep neural network model to obtain a circuit board drilling distribution detection feature map and a circuit board drilling distribution design feature map; S633, inputting the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map into a bidirectional global attention joint perception module based on fine-grained deconstruction to obtain the drilling distribution-drilling design joint perception alignment feature map.

[0043] More specifically, in S631, the circuit board drilling distribution image is scaled to obtain a scaled circuit board drilling distribution image having the same size as the circuit board drilling distribution design image. Considering that when detecting and comparing the circuit board actually produced with the design drawing, the image actually captured may change in size due to the angle, distance or other factors of the camera. If the size is not standardized, directly comparing the two images may lead to misjudgment, because even a circuit board without manufacturing defects may be mistakenly marked as unqualified due to inconsistent image sizes. Based on this, the circuit board drilling distribution image is further scaled to obtain a scaled circuit board drilling distribution image having the same size as the circuit board drilling distribution design image. By scaling the size, it is ensured that the circuit board drilling distribution image actually detected can be consistent in size with the designed circuit board drilling distribution design image, and it is possible to more accurately detect whether the drilling distribution on the circuit board meets the design requirements, thereby improving the accuracy of hole deviation detection.

[0044] More specifically, in S632, the scaled circuit board drilling distribution image and the circuit board drilling distribution design image are respectively input into a drilling distribution feature extractor based on a deep neural network model to obtain a circuit board drilling distribution detection feature map and a circuit board drilling distribution design feature map. Wherein, the drilling distribution feature extractor based on a deep neural network model is a drilling distribution feature extractor based on a hole convolutional neural network model. That is, in the technical solution of the present application, the scaled circuit board drilling distribution image and the circuit board drilling distribution design image are respectively input into a drilling distribution feature extractor based on a hole convolutional neural network model for feature mining, so as to respectively extract the detection features and design features of the scaled circuit board drilling distribution image and the circuit board drilling distribution design map with information on the circuit board drilling distribution, thereby obtaining a circuit board drilling distribution detection feature map and a circuit board drilling distribution design feature map. It is worth mentioning that a hole convolutional neural network (DCNN) is a convolutional neural network (CNN) that uses a hole convolution operation to expand the receptive field while maintaining spatial resolution. Wherein, the receptive field refers to the range of influence of neurons in a convolutional neural network on a specific area in an input image. Atrous convolution expands the receptive field by introducing holes without increasing the size of the convolution kernel.

[0045] More specifically, in S633, the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map are input into a two-way global attention joint perception module based on fine-grained deconstruction to obtain the drilling distribution-drilling design joint perception alignment feature map. It should be understood that since the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map respectively contain the detection features and design features of the circuit board drilling distribution, in order to be able to capture the subtle differences between the actual drilling distribution and the design more carefully, it is necessary to perform fine-grained joint two-way interactive perception of the drilling distribution detection features and design features of the circuit board. Based on this, in the technical solution of the present application, the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map are further input into a two-way global attention joint perception module based on fine-grained deconstruction to obtain the drilling distribution-drilling design joint perception alignment feature map. The bidirectional global attention joint perception module based on fine-grained deconstruction aims to improve the model's semantic joint perception ability between detection features and design features of circuit board drilling distribution through feature fine-grained deconstruction, global fine-grained interaction based on attention mechanism, feature coupling and feature fusion, thereby facilitating better perception and identification of drilling offset.

