PCB circuit board processing method, device and equipment based on technological parameter optimization

Abstract

The invention provides a PCB processing method, device and equipment based on technological parameter optimization, and the method comprises the steps: carrying out the feature extraction of PCB design data, and obtaining feature data; matching an adaptive initial process parameter combination according to the feature data; performing quality detection on the PCB which is processed for the first time based on the initial process parameter combination, and analyzing detection data by using a defect identification algorithm to obtain a quality evaluation result; based on a quality evaluation result, optimizing the initial process parameter combination by using an error back propagation algorithm to obtain an optimized process parameter combination; and controlling secondary processing of the PCB based on the optimized process parameter combination. According to the method, the initial process parameter combination is optimized according to the quality evaluation result of the circuit board, so that the defect that an existing processing method cannot perform self-adaptive optimization according to the actual condition of the circuit board is overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of circuit board processing, and in particular to a PCB circuit board processing method, device and equipment based on process parameter optimization. Background Art

[0002] PCB (Printed Circuit Board) circuit boards are key supporting components of electronic devices. From consumer electronic devices (such as smart phones and tablets) to industrial control equipment, communication equipment, and aerospace equipment, PCB circuit boards play a core role in carrying electronic components, realizing circuit connections and signal transmission. As electronic devices continue to develop in the direction of miniaturization, high performance, high integration, and high reliability, the requirements for PCB circuit boards are also increasing.

[0003] Traditional PCB circuit board processing usually follows a relatively fixed design and processing process. First, the circuit board is designed according to the circuit function requirements of the product, and then it is processed and manufactured according to the established process parameters. However, these process parameters are often selected based on experience or standard process manuals, and lack precise adaptability to specific design features. That is, the current processing method cannot be adaptively optimized according to the actual situation of the circuit board. Summary of the invention

[0004] The main purpose of the present invention is to provide a PCB circuit board processing method, device and equipment based on process parameter optimization, aiming to overcome the defect that the current processing method cannot be adaptively optimized according to the actual situation of the circuit board.

[0005] To achieve the above object, the present invention provides a PCB circuit board processing method based on process parameter optimization, comprising the following steps:

[0006] Extract features from PCB design data to obtain feature data; match an appropriate initial process parameter combination based on the feature data;

[0007] Performing quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyzing the inspection data using a defect recognition algorithm, and obtaining a quality assessment result;

[0008] Based on the quality assessment result, the initial process parameter combination is optimized using an error back propagation algorithm to obtain an optimized process parameter combination;

[0009] Based on the optimized process parameter combination, the secondary processing of the PCB circuit board is controlled.

[0010] Furthermore, the characteristic data includes circuit layout and aperture specifications.

[0011] Furthermore, the defect recognition algorithm is used to analyze the inspection data and obtain quality assessment results, including:

[0012] Separating the detection data of the PCB circuit board according to the circuit layer, the dielectric layer and the metal layer to obtain layered detection data;

[0013] Extracting feature points from each layer of data in the layered detection data to obtain data feature points;

[0014] Analyze the data feature points corresponding to each layer of data based on the pre-trained defect recognition model to obtain the corresponding defect type and severity;

[0015] The defect types and severity corresponding to each layer of data are summarized and weighted to obtain the quality assessment result.

[0016] Furthermore, the defect recognition algorithm is used to analyze the inspection data and obtain quality assessment results, including:

[0017] Based on multimodal data fusion technology, the detection data from different sensors are weighted and fused to form a high-dimensional comprehensive detection data set;

[0018] Using a sparse representation algorithm to represent the comprehensive detection data set as sparse data;

[0019] The sparse data is subjected to local binary pattern calculation to extract the texture information of the circuit board surface, and the various defect features of the circuit board surface are extracted by combining texture feature analysis.

[0020] Based on the defect features, the nodes and edges of the Bayesian network are constructed; each node represents a defect feature, and the edge represents the probability relationship between the defect features; the Bayesian theorem is used to calculate the probability of quality problems in the PCB circuit board under different defect feature combinations, and the quality status is divided into multiple quality levels according to the probability size to obtain the quality assessment result.

[0021] Furthermore, based on the quality evaluation result, the initial process parameter combination is optimized using an error back propagation algorithm to obtain an optimized process parameter combination, including:

[0022] Based on the pre-established correspondence between quality deviation and process parameter correction value, each deviation in the current quality assessment result is matched to the corresponding process parameter correction range;

[0023] Based on the process parameter correction range, an error back propagation algorithm is used to iteratively optimize the initial process parameter combination; wherein, in the error back propagation algorithm, the optimization step size is dynamically adjusted according to the real-time feedback of the deviation.

[0024] Further, after controlling the secondary processing of the PCB circuit board, it includes:

[0025] Conduct quality inspection on PCB circuit boards after secondary processing and analyze the secondary quality assessment results;

[0026] If the secondary quality assessment result meets the preset requirements, the characteristic data and the optimized process parameter combination are mapped to obtain a mapping relationship;

[0027] Generate a management code based on the characteristic data and the optimized process parameter combination;

[0028] The mapping relationship is stored, and the management authority of the mapping relationship is set based on the management code.

