Method and system for evaluating electromagnetic compatibility of PCB (Printed Circuit Board)

Abstract

The invention discloses a PCB electromagnetic compatibility evaluation method and system, and the method comprises the steps: obtaining first feature data corresponding to a target PCB, and analyzing the first feature data to obtain a first training data set of different dimensions, the dimensions comprising a lamination design, a layout design, a wiring design, a ground design, an interface filtering design and a shielding case design; performing parameterization processing on the first training data set to obtain a second training data set, and training the second training data set through a full-connection network to obtain an electromagnetic compatibility evaluation model; acquiring second feature data corresponding to the PCB to be evaluated, and analyzing the second feature data to obtain a feature vector; and inputting the feature vector into the electromagnetic compatibility evaluation model to obtain an evaluation result. According to the invention, the accuracy and efficiency of PCB electromagnetic compatibility evaluation can be improved, the quality and efficiency of PCB design are further improved, and the applicability is wide.

Description

Technical Field

[0001] The present application relates to the technical field of PCB, and in particular to a method and system for evaluating electromagnetic compatibility of PCB. Background Art

[0002] Printed Circuit Board (PCB) is a crucial electronic interconnection component in electronic products, responsible for connecting various electronic components, achieving circuit connectivity and signal transmission. Before the detailed design of PCB, the first step is to conduct a comprehensive feasibility analysis of PCB based on core elements such as board specifications and device configuration. In this process, electromagnetic compatibility issues become a core consideration that cannot be ignored, so it is crucial to conduct a scientific evaluation of the electromagnetic compatibility of PCB.

[0003] At present, the evaluation of PCB electromagnetic compatibility mainly relies on existing design criteria and engineers' practical experience for manual evaluation. However, when facing multi-layer PCB boards with complex structures, this method shows significant limitations. It not only greatly prolongs the design evaluation cycle of engineers and reduces work efficiency, but also easily introduces design flaws due to deviations in human judgment, thereby increasing the potential risks of products. Therefore, exploring more efficient and accurate PCB electromagnetic compatibility evaluation methods is particularly important for improving the quality and efficiency of multi-layer PCB design.

[0004] Application Contents

[0005] The present application provides a PCB electromagnetic compatibility assessment method and system to improve the accuracy and efficiency of PCB electromagnetic compatibility assessment, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0006] In a first aspect, the present application provides a method for evaluating electromagnetic compatibility of a PCB, comprising:

[0007] Acquire first feature data corresponding to the target PCB board, and parse the first feature data to obtain first training data sets of different dimensions, wherein the dimensions include stack-up design, layout design, wiring design, ground design, interface filtering design, and shielding cover design;

[0008] Performing parameterization processing on the first training data set to obtain a second training data set, and training the second training data set through a fully connected network to obtain an electromagnetic compatibility assessment model;

[0009] Acquire second feature data corresponding to the PCB board to be evaluated, and parse the second feature data to obtain a feature vector;

[0010] The characteristic vector is input into the electromagnetic compatibility evaluation model to obtain an evaluation result.

[0011] The embodiment of the present application performs multi-dimensional decomposition on the feature data corresponding to the extracted target PCB board, which can be accurately decomposed into specific dimensions, which is beneficial to the subsequent training of an electromagnetic compatibility evaluation model that can evaluate multiple dimensions; the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB; the second feature data corresponding to the PCB board to be evaluated is analyzed, which is convenient for subsequent input into the PCB electromagnetic compatibility evaluation model for evaluation; the electromagnetic compatibility evaluation model can accurately and efficiently evaluate the compatibility of the model to be evaluated, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0012] Furthermore, before training the second training data set through the fully connected network, the method further includes:

[0013] Identify abnormal data in the second training data set, and remove the abnormal data in the second training data set.

[0014] In this way, by eliminating abnormal data in the second training data set, the dimension of the training data can be reduced, which helps to improve the accuracy of subsequent model training.

[0015] Furthermore, the second training data set is trained by a fully connected network to obtain an electromagnetic compatibility assessment model, specifically:

[0016] Inputting the second training data set into the fully connected network to obtain an initial evaluation model;

[0017] Calculate the loss value between the initial evaluation model and the true model by using the MSE loss function;

[0018] Based on the loss value, the parameter gradient in the fully connected network is calculated by a back propagation algorithm, and the weight of the fully connected network is updated based on the parameter gradient until the weight of the fully connected network meets a preset threshold, thereby obtaining an electromagnetic compatibility assessment model.

[0019] In this way, the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB.

[0020] Furthermore, the characteristic vector is input into the electromagnetic compatibility evaluation model to obtain an evaluation result, specifically:

[0021] Determining whether the feature vector meets a preset condition, wherein the preset condition includes a design rule and a corresponding scoring standard;

[0022] If the preset conditions are met, the evaluation result corresponding to the feature vector is output.