[0046] In an embodiment of the present application, the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map are input into a bidirectional global attention joint perception module based on fine-grained deconstruction to obtain the drilling distribution-drilling design joint perception alignment feature map, including: first, the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map are subjected to feature fine-grained deconstruction to obtain a set of local feature matrices of the circuit board drilling distribution detection channel dimension and a set of local feature matrices of the circuit board drilling distribution design channel dimension; it should be understood that the process of fine-grained deconstruction along the channel dimension can allow the model to detect the circuit board drilling distribution. The detection feature map and the circuit board drilling distribution design feature map are segmented and semantically understood in more detail, so that more local details of the circuit board drilling distribution can be captured, which is helpful for the subsequent attention mechanism to more accurately identify the features of important areas and compare the detection drilling distribution and design drilling distribution features; then, each circuit board drilling distribution detection channel dimension local feature matrix in the set of the circuit board drilling distribution detection channel dimension local feature matrix is ​​used as the query feature matrix, and the set of the circuit board drilling distribution design channel dimension local feature matrix is ​​used as the set of key feature matrices, and the query feature matrix and the key feature matrix are combined. The collection inputs a one-way global attention interaction module based on the first converter structure to obtain a collection of one-way global attention optimized circuit board drilling distribution detection channel dimension local feature matrices; that is, after fine-grained deconstruction of the feature graph, the one-way global attention interaction module based on the converter structure can measure the correlation between different feature matrices by utilizing the multi-head attention mechanism in the Transformer structure, thereby strengthening the relationship between the circuit board drilling distribution detection features and the circuit board drilling distribution design features, so that the model can pay more attention to information-rich feature areas and ignore irrelevant background information, so that the model can be used in complex environments. More accurately identify key information, such as the position semantics and association relationship of drilling distribution, so as to identify and detect drilling offsets; similarly, each circuit board drilling distribution design channel dimension local feature matrix in the set of the circuit board drilling distribution design channel dimension local feature matrix is ​​used as a query feature matrix, and the set of the circuit board drilling distribution detection channel dimension local feature matrix is ​​used as a set of key feature matrices, and the query feature matrix and the set of the key feature matrix are input into a one-way global attention interaction module based on the second converter structure to obtain a set of one-way global attention optimized circuit board drilling distribution design channel dimension local feature matrices;In particular, in the technical solution of the present application, each local feature matrix of the circuit board drilling distribution detection channel dimension of the circuit board drilling distribution detection feature map can respectively perform one-way semantic interaction with all local feature matrices of the circuit board drilling distribution design channel dimension of the circuit board drilling distribution design feature map to more comprehensively capture the global semantic association information between each local feature matrix of the circuit board drilling distribution detection channel dimension of the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map. Similarly, each local feature matrix of the circuit board drilling distribution design channel dimension of the circuit board drilling distribution design feature map can respectively perform one-way semantic interaction with all local feature matrices of the circuit board drilling distribution detection channel dimension of the circuit board drilling distribution detection feature map to more comprehensively capture the global semantic association information between each local feature matrix of the circuit board drilling distribution design channel dimension of the circuit board drilling distribution detection feature map and the circuit board drilling distribution detection feature map. It should be understood that the effect of the one-way global attention interaction is to enhance the fine-grained channel relationship within the feature map, so that the model can pay more attention to the key information related to the hole position detection and drilling offset identification tasks in the feature map, improve the quality and discrimination of the feature representation, and thus help improve the performance of the model in subsequent tasks; further, the set of local feature matrices of the one-way global attention optimized circuit board drilling distribution detection channel dimension and the set of local feature matrices of the one-way global attention optimized circuit board drilling distribution design channel dimension are coupled along the channel dimension to obtain a one-way global interaction optimized circuit board drilling distribution detection feature map and a one-way global interaction optimized circuit board drilling distribution design feature map; through the coupling processing along the channel dimension, the local feature matrices of the circuit board drilling distribution detection and design optimized by attention can be recombined along the channel dimension to form a complete feature map, providing a complete feature representation for the final fusion; finally, the position-weighted sum between the one-way global interaction optimized circuit board drilling distribution detection feature map and the one-way global interaction optimized circuit board drilling distribution design feature map is calculated to obtain the drilling distribution-drilling design joint perception alignment feature map. Here, the weighted sum operation combines the key information in the two feature maps and generates a joint perception feature representation that contains rich interactive information between drilling distribution detection features and drilling distribution design features. In particular, the final generated "drilling distribution-drilling design joint perception alignment feature map" is a result that integrates the fine-grained semantic interactive information of drilling distribution in detection and design, which is conducive to more accurate fine-grained detection of hole deviation in the subsequent process. ;