[0029] Furthermore, based on the characteristic data and the optimized process parameter combination, a management code is generated, including:

[0030] Performing data cleaning on the characteristic data and the optimized process parameter combination to obtain a data string;

[0031] Adding characters in the data string to the matrix to generate a character matrix;

[0032] Based on the characteristics of the characters at each matrix position in the character matrix, a preset coding table is adjusted for a second time to obtain an adjusted coding table;

[0033] Based on the adjustment coding table, encoding the data string to obtain a coded data string;

[0034] The management code is generated based on the multi-dimensional features of the encoded data string.

[0035] The present invention also provides a PCB circuit board processing device based on process parameter optimization, comprising:

[0036] A matching unit is used to extract features from PCB design data to obtain feature data; and to match an adaptive initial process parameter combination according to the feature data;

[0037] An evaluation unit, used to perform quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyze the inspection data using a defect recognition algorithm, and obtain a quality evaluation result;

[0038] An optimization unit, configured to optimize the initial process parameter combination based on the quality evaluation result by using an error back propagation algorithm to obtain an optimized process parameter combination;

[0039] A control unit is used to control the secondary processing of the PCB circuit board based on the optimized process parameter combination.

[0040] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0041] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0042] The PCB circuit board processing method, device and equipment based on process parameter optimization provided by the present invention include: extracting features from PCB circuit board design data to obtain feature data; matching an adaptive initial process parameter combination according to the feature data; performing quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyzing the inspection data using a defect recognition algorithm, and obtaining a quality assessment result; optimizing the initial process parameter combination based on the quality assessment result using an error back propagation algorithm to obtain an optimized process parameter combination; and controlling the secondary processing of the PCB circuit board based on the optimized process parameter combination. In the present invention, by optimizing the initial process parameter combination according to the quality assessment result of the circuit board, the defect that the current processing method cannot perform adaptive optimization according to the actual situation of the circuit board is overcome. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a schematic diagram of the steps of a PCB circuit board processing method based on process parameter optimization in one embodiment of the present invention;

[0044] Figure 2 It is a structural block diagram of a PCB circuit board processing device based on process parameter optimization in one embodiment of the present invention;

[0045] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0046] The implementation, functional features and advantages of the present invention will be further described in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0048] Reference Figure 1 In one embodiment of the present invention, a PCB circuit board processing method based on process parameter optimization is provided, comprising the following steps:

[0049] Step S1, extracting features from PCB design data to obtain feature data; matching an appropriate initial process parameter combination according to the feature data;

[0050] Step S2, performing quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyzing the inspection data using a defect recognition algorithm, and obtaining a quality assessment result;

[0051] Step S3, based on the quality evaluation result, using an error back propagation algorithm to optimize the initial process parameter combination to obtain an optimized process parameter combination;

[0052] Step S4, controlling the secondary processing of the PCB circuit board based on the optimized process parameter combination.

[0053] In this embodiment, as described in step S1 above, the PCB circuit board design data includes information such as circuit layout, via position, number of layers, etc. Feature extraction is to use specific algorithms and techniques to extract key points from these data, such as circuit width, length, spacing, etc. The above key information is an important basis for subsequent processing. Different circuit features correspond to different processing techniques and parameters. Accurate extraction can provide strong data support for subsequent steps.

[0054] After acquiring the characteristic data, the initial process parameter combination is matched based on these data. PCB circuit board processing includes etching, drilling, electroplating and other links, each of which has many parameters, such as etching time, temperature, etching liquid concentration, drilling speed, feed rate, etc. Through the constructed parameter matching model, the characteristic data is compared and calculated with the preset process parameter library to find the initial process parameter combination that best fits the current circuit board characteristics, laying the foundation for the first processing.

[0055] As described in step S2 above, after the first processing is completed based on the initial process parameter combination, it is very important to perform quality inspection on the PCB circuit board. Use optical inspection equipment, electronic testing equipment, etc. to comprehensively check whether the circuit board has short circuits, open circuits, line defects, poor vias, etc.

[0056] Then, the defect recognition algorithm is used to deeply analyze the test data. The algorithm can quickly and accurately identify the possible defect types and locations from a large amount of test data. It analyzes the test data features and compares them with the preset defect pattern to determine whether the circuit board has defects and the severity of the defects. For example, by analyzing the circuit image features, it can determine whether the circuit width meets the requirements and whether the circuit edges are neat.

[0057] The quality assessment result is obtained based on the analysis results of the defect recognition algorithm. This result is a comprehensive evaluation of the quality of the circuit board after the first processing. It can not only determine whether the circuit board is qualified, but also provide detailed feedback for subsequent process parameter optimization and clarify the direction of improvement and optimization.

[0058] As described in step S3 above, the quality assessment result reflects the quality status of the circuit board under the initial process parameter combination. Based on this, the initial process parameter combination is optimized using the error back propagation algorithm. The error back propagation algorithm can reversely deduce and adjust the process parameters based on the error information in the quality assessment result, so that the subsequent processing is as close to the ideal state as possible. Through the continuous iterative calculation of the error back propagation algorithm, the values ​​of each process parameter are gradually adjusted, and finally the optimized process parameter combination is obtained. This combination is the result adjusted by the optimization algorithm based on the actual quality feedback, and it is more likely to produce a PCB circuit board that meets high quality standards.

[0059] As described in step S4 above, after obtaining the optimized process parameter combination, the secondary processing of the PCB circuit board is controlled according to the above parameters. This step is to put the optimization plan into practice and process the circuit board again with new process parameters, hoping to improve the quality of the circuit board, reduce or eliminate the defects in the first processing, and produce products that better meet the design requirements and quality standards.