[0023] In this way, the electromagnetic compatibility evaluation model can accurately and efficiently evaluate the compatibility of the evaluation model, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0024] Furthermore, after obtaining the evaluation result, the method further includes:

[0025] Performing risk analysis on the assessment result to obtain the risk probability corresponding to the feature vector;

[0026] Determine whether the assessment result is less than a preset scoring threshold or whether the risk probability is greater than a preset risk threshold; if the assessment result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, automatically generate a risk optimization plan corresponding to the feature vector.

[0027] In this way, by performing risk analysis on the evaluation results, it is helpful to subsequently generate corresponding risk optimization solutions in a targeted manner, thereby improving the quality and efficiency of PCB design, and the applicability is wide.

[0028] In a second aspect, the present application provides an electromagnetic compatibility evaluation system for a PCB, including: an acquisition module, a training module, a parsing module, and an evaluation module;

[0029] The acquisition module is used to acquire first feature data corresponding to the target PCB board, and parse the first feature data to obtain first training data sets of different dimensions, wherein the dimensions include stacking design, layout design, wiring design, ground design, interface filtering design and shielding cover design;

[0030] The training module is used to perform parameterization processing on the first training data set to obtain a second training data set, and train the second training data set through a fully connected network to obtain an electromagnetic compatibility assessment model;

[0031] The analysis module is used to obtain second feature data corresponding to the PCB board to be evaluated, and analyze the second feature data to obtain a feature vector;

[0032] The evaluation module is used to input the feature vector into the electromagnetic compatibility evaluation model to obtain an evaluation result.

[0033] The embodiment of the present application performs multi-dimensional decomposition on the feature data corresponding to the extracted target PCB board, which can be accurately decomposed into specific dimensions, which is beneficial to the subsequent training of an electromagnetic compatibility evaluation model that can evaluate multiple dimensions; the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB; the second feature data corresponding to the PCB board to be evaluated is analyzed, which is convenient for subsequent input into the PCB electromagnetic compatibility evaluation model for evaluation; the electromagnetic compatibility evaluation model can accurately and efficiently evaluate the compatibility of the model to be evaluated, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0034] Furthermore, the electromagnetic compatibility evaluation system of the PCB also includes: a rejection module;

[0035] The elimination module is used to identify abnormal data in the second training data set and eliminate the abnormal data in the second training data set.

[0036] Furthermore, the training module includes: an input unit, a calculation unit and a training unit;

[0037] The input unit is used to input the second training data set into the fully connected network to obtain an initial evaluation model;

[0038] The calculation unit is used to calculate the loss value between the initial evaluation model and the true model through the MSE loss function;

[0039] The training unit is used to calculate the parameter gradient in the fully connected network through a back propagation algorithm based on the loss value, and update the weight of the fully connected network based on the parameter gradient until the weight of the fully connected network meets a preset threshold, thereby obtaining an electromagnetic compatibility assessment model.

[0040] Further, the evaluation module includes: a judgment unit and an output unit;

[0041] The judging unit is used to judge whether the feature vector meets a preset condition, wherein the preset condition includes a design rule and a corresponding scoring standard;

[0042] If the preset conditions are met, the evaluation result corresponding to the feature vector is output.

[0043] Furthermore, the electromagnetic compatibility evaluation system of the PCB further includes: an analysis module, the analysis module including an analysis unit and a generation unit:

[0044] The analysis unit is used to perform risk analysis on the assessment result to obtain the risk probability corresponding to the feature vector;

[0045] The generation unit is used to determine whether the evaluation result is less than a preset scoring threshold or whether the risk probability is greater than a preset risk threshold. If the evaluation result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, a risk optimization plan corresponding to the feature vector is automatically generated. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a flow chart of an embodiment of a PCB electromagnetic compatibility evaluation method provided by the present application;

[0047] Figure 2 It is a schematic diagram of the training process of the electromagnetic compatibility assessment model provided in this application;

[0048] Figure 3 It is a structural schematic diagram of an embodiment of a PCB electromagnetic compatibility evaluation system provided by the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] It should be understood that the step numbers used in this article are only for the convenience of description and are not intended to limit the order in which the steps are executed.

[0051] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0052] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0053] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0054] PCB is a key interconnection component of electronic products, responsible for connecting components, realizing circuit and signal transmission. Before designing PCB, it is necessary to conduct a feasibility analysis of PCB based on board type, devices and other factors, among which electromagnetic compatibility is the core consideration, so it is very important to scientifically evaluate the electromagnetic compatibility of PCB. At present, the evaluation of electromagnetic compatibility of PCB mainly relies on design criteria and engineer experience. However, it has great limitations when facing complex multi-layer boards, which not only greatly prolongs the design evaluation cycle of engineers and reduces work efficiency, but also easily introduces design defects due to deviations in human judgment, thereby increasing the potential risks of products. Therefore, exploring efficient and accurate electromagnetic compatibility evaluation methods is crucial to improving the quality and efficiency of multi-layer PCB design.

[0055] Next, the nouns involved in this application are analyzed:

[0056] Mean Squared Error (MSE) is a commonly used statistic. The mean squared error is a measure of the difference between the estimator and the estimated quantity. In statistics, if there is an estimator t of a population parameter θ, then the mathematical expectation of (θ-t)2 is called the mean squared error of the estimator t.