[0047] Among them, each circuit board drilling distribution detection channel dimension local feature matrix in the set of the circuit board drilling distribution detection channel dimension local feature matrix is ​​used as a query feature matrix, and the set of the circuit board drilling distribution design channel dimension local feature matrix is ​​used as a set of key feature matrices, and the query feature matrix and the set of the key feature matrix are input into a one-way global attention interaction module based on a first converter structure to obtain a set of one-way global attention optimized circuit board drilling distribution detection channel dimension local feature matrix. The process includes: selecting a predetermined circuit board drilling distribution detection channel dimension local feature matrix from the set of the circuit board drilling distribution detection channel dimension local feature matrix as a query feature matrix; To query the feature matrix; calculate the product of the predetermined circuit board drilling distribution detection channel dimension local feature matrix and the transposed matrix of each circuit board drilling distribution design channel dimension local feature matrix in the set of the circuit board drilling distribution design channel dimension local feature matrix to obtain a set of circuit board drilling distribution detection-design channel dimension local semantic interaction feature matrices; divide each circuit board drilling distribution detection-design channel dimension local semantic interaction feature matrix in the set of the circuit board drilling distribution detection-design channel dimension local semantic interaction feature matrix by the square root of the scale of the predetermined circuit board drilling distribution detection channel dimension local feature matrix according to the position point, and then use The function performs soft maximum normalization processing on each feature matrix in the obtained set of feature matrices to obtain a set of local weight matrices of the circuit board drilling distribution detection channel dimension; using the set of local weight matrices of the circuit board drilling distribution detection channel dimension as weighted weights, calculate the position-weighted sum between each local feature matrix of the circuit board drilling distribution detection channel dimension in the set of local feature matrices of the circuit board drilling distribution detection channel dimension to obtain the one-way global attention optimization circuit board drilling distribution detection channel dimension local feature matrix. And, using each local feature matrix of the circuit board drilling distribution design channel dimension in the set of local feature matrices of the circuit board drilling distribution design channel dimension as a query feature matrix, using the set of local feature matrices of the circuit board drilling distribution detection channel dimension as a set of key feature matrices, inputting the query feature matrix and the set of key feature matrices into a one-way global attention interaction module based on a second converter structure to obtain a set of local feature matrices of the circuit board drilling distribution design channel dimension for one-way global attention optimization. The process includes: selecting a predetermined local feature matrix of the circuit board drilling distribution design channel dimension from the set of local feature matrices of the circuit board drilling distribution design channel dimension as a query feature matrix, and using the set of local feature matrices of the circuit board drilling distribution detection channel dimension as a key feature matrix. To query the feature matrix; calculate the product of the predetermined circuit board drilling distribution design channel dimension local feature matrix and the transposed matrix of each circuit board drilling distribution detection channel dimension local feature matrix in the set of the circuit board drilling distribution detection channel dimension local feature matrix to obtain a set of circuit board drilling distribution design-detection channel dimension local semantic interaction feature matrices; divide each circuit board drilling distribution design-detection channel dimension local semantic interaction feature matrix in the set of the circuit board drilling distribution design-detection channel dimension local semantic interaction feature matrix by the square root of the scale of the predetermined circuit board drilling distribution design channel dimension local feature matrix according to the position point, and then use The function performs soft maximum normalization processing on each feature matrix in the obtained set of feature matrices to obtain a set of local weight matrices of the circuit board drilling distribution design channel dimension; using the set of local weight matrices of the circuit board drilling distribution design channel dimension as weighted weights, calculates the position-weighted sum of the local feature matrices of the circuit board drilling distribution design channel dimension in the set of local feature matrices of the circuit board drilling distribution design channel dimension to obtain the unidirectional global attention optimized local feature matrix of the circuit board drilling distribution design channel dimension.