[0060] In one embodiment, the characteristic data includes circuit layout and aperture specifications.

[0061] In one embodiment, the defect recognition algorithm is used to analyze the inspection data to obtain a quality assessment result, including:

[0062] Separating the detection data of the PCB circuit board according to the circuit layer, the dielectric layer and the metal layer to obtain layered detection data;

[0063] Extracting feature points from each layer of data in the layered detection data to obtain data feature points;

[0064] Analyze the data feature points corresponding to each layer of data based on the pre-trained defect recognition model to obtain the corresponding defect type and severity;

[0065] The defect types and severity corresponding to each layer of data are summarized and weighted to obtain the quality assessment result.

[0066] In this embodiment, the PCB circuit board is a complex multi-layer structure, and different layers have differences in function and manufacturing process, and their potential defect types and impacts are also different. Separating the detection data according to the circuit layer, dielectric layer and metal layer can make the subsequent analysis more targeted, avoid interference between data of different layers, and more accurately locate and analyze possible problems in each layer. Compared with the traditional overall detection data analysis method, this layered processing method takes into account the structural characteristics of the PCB circuit board, is an innovation in the detection data processing method, and helps to improve the accuracy and efficiency of defect identification.

[0067] The purpose of extracting feature points from the layered inspection data of each layer is to convert a large amount of raw inspection data into representative key information. These feature points can reflect the important characteristics of the circuit board of this layer, such as the width, spacing, size and position of the lines, etc. By analyzing these feature points, it is possible to quickly determine whether there are defects in this layer and the approximate situation of the defects. Accurate and targeted feature point extraction method is one of the key innovations of this technical solution. By reasonably selecting and extracting feature points, the amount of data processing can be effectively reduced, while improving the accuracy of defect identification. Compared with traditional random or broad data extraction methods, it is more scientific and efficient.

[0068] By using the pre-trained defect recognition model to analyze the feature points corresponding to each layer of data, the model's powerful learning and recognition capabilities can be used to quickly and accurately determine the defect type and severity corresponding to the layer of data. This model-based analysis method has higher accuracy and consistency than manual experience judgment.

[0069] The introduction of a pre-trained defect recognition model is a technological innovation. The model can learn the characteristic patterns of various defects through a large amount of sample data, so that it can quickly make accurate judgments when faced with actual inspection data. Moreover, with the development of technology and the accumulation of data, the model can be continuously optimized and updated to improve recognition capabilities.

[0070] By summarizing the defect type and severity corresponding to each layer of data, we can fully understand the defect situation of the entire PCB circuit board. The weighted calculation takes into account the different degrees of influence of defects in different layers on the overall quality of the circuit board. By assigning different weights to defects in different layers, we can more scientifically and comprehensively evaluate the quality of the circuit board and obtain an evaluation result that can accurately reflect the overall quality status of the circuit board.

[0071] The weighted calculation method fully considers the differences in the importance of defects in different layers. It is an innovation in the quality assessment method, making the quality assessment results more objective, accurate and more in line with actual application needs.

[0072] In summary, the various steps of defect identification and quality assessment in this technical solution are closely linked and progressive. Through innovative data processing, analysis and evaluation methods, the quality of PCB circuit boards can be accurately and effectively evaluated, laying a solid foundation for subsequent process improvements and product quality improvements.

[0073] In one embodiment, the defect recognition algorithm is used to analyze the inspection data to obtain a quality assessment result, including:

[0074] Based on multimodal data fusion technology, the detection data from different sensors are weighted and fused to form a high-dimensional comprehensive detection data set;

[0075] Using a sparse representation algorithm to represent the comprehensive detection data set as sparse data;

[0076] The sparse data is subjected to local binary pattern calculation to extract the texture information of the circuit board surface, and the various defect features of the circuit board surface are extracted by combining texture feature analysis.

[0077] Based on the defect features, the nodes and edges of the Bayesian network are constructed; each node represents a defect feature, and the edge represents the probability relationship between the defect features; the Bayesian theorem is used to calculate the probability of quality problems in the PCB circuit board under different defect feature combinations, and the quality status is divided into multiple quality levels according to the probability size to obtain the quality assessment result.

[0078] In this embodiment, in the quality inspection process of the PCB circuit board, a variety of different types of sensors (such as optical sensors, electronic test sensors, ultrasonic sensors, etc.) are used to collect detection data. The data provided by the above sensors may have different characteristics and advantages. For example, optical sensors can capture image information on the surface of the circuit board, electronic test sensors can detect the electrical performance of the circuit board, and ultrasonic sensors can detect internal structures. The detection data from different sensors are weighted and fused in order to make full use of the data advantages of various sensors to form a more comprehensive and more informative high-dimensional comprehensive detection data set. Through weighted fusion, the data of each sensor can be assigned corresponding weights according to the importance and reliability of different sensor data, so that the final comprehensive data set can more accurately reflect the true state of the circuit board. This helps to overcome the limitations of single sensor data and improve the accuracy and integrity of detection data.

[0079] Traditional detection data processing usually relies on only a single type of sensor data or simply stitches together different sensor data, lacking data fusion and optimization. The multimodal data fusion technology adopted by this technical solution fully considers the importance of different sensor data through weighted fusion, which is an innovative way of data processing. This method can integrate multiple data sources together, so that subsequent analysis can be based on more comprehensive data, providing a more reliable basis for subsequent quality assessment.

[0080] Comprehensive inspection data sets are often high-dimensional and highly redundant, and contain a large amount of data information, some of which may not be critical or redundant for defect identification and quality assessment. Using sparse representation algorithms to represent them as sparse data can remove a large amount of redundant information while retaining important information, reducing data complexity and computational complexity. Sparse data representation is easier to store, transmit, and process, and also helps to highlight key information in the data, providing a more concise and effective data representation for subsequent feature extraction and analysis, and improving the computational efficiency and accuracy of subsequent steps.

[0081] Compared with directly processing high-dimensional comprehensive detection data sets, the application of sparse representation algorithms reflects the innovation of data processing. It can present complex data in a more concise form, reduce the resource consumption and time cost of data processing, and also improve the pertinence of subsequent analysis, making key information easier to mine and process. This is an innovation in data preprocessing, which improves the performance and efficiency of the entire technical solution.

[0082] By calculating the local binary pattern of sparse data, the texture information of the circuit board surface can be effectively extracted. The texture characteristics of the circuit board surface contain rich quality information, such as surface smoothness, uniformity, scratches, defects, etc. This information is very important for judging the quality of the circuit board.

[0083] By combining texture feature analysis, we can further extract various defect features on the surface of the circuit board. These defect features may include physical defects on the surface (such as scratches, pits, etc.) and some microscopic structural defects, providing more detailed information for quality assessment and helping to more accurately discover and locate potential quality problems. Using local binary pattern calculation to extract texture information and combining texture feature analysis to mine defect features is an innovative method for circuit board surface defect detection. It makes full use of the inherent connection between the surface texture features of the circuit board and quality problems, providing a new perspective and means for more accurate quality assessment. Compared with traditional methods based only on macroscopic inspection or simple image analysis, it can more deeply mine the microscopic defect information on the surface.

[0084] Based on the extracted defect features, a Bayesian network is constructed, with each defect feature as a node in the network. The edges represent the probability relationship between the defect features, which can clearly show the relationship between different defect features. The probability of quality problems in PCB circuit boards under different defect feature combinations is calculated by Bayesian theorem. The quality status can be divided into multiple quality levels according to the probability, and the quantitative evaluation of the quality of PCB circuit boards can be achieved. This method connects various defect features through probability relationships, takes into account the mutual influence and joint occurrence of different defect features, and can evaluate the quality status of circuit boards more scientifically and systematically, avoiding the problem of inaccurate evaluation caused by isolating each defect feature.

[0085] Using Bayesian networks to evaluate the quality of PCB circuit boards is an innovative evaluation method. Traditional quality evaluation is usually based on experience or simple rule judgment, while Bayesian networks use probabilistic reasoning to comprehensively consider multiple defect characteristics and their interrelationships. It can handle complex defect combinations more flexibly and give more reasonable quality grade divisions based on probabilistic reasoning, making the quality evaluation results more scientific and reliable.

[0086] In one embodiment, based on the quality evaluation result, the initial process parameter combination is optimized using an error back propagation algorithm to obtain an optimized process parameter combination, including:

[0087] Based on the pre-established correspondence between quality deviation and process parameter correction value, each deviation in the current quality assessment result is matched to the corresponding process parameter correction range;

[0088] Based on the process parameter correction range, an error back propagation algorithm is used to iteratively optimize the initial process parameter combination; wherein, in the error back propagation algorithm, the optimization step size is dynamically adjusted according to the real-time feedback of the deviation.

[0089] In this embodiment, in the PCB circuit board processing, different quality assessment results will reflect various deviations, which involve multiple aspects such as circuit size, electrical performance, physical structure, etc. The pre-constructed correspondence between quality deviation and process parameter correction value is a knowledge base established based on a large amount of experimental data, experience or theoretical analysis. Matching each deviation in the current quality assessment result to the corresponding process parameter correction range is to determine which process parameters need to be adjusted and the approximate adjustment direction.

[0090] The above matching operation can help quickly locate the process parameters related to quality issues, making the subsequent optimization process more targeted. For example, if the quality assessment results show that there is a deviation in the line width, through this correspondence, the correction range of process parameters that may affect the line width, such as etching time, etching solution concentration, etc., can be found to avoid blindly adjusting all process parameters, thereby improving optimization efficiency and accuracy.

[0091] Traditional process parameter adjustments are often made through trial-and-error local adjustments based on experience. However, this technical solution establishes a correspondence between the system's quality deviation and the process parameter correction value, links the quality assessment results with the process parameter adjustment, and achieves a more scientific and systematic preliminary preparation for parameter adjustment. This method is highly targeted, reduces blindness in the optimization process, and is an innovative shift from experience-based adjustment to data-driven adjustment.

[0092] The error back propagation algorithm is a powerful optimization algorithm. In this technical solution, it is used to optimize the initial process parameter combination according to the process parameter correction range determined in the previous step. Through continuous iteration, the process parameters are gradually adjusted to improve the process parameters in the direction of reducing or eliminating quality deviations, so as to achieve the purpose of improving the quality of PCB circuit boards. During the optimization process, it takes into account the complex relationship and mutual influence between process parameters, rather than adjusting each parameter in isolation, so as to achieve an overall optimization effect. For example, adjusting the etching time may affect the etching depth, which in turn affects the resistance and capacitance of the circuit. The error back propagation algorithm can comprehensively consider these factors and find the best parameter combination.