[0048] In summary, in the above embodiment, the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map are input into the fine-grained deconstruction-based bidirectional global attention joint perception module and processed by the following bidirectional global attention joint perception formula to obtain the drilling distribution-drilling design joint perception alignment feature map; wherein the bidirectional global attention joint perception formula is:

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055]

[0056] in, and respectively represent the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map, It is a feature fine-grained deconstruction operation. are respectively the first, second, and third components along the channel dimension in the circuit board drilling distribution detection feature map. and The local feature matrix of the circuit board drilling distribution detection channel dimension, are respectively the first, second, and third holes along the channel dimension in the circuit board drilling distribution design feature diagram. and The local feature matrix of the channel dimension for the design of the drilling distribution of the circuit board, For the The scale of the local feature matrix of the circuit board drilling distribution detection channel dimension, is matrix multiplication, for function, is the number of local feature matrices of the circuit board drilling distribution detection channel dimension in the circuit board drilling distribution detection feature map, For the The one-way global attention optimization circuit board drilling distribution detection channel dimension local feature matrix corresponding to the circuit board drilling distribution detection channel dimension local feature matrix, For the The one-way global attention optimization circuit board drilling distribution design channel dimension local feature matrix corresponding to the channel dimension local feature matrix, Indicates that the features are coupled along the channel dimension. and They are one-way global interactive optimization of circuit board drilling distribution detection feature map and one-way global interactive optimization of circuit board drilling distribution design feature map. and are weighted hyper parameters of the one-way global interactive optimization circuit board drilling distribution detection feature map and the one-way global interactive optimization circuit board drilling distribution design feature map, respectively. A joint perceptual alignment feature map is designed for the drilling distribution-drilling.

[0057] It is worth mentioning that in other examples of the present application, the following steps can be used to extract features from the circuit board drilling distribution image and the circuit board drilling distribution design image and perform joint analysis based on fine-grained deconstruction to obtain a drilling distribution-drilling design joint perception alignment feature map, for example: first, various feature extraction techniques, such as convolutional neural networks (CNN), local binary patterns (LBP) or scale-invariant feature transform (SIFT) are used to extract features of the circuit board drilling distribution image and the circuit board drilling distribution design image, and the extracted features are decomposed into finer-grained representations using mechanisms such as self-attention mechanisms or graph neural networks, and the fine-grained features from the circuit board drilling distribution image and the circuit board drilling distribution design image are jointly analyzed to reveal the potential relationship and interdependence between drilling distribution and drilling design, and the features obtained through the joint analysis are used to obtain the drilling distribution-drilling design joint perception alignment feature map.

[0058] Specifically, the S64 performs hole deviation identification based on the drilling distribution-drilling design joint perception alignment feature map to determine whether drilling deviation occurs. In a specific example of the present application, the drilling distribution-drilling design joint perception alignment feature map is input into a hole deviation identifier based on a classifier to obtain a recognition result, and the recognition result is used to indicate whether drilling deviation occurs. That is, the classification process is performed using the joint perception alignment semantics between the detection features of the hole distribution and the design features to automatically identify the hole deviation. In this way, the intelligent identification and detection of hole deviation can be realized in the manufacturing process of thick copper and high-density circuit boards, so as to promptly discover and correct the hole position deviation problem in circuit board manufacturing, reduce the scrap rate caused by hole position deviation, and ensure the reliability of circuit board preparation. More specifically, the borehole distribution-drilling design joint perception alignment feature map is input into a hole deviation identifier based on a classifier to obtain a recognition result, and the recognition result is used to indicate whether a borehole deviation occurs, including: expanding the borehole distribution-drilling design joint perception alignment feature map into a classification feature vector based on a row vector or a column vector; using multiple fully connected layers of the classifier to fully connect encode the classification feature vector to obtain an encoded classification feature vector; and passing the encoded classification feature vector through the Softmax classification function of the classifier to obtain the recognition result.