[0093] One of the core innovations of this technical solution is to use the error back propagation algorithm to optimize process parameters. In the field of PCB circuit board processing, traditional parameter optimization is usually based on simple linear adjustment or local adjustment, while the error back propagation algorithm can find the optimal solution in a complex parameter space. It can handle the nonlinear relationship between multiple parameters and can automatically adjust parameters according to the feedback information of quality evaluation results, which improves the accuracy and reliability of optimization.

[0094] In the iterative process of the error back propagation algorithm, the selection of the optimization step size is crucial. If the step size is too large, the optimization process may be unstable and the optimal solution may be missed; if the step size is too small, the optimization process will be slow and inefficient. By dynamically adjusting the optimization step size based on the real-time feedback of the deviation, the optimization speed and accuracy can be flexibly adjusted according to the current quality deviation.

[0095] When the quality deviation is large, increasing the optimization step size can speed up the optimization process and make the process parameters approach the optimal value more quickly; when the quality deviation is small, reducing the optimization step size can achieve more precise adjustments, avoid excessive adjustments and repeated oscillations, and improve the accuracy and stability of the final optimization results.

[0096] Traditional error back propagation algorithms usually use a fixed optimization step size, while the method of dynamically adjusting the optimization step size in this technical solution reflects an innovative improvement to the algorithm. This dynamic adjustment method flexibly changes the optimization rhythm according to actual conditions, making the algorithm more adaptable to different quality deviation situations, improving the adaptability and optimization performance of the algorithm, and is an optimization and refinement of the error back propagation algorithm, enhancing the applicability of the algorithm in PCB circuit board processing parameter optimization.

[0097] In one embodiment, after controlling the secondary processing of the PCB circuit board, the method includes:

[0098] Conduct quality inspection on PCB circuit boards after secondary processing and analyze the secondary quality assessment results;

[0099] If the secondary quality assessment result meets the preset requirements, the characteristic data and the optimized process parameter combination are mapped to obtain a mapping relationship;

[0100] Generate a management code based on the characteristic data and the optimized process parameter combination;

[0101] The mapping relationship is stored, and the management authority of the mapping relationship is set based on the management code.

[0102] In this embodiment, after the PCB circuit board is subjected to secondary processing, quality inspection is a key link to ensure the quality of the final product. Through various detection means, such as appearance inspection, electrical performance test, functional test, etc., a comprehensive assessment is made as to whether the PCB circuit board after secondary processing meets the expected quality standards. By analyzing the results of the secondary quality assessment, it can be determined whether the secondary processing has successfully eliminated the problems existing in the first processing, and whether new quality problems have been introduced. It is intended to provide a basis for subsequent operations to determine whether the quality requirements of the product are met, and then determine whether the product is qualified or needs further improvement. For example, the electrical performance of the PCB circuit board is evaluated by detecting the connectivity, insulation performance, signal integrity, etc. of the circuit; its physical structure is evaluated by checking the width, spacing, size and position of the vias, etc.

[0103] When the secondary quality assessment results meet the preset requirements, it indicates that the product has reached the expected quality standard. The feature data (such as the design information and layer structure information of the circuit board) and the optimized process parameter combination (such as etching time, drilling parameters, etc.) are mapped to establish an association relationship. This mapping relationship can reflect what kind of design features can produce products that meet the quality requirements under what kind of optimized process parameters, providing valuable data support for subsequent production and quality control.

[0104] The management code is generated to facilitate the effective management and traceability of PCB products. The characteristic data and optimized process parameters are combined into a unique management code through a specific algorithm or coding rule. The management code can be used as a product identifier for the full life cycle management of the product. It can facilitate rapid identification, information query and status tracking of products during production, sales, use and maintenance, improve the efficiency of production management and product traceability. For example, when a product has quality problems, the management code can be used to quickly locate the product's design information and processing technology to find the root cause of the problem.

[0105] Using characteristic data and optimized process parameters to generate management codes is an innovative product management method. Different from the traditional management methods that only use simple numbers or barcodes, this management code contains richer product information and integrates key product information through a unique coding method, improving the accuracy of management and the richness of information. This management code can prevent the loss and confusion of product information, and provides a more advanced tool for product quality management and traceability. It is an innovation and improvement of traditional product management methods.

[0106] The purpose of storing mapping relationships is to save key information in the production process and accumulate production experience and data assets for the enterprise. At the same time, setting management permissions for mapping relationships based on management codes can ensure the security and controllability of these key data. Different users or departments access and use this information according to their permissions to prevent information leakage and misoperation. It can protect the core production data of the enterprise, and at the same time, according to different management needs, flexibly allocate data usage permissions to ensure that information is reasonably used in different production, quality control and management links.

[0107] In one embodiment, generating a management code based on the characteristic data and the optimized process parameter combination includes:

[0108] Performing data cleaning on the characteristic data and the optimized process parameter combination to obtain a data string;

[0109] Adding characters in the data string to the matrix to generate a character matrix;

[0110] Based on the characteristics of the characters at each matrix position in the character matrix, a preset coding table is adjusted for a second time to obtain an adjusted coding table;

[0111] Based on the adjustment coding table, encoding the data string to obtain a coded data string;

[0112] The management code is generated based on the multi-dimensional features of the encoded data string.

[0113] In this embodiment, during the PCB circuit board processing, the feature data and the optimized process parameter combination may contain various types of data, which may contain some noise, redundant information or non-standard data formats. The purpose of data cleaning is to remove these unnecessary or interfering information, extract useful data and organize them into a standardized data string. For example, remove the abnormal values ​​that may exist in the feature data, and clean up unnecessary spaces and special symbols in the process parameter combination to make the data purer and more unified. The accuracy and efficiency of subsequent processing steps can be improved to ensure that only valid data participates in the subsequent encoding and management code generation process. Through data cleaning, it can be ensured that the data on which the final management code is based is high quality and reliable.

[0114] Traditional data processing directly uses raw data for encoding, while this technical solution improves data quality through the pre-step of data cleaning. This reflects the emphasis and innovation on data preprocessing, lays the foundation for the subsequent generation of more accurate and valuable management codes, and avoids encoding errors or low management code quality caused by irregularities or noise in the raw data.

[0115] The purpose of adding characters in the data string to the matrix is ​​to convert the one-dimensional data form into a two-dimensional matrix form, providing more operating space and structural information for subsequent encoding operations. The use of matrices can better organize and represent data, facilitating complex operations and analysis.

[0116] In the form of a matrix, the position information of the rows and columns of the matrix can be used to explore the relationship and characteristics between characters, which facilitates further encoding and adjustment. For example, the position information of the characters can be combined with the structural characteristics of the matrix to provide a basis for adjusting the encoding table according to the character characteristics.

[0117] The conversion from data string to character matrix is ​​an innovative way of data processing. This structural conversion brings a new perspective to data processing. Displaying data in the form of a matrix can take advantage of matrix operations and matrix analysis. It is an extension of the traditional one-dimensional data processing method and provides a unique idea for subsequent use of matrix characteristics for coding adjustment and generation of management codes.

[0118] The preset coding table is a universal or initial coding scheme. The coding table is adjusted twice according to the characteristics of the characters in the character matrix, such as the frequency, position, and relationship between adjacent characters of the characters (the above adjustment method can be customized in advance). The above adjustment can improve the pertinence and accuracy of the coding, so that the final coded data string can have stronger uniqueness and security. For example, according to the position and frequency of occurrence of the characters in the matrix, the corresponding coding rules in the coding table are modified to make the coding more personalized, targeted, and unique. Specifically, the maximum and minimum values ​​of the characters in the character matrix can be identified; based on the maximum and minimum values, the preset coding table is replaced or translated. This makes the coding table unique and improves the security and uniqueness of subsequent coding.

[0119] The secondary adjustment of the coding table based on data characteristics is an important innovation of this technical solution. The traditional coding process usually uses a fixed coding table, while this solution optimizes the coding process according to specific data characteristics through dynamic adjustment of the coding table, thereby improving the security and uniqueness of the coding.

[0120] The purpose of encoding the data string using the adjusted coding table is to convert the data string into a coded data string with higher information density and confidentiality. By encoding, the character information can be converted into a new representation form, increasing the complexity and security of the information.

[0121] The encoding process can encrypt data or hide information to prevent sensitive information leakage, and can also make the management code more unique and representative, facilitating subsequent storage, management and use. For example, through special encoding rules, the characters in the data string are converted into a set of specific coding sequences, making the management code more recognizable and operable.

[0122] Coding based on the adjusted coding table is an innovative coding method, which utilizes the optimization of the coding table in the previous steps to make the coding result more unique and valuable. Compared with the traditional fixed coding method, this coding method fully considers the data characteristics, improves the security, flexibility and applicability of coding, and is an innovative application of coding technology.

[0123] The coded data string has multi-dimensional features, such as length, character frequency, character combination pattern, etc. The management code is generated based on these multi-dimensional features, aiming to further refine and synthesize the coded data string to produce a final, concise and unique management code. The above management code will serve as the identifier of the PCB circuit board, comprehensively reflecting its characteristic data and optimized process parameter combination information, facilitating the effective management, tracking and tracing of the PCB circuit board, and also providing a basis for subsequent information storage and authority management.

[0124] Using the multi-dimensional characteristics of the coded data string to generate the management code is an innovative management code generation method. It does not simply use the coded data string as the management code, but comprehensively refines it, taking into account information from multiple dimensions, making the management code more representative and unique, and providing a more unique means for product management and traceability, which is different from the traditional simple coding or numbering method.

[0125] In one embodiment, generating a management code based on the characteristic data and the optimized process parameter combination includes:

[0126] For feature data, a three-dimensional matrix is ​​created according to different levels and regions of the PCB circuit board. Different dimensions correspond to different levels, regions and feature types. In the three-dimensional matrix, each element stores specific feature information at that position. For example, in the part representing the line feature, the element can store information such as the width, length, and resistance of the line; for the via feature part, the element stores information such as the diameter and depth of the via.

[0127] For the optimized process parameter combination, a two-dimensional matrix is ​​constructed, wherein the rows correspond to different process steps, such as etching, drilling, electroplating, etc., and the columns correspond to various parameters of the corresponding process steps, such as time, temperature, pressure, etc. The matrix elements store specific parameter values.

[0128] The three-dimensional matrix is ​​transformed to obtain a transformation matrix; the transformation may be a rearrangement of the matrix elements, grouping elements with similar characteristics together, or grouping data at different levels and regions according to a certain rule to form a new matrix structure. From this new matrix, the concentration trend and dispersion degree of the characteristic data in different aspects can be observed.