[0059] In a preferred example, inputting the drilling distribution-drilling design joint perception alignment feature map into a classifier-based hole deviation identifier to obtain an identification result includes:

[0060] Determine a maximum eigenvalue of the drilling distribution-drilling design joint perceptual alignment and a minimum eigenvalue of the drilling distribution-drilling design joint perceptual alignment of the drilling distribution-drilling design joint perceptual alignment feature map;

[0061] Calculating a borehole distribution-drilling design joint perceptual alignment mean and a borehole distribution-drilling design joint perceptual alignment standard deviation of a feature set of the borehole distribution-drilling design joint perceptual alignment feature graph, and calculating a quotient of the borehole distribution-drilling design joint perceptual alignment mean and the borehole distribution-drilling design joint perceptual alignment standard deviation to obtain a borehole distribution-drilling design joint perceptual alignment statistical normalization value;

[0062] Calculate the inverse of each eigenvalue of the drilling distribution-drilling design joint perceptual alignment feature map, multiply it by the difference between the maximum eigenvalue of the drilling distribution-drilling design joint perceptual alignment and the minimum eigenvalue of the drilling distribution-drilling design joint perceptual alignment, and then subtract it from the drilling distribution-drilling design joint perceptual alignment statistical normalization value to obtain a drilling distribution-drilling design joint perceptual alignment distribution approximation feature map;

[0063] Calculate an exponential function with a natural constant as the base, with each eigenvalue of the drilling distribution-drilling design joint-aware alignment distribution approximation feature map as the exponent to obtain a drilling distribution-drilling design joint-aware alignment distribution class approximation feature map;

[0064] Performing point addition on the borehole distribution-drilling design joint perceptual alignment distribution class approximation feature map and the borehole distribution-drilling design joint perceptual alignment statistical normalization value, and calculating the base-two logarithm of the absolute value of each eigenvalue of the point-added vector to obtain an optimized borehole distribution-drilling design joint perceptual alignment feature map; and

[0065] The optimized drilling distribution-drilling design joint perception alignment feature map is input into a classifier-based hole deviation identifier to obtain a recognition result.

[0066] The drilling distribution-drilling design joint perception alignment feature map is denoted as The optimization expression is:

[0067]

[0068]

[0069]

[0070] in is the drilling distribution-drilling design joint perception alignment feature map, and are respectively the maximum eigenvalue of the drilling distribution-drilling design joint perceptual alignment and the minimum eigenvalue of the drilling distribution-drilling design joint perceptual alignment of the drilling distribution-drilling design joint perceptual alignment feature map, is the difference between the maximum eigenvalue of the joint perceptual alignment of the borehole distribution and the borehole design and the minimum eigenvalue of the joint perceptual alignment of the borehole distribution and the borehole design, is the drill hole distribution-drill hole design joint perceptual alignment mean, is the standard deviation of the borehole distribution-drillhole design joint perceptual alignment, and Normalize the values ​​for the joint perceptual alignment of the borehole distribution-drillhole design statistics, is the drilling distribution-drilling design joint perception alignment distribution class approximate feature map, () is the exponential operation, is the logarithm to base 2, is the dot product, It's a point reduction. It's a little plus. It is the optimized drilling distribution-drilling design joint-aware alignment feature map.

[0071] Here, in the preferred example, since the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map respectively represent the image semantic features of the circuit board drilling distribution image and the circuit board drilling distribution design image, when they are input into the bidirectional global attention joint perception module based on fine-grained deconstruction, the image semantic feature differences caused by the semantic differences of the source images will have different attention weights based on the fine-grained decoupling of the spatial distribution of the image semantic features. Therefore, the drilling distribution-drilling design joint perception alignment feature map will also have a diversified set expression distribution of bidirectional global aggregation features. Therefore, it is expected to improve the balance between the regression mapping accuracy and completeness when the drilling distribution-drilling design joint perception alignment feature map is input into the classifier-based hole bias identifier for class regression, thereby improving the accuracy of the recognition results.