[0129] The two-dimensional matrix is ​​converted into the frequency domain to obtain the frequency amplitude diagram. By converting the distribution of process parameters in different time or space into frequency domain representation, the periodicity and correlation between different process parameters can be found. According to the frequency domain information, the corresponding frequency amplitude diagram can be drawn, which can intuitively show the frequency characteristics of the process parameters.

[0130] Perform a projection operation on the transformation matrix and project it onto a two-dimensional plane to generate a projection image.

[0131] Extract key information from the transformation matrix, frequency amplitude map and projection map. This key information may include the clustering area of ​​elements, the range of elements, the peak and valley values ​​of the frequency amplitude map, the shape and distribution characteristics of the projection map, etc. This information is integrated with the important elements in the feature data matrix and the process parameter matrix to form a comprehensive information set.

[0132] This comprehensive information set is subjected to dimensionality reduction processing to convert high-dimensional information into low-dimensional information. This dimensionality reduction operation is an intelligent data compression method that removes redundant information based on the importance and relevance of the information and retains only the part that best represents the original information. The reduced-dimensional information is converted into a coding sequence according to the preset coding rules. Different information ranges and features will correspond to different coding methods to ensure the uniqueness and integrity of the coding.

[0133] The generated coding sequence is encrypted using an encryption algorithm based on a chaotic system. This encryption algorithm uses the randomness of the chaotic system and its sensitivity to initial conditions, takes the coding sequence as input, and generates an encrypted coding sequence as a management code through iterative operations of the chaotic system. During the encryption process, each iteration will cause complex changes in the coding sequence according to the characteristics of the chaotic system, ensuring the security of the management code.

[0134] In one embodiment, generating a management code based on the characteristic data and the optimized process parameter combination includes:

[0135] The characteristic data of the PCB circuit board is constructed into a graph structure; the nodes of the graph represent different elements of the PCB circuit board, such as circuits, vias, pads, etc. The edges between the nodes represent the connection relationship between these elements. The nodes have multiple attributes. For example, the node representing the circuit can contain the length, width, number of layers of the circuit, etc.; the node representing the via can contain the diameter, depth, position, etc. of the via. Through this graph structure, the physical structure of the PCB circuit board and the connection between the elements can be clearly displayed.

[0136] Combine the optimized process parameters to construct another graph structure. The nodes of this graph represent different process steps, such as etching, drilling, electroplating, etc. The attributes of the node contain various parameters involved in the process step. For example, the attributes of the etching node can include the concentration of the etching solution, etching time, etching temperature, etc. The edge represents the relationship between different process steps. For example, the edge from the etching node to the cleaning node indicates that the cleaning operation is required after the etching is completed. The edge can have attributes indicating the order or importance to indicate the relationship between different process steps.

[0137] Perform information mining on the graph structure representing feature data to obtain information about key nodes and edges. By traversing the nodes and edges of the graph, find nodes and edges with important features. For example, identify via nodes that connect many lines, which may be key nodes; or find edges that connect key lines, which have an important impact on the performance of the circuit. By observing the properties of nodes and edges, mark those nodes and edges with special properties, such as high-resistance lines or large-diameter vias.

[0138] For the graph structure representing process parameters, analyze the association of each process step node to obtain key process step information. Observe which process steps are connected to more other process steps, indicating that the process step is in a core position. At the same time, based on the attributes of the edges, judge the closeness between different process steps, such as which process steps cannot be changed at will, and which process steps have a greater impact on subsequent steps. Through these analyses, this information is organized into a list of important information about key process steps.

[0139] The information of the mined key nodes and edges, as well as the key process step information, are integrated to form a comprehensive information set. They can be classified and sorted, for example, the key physical element information and the key process step information are stored separately to form a comprehensive information set. This set contains various information that has an important impact on the performance and quality of the PCB circuit board.

[0140] Screen and organize the comprehensive information set to remove some unimportant or redundant information. According to the actual production experience and quality standards of PCB circuit boards, determine which information must be retained, for example, only retaining information that may affect the performance, reliability or electrical characteristics of the final product, making the comprehensive information set more refined and practical.

[0141] The screened comprehensive information set is encoded to obtain a coding sequence. Different encoding methods are used according to the different categories and importance of information. For physical element information, one encoding rule can be used to convert its element attributes into a segment of code according to a certain order and rule; for process step information, another encoding rule is used to convert the process steps and their parameter attributes into another segment of code. Finally, these different coding fragments are spliced ​​together to form a complete coding sequence.

[0142] The coding sequence is encrypted to obtain the management code. An innovative encryption algorithm is used, which can perform encryption operations based on the length of the coding sequence and the characteristics of the elements therein, making the coding sequence more secure. For example, according to the arrangement order and frequency of occurrence of the elements in the coding sequence, the elements are rearranged or replaced to enhance its security.

[0143] Reference Figure 2 In another embodiment of the present invention, a PCB circuit board processing device based on process parameter optimization is provided, comprising:

[0144] A matching unit is used to extract features from PCB design data to obtain feature data; and to match an adaptive initial process parameter combination according to the feature data;

[0145] An evaluation unit, used to perform quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyze the inspection data using a defect recognition algorithm, and obtain a quality evaluation result;

[0146] An optimization unit, configured to optimize the initial process parameter combination based on the quality evaluation result by using an error back propagation algorithm to obtain an optimized process parameter combination;

[0147] A control unit is used to control the secondary processing of the PCB circuit board based on the optimized process parameter combination.

[0148] In this embodiment, for the specific implementation of each unit in the above device embodiment, please refer to the above method embodiment, which will not be repeated here.

[0149] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0150] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0151] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0152] In summary, the PCB circuit board processing method, device and equipment based on process parameter optimization provided in the embodiment of the present invention include: extracting features from PCB circuit board design data to obtain feature data; matching an adapted initial process parameter combination according to the feature data; performing quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyzing the inspection data using a defect recognition algorithm, and obtaining a quality assessment result; based on the quality assessment result, optimizing the initial process parameter combination using an error back propagation algorithm to obtain an optimized process parameter combination; and controlling the secondary processing of the PCB circuit board based on the optimized process parameter combination. In the present invention, by optimizing the initial process parameter combination according to the quality assessment result of the circuit board, the defect that the current processing method cannot perform adaptive optimization according to the actual situation of the circuit board is overcome.

[0153] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0154] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0155] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A PCB circuit board processing method based on process parameter optimization, characterized in that: The following steps are involved: Extract features from PCB design data to obtain feature data; According to the characteristic data, matching an adaptive initial process parameter combination; Performing quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyzing the inspection data using a defect recognition algorithm, and obtaining a quality assessment result; Based on the quality assessment result, the initial process parameter combination is optimized using an error back propagation algorithm to obtain an optimized process parameter combination; Based on the optimized process parameter combination, the secondary processing of the PCB circuit board is controlled.

2. The PCB circuit board processing method based on process parameter optimization according to claim 1 is characterized in that: The characteristic data includes circuit layout and aperture specifications.

3. The PCB circuit board processing method based on process parameter optimization according to claim 1 is characterized in that: Analyze inspection data using defect recognition algorithms to produce quality assessment results, including: Separating the detection data of the PCB circuit board according to the circuit layer, the dielectric layer and the metal layer to obtain layered detection data; Extracting feature points from each layer of data in the layered detection data to obtain data feature points; Analyze the data feature points corresponding to each layer of data based on the pre-trained defect recognition model to obtain the corresponding defect type and severity; The defect types and severity corresponding to each layer of data are summarized and weighted to obtain the quality assessment result.

4. The PCB circuit board processing method based on process parameter optimization according to claim 1 is characterized in that: Analyze inspection data using defect recognition algorithms to produce quality assessment results, including: Based on multimodal data fusion technology, the detection data from different sensors are weighted and fused to form a high-dimensional comprehensive detection data set; Using a sparse representation algorithm to represent the comprehensive detection data set as sparse data; The sparse data is subjected to local binary pattern calculation to extract the texture information of the circuit board surface, and the various defect features of the circuit board surface are extracted by combining texture feature analysis. Based on the defect features, the nodes and edges of the Bayesian network are constructed; each node represents a defect feature, and the edge represents the probability relationship between the defect features; the Bayesian theorem is used to calculate the probability of quality problems in the PCB circuit board under different defect feature combinations, and the quality status is divided into multiple quality levels according to the probability size to obtain the quality assessment result.

5. The PCB circuit board processing method based on process parameter optimization according to claim 1 is characterized in that: Based on the quality assessment result, the initial process parameter combination is optimized using an error back propagation algorithm to obtain an optimized process parameter combination, including: Based on the pre-established correspondence between quality deviation and process parameter correction value, each deviation in the current quality assessment result is matched to the corresponding process parameter correction range; Based on the process parameter correction range, an error back propagation algorithm is used to iteratively optimize the initial process parameter combination; wherein, in the error back propagation algorithm, the optimization step size is dynamically adjusted according to the real-time feedback of the deviation.

6. The PCB circuit board processing method based on process parameter optimization according to claim 1 is characterized in that: After controlling the secondary processing of the PCB circuit board, it includes: Conduct quality inspection on PCB circuit boards after secondary processing and analyze the secondary quality assessment results; If the secondary quality assessment result meets the preset requirements, the characteristic data and the optimized process parameter combination are mapped to obtain a mapping relationship; Generate a management code based on the characteristic data and the optimized process parameter combination; The mapping relationship is stored, and the management authority of the mapping relationship is set based on the management code.

7. The PCB circuit board processing method based on process parameter optimization according to claim 6 is characterized in that: Based on the characteristic data and the optimized process parameter combination, a management code is generated, including: Performing data cleaning on the characteristic data and the optimized process parameter combination to obtain a data string; Adding characters in the data string to the matrix to generate a character matrix; Based on the characteristics of the characters at each matrix position in the character matrix, a preset coding table is adjusted for a second time to obtain an adjusted coding table; Based on the adjustment coding table, encoding the data string to obtain a coded data string; The management code is generated based on the multi-dimensional features of the encoded data string.

8. A PCB circuit board processing device based on process parameter optimization, characterized in that: include: The matching unit is used to extract features from the PCB design data to obtain feature data; According to the characteristic data, matching an adaptive initial process parameter combination; An evaluation unit, used to perform quality inspection on the PCB circuit board that is first processed based on the initial process parameter combination, analyze the inspection data using a defect recognition algorithm, and obtain a quality evaluation result; An optimization unit, configured to optimize the initial process parameter combination based on the quality evaluation result by using an error back propagation algorithm to obtain an optimized process parameter combination; A control unit is used to control the secondary processing of the PCB circuit board based on the optimized process parameter combination.

9. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images