[0072] Based on this, by performing random statistical normalization on the diversified feature set of the drilling distribution-drilling design joint perception alignment feature map, the confidence space constructed based on the overall feature value of the drilling distribution-drilling design joint perception alignment feature map is approximated by the standardized continuous probability density distribution relative to the response hypothesis test of each feature value of the drilling distribution-drilling design joint perception alignment feature map, thereby establishing the target reachability of the diversified feature distribution of the drilling distribution-drilling design joint perception alignment feature map to the unified regression target, so as to achieve the balanced executableness between mapping accuracy and mapping completeness in the class regression process based on the diversified feature distribution of the drilling distribution-drilling design joint perception alignment feature map, and improve the accuracy of the recognition result obtained by the hole deviation identifier based on the classifier of the drilling distribution-drilling design joint perception alignment feature map input. In this way, the hole deviation recognition detection can be performed more accurately in the manufacturing process of thick copper high-density circuit boards, so as to timely discover and correct the hole position deviation problem in the circuit board manufacturing, reduce the scrap rate caused by the hole position deviation, and ensure the reliability of the circuit board preparation.

[0073] In particular, in S7, a current shunt is arranged between the thick copper circuit pattern via and the high-density circuit pattern via to form a high-density thick copper circuit board. The current shunt can evenly distribute the current from the thick copper layer to the high-density layer, thereby reducing local overcurrent and heat accumulation. It is worth mentioning that the current shunt is placed around the via to ensure that the current is evenly distributed to all vias.

[0074] In particular, in S8, a power management chip is installed on the high-density thick copper circuit board, and the power management chip is used to monitor the battery status. The power management chip is used to monitor the battery status to prevent overcharging or over-discharging, so as to take into account the fast charging and battery life of the thick copper high-density circuit board, thereby extending the service life of the circuit board and the battery.

[0075] In summary, the method of the thick copper high-density circuit board based on fast charging and endurance according to the embodiment of the present application is explained, which collects the circuit board drilling distribution image through the camera, and extracts the circuit board drilling distribution design image from the database, and then introduces the image processing and analysis algorithm based on artificial intelligence and machine vision at the back end to analyze the circuit board drilling distribution image and the circuit board drilling distribution design image, so as to learn and characterize the bidirectional fine-grained global joint perception semantics between the detection drilling distribution characteristics and the design drilling distribution characteristics, so as to automatically identify the hole deviation. In this way, the intelligent identification and detection of hole deviation can be realized in the manufacturing process of thick copper high-density circuit boards, so as to timely discover and correct the hole position deviation problem in the circuit board manufacturing, reduce the scrap rate caused by the hole position deviation, and ensure the reliability of the circuit board preparation.

[0076] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for a thick copper high-density circuit board based on fast charging and battery life, characterized in that: include: Taking the copper clad board and making a high density circuit pattern to form a daughter board pattern board; Laminating an insulating dielectric layer to a surface of the daughterboard graphic board to form a laminated daughterboard; Cutting the laminated sub-boards to form high-density sub-core boards; Taking a thick copper clad board and making a thick copper circuit pattern to form a thick copper pattern board; Placing the high-density sub-core board in the embedding area of ​​the thick copper graphic board, and laminating it with the thick copper insulating medium layer and the surface copper foil layer to form a laminate; Drilling holes in the area where the thick copper circuit pattern of the laminate is located, and drilling holes in the area where the high-density circuit pattern of the laminate is located, and then performing hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes; A current shunt is provided between the thick copper circuit pattern via hole and the high-density circuit pattern via hole to form a high-density thick copper circuit board; Installing a power management chip on the high-density thick copper circuit board, wherein the power management chip is used to monitor the battery status; The method comprises drilling holes in the area where the thick copper circuit pattern of the laminate is located, drilling holes in the area where the high-density circuit pattern of the laminate is located, and then performing hole metallization processing to form thick copper circuit pattern via holes and high-density circuit pattern via holes, including: Acquire a circuit board drilling distribution image captured by a camera; Extracting a circuit board drilling distribution design image from a database; Performing feature extraction and joint analysis based on fine-grained deconstruction on the circuit board drilling distribution image and the circuit board drilling distribution design image to obtain a drilling distribution-drilling design joint perception alignment feature map; Based on the drilling distribution-drilling design joint perception alignment feature map, hole deviation identification is performed to determine whether drilling deviation occurs.

2. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 1 is characterized in that: The circuit board drilling distribution image and the circuit board drilling distribution design image are subjected to feature extraction and joint analysis based on fine-grained deconstruction to obtain a drilling distribution-drilling design joint perception alignment feature map, including: Scaling the circuit board drilling distribution image to obtain a scaled circuit board drilling distribution image having the same size as the circuit board drilling distribution design image; Inputting the scaled circuit board drilling distribution image and the circuit board drilling distribution design image into a drilling distribution feature extractor based on a deep neural network model to obtain a circuit board drilling distribution detection feature map and a circuit board drilling distribution design feature map respectively; The circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map are input into a bidirectional global attention joint perception module based on fine-grained deconstruction to obtain the drilling distribution-drilling design joint perception alignment feature map.

3. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 2, characterized in that: The drilling distribution feature extractor based on the deep neural network model is a drilling distribution feature extractor based on the void convolutional neural network model.

4. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 3 is characterized in that: Inputting the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map into a bidirectional global attention joint perception module based on fine-grained deconstruction to obtain the drilling distribution-drilling design joint perception alignment feature map, including: Performing feature fine-grained deconstruction on the circuit board drilling distribution detection feature map and the circuit board drilling distribution design feature map to obtain a set of local feature matrices of the circuit board drilling distribution detection channel dimension and a set of local feature matrices of the circuit board drilling distribution design channel dimension; Using each local feature matrix of the circuit board drilling distribution detection channel dimension in the set of local feature matrices of the circuit board drilling distribution detection channel dimension as a query feature matrix, using the set of local feature matrices of the circuit board drilling distribution design channel dimension as a set of key feature matrices, inputting the query feature matrix and the set of key feature matrices into a one-way global attention interaction module based on the first converter structure to obtain a set of local feature matrices of the circuit board drilling distribution detection channel dimension optimized by one-way global attention; Using each local feature matrix of the circuit board drilling distribution design channel dimension in the set of local feature matrices of the circuit board drilling distribution design channel dimension as a query feature matrix, using the set of local feature matrices of the circuit board drilling distribution detection channel dimension as a set of key feature matrices, inputting the query feature matrix and the set of key feature matrices into a one-way global attention interaction module based on a second converter structure to obtain a set of local feature matrices of the circuit board drilling distribution design channel dimension optimized by one-way global attention; The set of local feature matrices of the one-way global attention optimization circuit board drilling distribution detection channel dimension and the set of local feature matrices of the one-way global attention optimization circuit board drilling distribution design channel dimension are coupled along the channel dimension to obtain a one-way global interaction optimization circuit board drilling distribution detection feature map and a one-way global interaction optimization circuit board drilling distribution design feature map; The position-weighted sum between the one-way global interactive optimization circuit board drilling distribution detection feature map and the one-way global interactive optimization circuit board drilling distribution design feature map is calculated to obtain the drilling distribution-drilling design joint perception alignment feature map.

5. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 4, characterized in that: Each local feature matrix of the circuit board drilling distribution detection channel dimension in the set of local feature matrices of the circuit board drilling distribution detection channel dimension is used as a query feature matrix, and the set of local feature matrices of the circuit board drilling distribution design channel dimension is used as a set of key feature matrices, and the query feature matrix and the set of key feature matrices are input into a one-way global attention interaction module based on a first converter structure to obtain a set of one-way global attention optimized circuit board drilling distribution detection channel dimension local feature matrices, including: Selecting a predetermined circuit board drilling distribution detection channel dimension local feature matrix from the set of circuit board drilling distribution detection channel dimension local feature matrix as a query feature matrix; Calculate the product of the predetermined circuit board drilling distribution detection channel dimension local feature matrix and the transposed matrix of each circuit board drilling distribution design channel dimension local feature matrix in the set of circuit board drilling distribution design channel dimension local feature matrix to obtain a set of circuit board drilling distribution detection-design channel dimension local semantic interaction feature matrices; Each circuit board drilling distribution detection-design channel dimension local semantic interaction feature matrix in the set of the circuit board drilling distribution detection-design channel dimension local semantic interaction feature matrix is ​​divided by the scale square root of the predetermined circuit board drilling distribution detection channel dimension local feature matrix according to the position point, and then used The function performs soft maximum normalization processing on each feature matrix in the obtained set of feature matrices to obtain a set of local weight matrices of the circuit board drilling distribution detection channel dimension; Taking the set of local weight matrices of the circuit board drilling distribution detection channel dimension as weighted weights, the position-weighted sum between the local feature matrices of the circuit board drilling distribution detection channel dimension in the set of local feature matrices of the circuit board drilling distribution detection channel dimension is calculated to obtain the unidirectional global attention optimized local feature matrix of the circuit board drilling distribution detection channel dimension.

6. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 5, characterized in that: Each local feature matrix of the circuit board drilling distribution design channel dimension in the set of local feature matrices of the circuit board drilling distribution design channel dimension is used as a query feature matrix, and the set of local feature matrices of the circuit board drilling distribution detection channel dimension is used as a set of key feature matrices, and the query feature matrix and the set of key feature matrices are input into a one-way global attention interaction module based on a second converter structure to obtain a set of one-way global attention optimized circuit board drilling distribution design channel dimension local feature matrices, including: Selecting a predetermined circuit board drilling distribution design channel dimension local feature matrix from the set of circuit board drilling distribution design channel dimension local feature matrices as a query feature matrix; Calculate the product of the predetermined circuit board drilling distribution design channel dimension local feature matrix and the transposed matrix of each circuit board drilling distribution detection channel dimension local feature matrix in the set of circuit board drilling distribution detection channel dimension local feature matrix to obtain a set of circuit board drilling distribution design-detection channel dimension local semantic interaction feature matrices; The circuit board drilling distribution design-detection channel dimension local semantic interaction feature matrix in the set of the circuit board drilling distribution design-detection channel dimension local semantic interaction feature matrix is ​​divided by the scale square root of the predetermined circuit board drilling distribution design channel dimension local feature matrix according to the position point, and then used The function performs a soft maximum normalization process on each feature matrix in the obtained set of feature matrices to obtain a set of local weight matrices of the channel dimension of the circuit board drilling distribution design; Taking the set of local weight matrices of the circuit board drilling distribution design channel dimension as weighted weights, the position-weighted sum between the local feature matrices of the circuit board drilling distribution design channel dimension in the set of local feature matrices of the circuit board drilling distribution design channel dimension is calculated to obtain the unidirectional global attention optimized local feature matrix of the circuit board drilling distribution design channel dimension.

7. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 6, characterized in that: Based on the drilling distribution-drilling design joint perception alignment feature map, hole deviation identification is performed to determine whether drilling deviation occurs, including: inputting the drilling distribution-drilling design joint perception alignment feature map into a classifier-based hole deviation identifier to obtain an identification result, and the identification result is used to indicate whether drilling deviation occurs.

8. The method for thick copper high-density circuit board based on fast charging and battery life according to claim 7, characterized in that: The drilling distribution-drilling design joint perception alignment feature map is input into a classifier-based hole deviation identifier to obtain a recognition result, and the recognition result is used to indicate whether drilling deviation occurs, including: Expanding the drilling distribution-drilling design joint perception alignment feature map into a classification feature vector based on a row vector or a column vector; Performing full connection encoding on the classification feature vector using multiple fully connected layers of the classifier to obtain an encoded classification feature vector; and The encoded classification feature vector is passed through the Softmax classification function of the classifier to obtain the recognition result.

9. A thick copper high-density circuit board based on fast charging and battery life, characterized in that: The thick copper high-density circuit board based on fast charging and long battery life is manufactured by the method for a thick copper high-density circuit board based on fast charging and long battery life as described in any one of claims 1-8.

Images