[0057] Based on this, the embodiment of the present application provides a PCB electromagnetic compatibility assessment method and system, which can improve the accuracy and efficiency of PCB electromagnetic compatibility assessment, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0058] An electromagnetic compatibility evaluation method and system for a PCB provided in an embodiment of the present application are specifically described through the following embodiments. First, the electromagnetic compatibility evaluation method for a PCB in an embodiment of the present application is described.

[0059] The electromagnetic compatibility assessment method of PCB provided in the embodiment of the present application relates to the technical field of PCB. The electromagnetic compatibility assessment method of PCB provided in the embodiment of the present application can be applied to the terminal, can also be applied to the server side, and can also be software running in the terminal or the server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or can be configured as a server cluster or a distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the electromagnetic compatibility assessment method of PCB, etc., but is not limited to the above forms.

[0060] The present application can also be used in numerous general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0061] Embodiment 1

[0062] Please refer to Figure 1 , Figure 1 It is a flow chart of an embodiment of a PCB electromagnetic compatibility evaluation method provided by the present application, including steps S101 to S104;

[0063] Step S101, obtaining first feature data corresponding to a target PCB board, and parsing the first feature data to obtain a first training data set of different dimensions, wherein the dimensions include stack-up design, layout design, wiring design, ground design, interface filtering design, and shielding cover design;

[0064] It is understandable that the target PCB board needs to be determined first. If there is a design file corresponding to the PCB board, the design file of the target PCB board can be opened using electronic design automation software to obtain the information corresponding to the PCB board. After obtaining the information of the PCB board, it is necessary to perform data preprocessing on the information, and extract features related to stack design, layout design, wiring design, ground layer design, interface filter design and shielding cover design from the preprocessed data to obtain the first feature data, wherein the stack design includes features such as the top layer, power layer, signal layer and bottom layer, the layout design includes features such as important components (such as chips and crystal oscillators) and ordinary components, the wiring design includes features such as line length and line width, the ground design includes features such as analog ground, digital ground and power ground, the interface filter design includes features such as interface filtering, and the shielding cover design includes features such as shielding cover.

[0065] After obtaining the first characteristic data, it is necessary to analyze the first characteristic data and divide it according to different dimensions to form multiple first training data sets, wherein each first training data set corresponds to a specific division dimension, such as stackup design, layout design, wiring design, formation design, interface filtering design and shielding cover design.

[0066] It should be noted that if the data volume of the first training data set is insufficient, data enhancement technology may be used to increase the diversity and quantity of the data, wherein the data enhancement technology may include data transformation, noise addition, data interpolation, etc.

[0067] Step S102, parameterizing the first training data set to obtain a second training data set, and training the second training data set through a fully connected network to obtain an electromagnetic compatibility assessment model;

[0068] It is understandable that after obtaining the first training sample data set containing PCB features, it is necessary to further analyze the first training sample to determine the parameters corresponding to each feature in the first training sample, and convert each sample in the first training data set into a parameterized representation, and integrate all parameterized samples into a new data set, namely the second training data set, to ensure that each sample in the data set contains all necessary parameter information.

[0069] After the second training data set is obtained, it is necessary to identify abnormal data in the second training data set and remove the abnormal data in the second training data set, wherein the abnormal data may be a null value or data beyond a normal parameter range.

[0070] Table 1 Parameters

[0071]

[0072] It can be understood that, first, it is necessary to construct a PCB design structure that meets the electromagnetic compatibility design requirements as the objective function, and input the second training data set into the fully connected network to obtain an initial evaluation model; secondly, the loss value between the initial evaluation model and the true model is calculated through the MSE loss function; finally, based on the loss value, the parameter gradient in the fully connected network is calculated through the back propagation algorithm, and the Adam optimizer is used to update the weight of the fully connected network based on the parameter gradient until the weight of the fully connected network meets the preset threshold, and the optimal weight is obtained. The parameters of the objective function are updated based on the optimal weight, and finally the electromagnetic compatibility evaluation model is obtained. Among them, the schematic diagram of the training process of the electromagnetic compatibility evaluation model is shown in Figure 2 As shown, the calculation formula of the MES loss function is:

[0073]

[0074] In the formula, n represents the number of samples, y represents the true value, Indicates the predicted value of the model. Generally speaking, the smaller the MSE, the more accurate the model's prediction.

[0075] In this way, the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB.

[0076] It should be noted that in order to verify the effectiveness of the electromagnetic compatibility evaluation model, multiple PCBs with good electromagnetic compatibility performance and PCBs with electromagnetic interference problems can be selected and input into the trained electromagnetic compatibility evaluation model. By judging whether they meet the requirements of electromagnetic compatibility, if they meet the requirements of electromagnetic compatibility, a good PCB is output; if they do not meet the requirements of electromagnetic compatibility, a PCB with electromagnetic interference problems is output.

[0077] Step S103, obtaining second characteristic data corresponding to the PCB board to be evaluated, and parsing the second characteristic data to obtain a characteristic vector;

[0078] Since the steps of extracting feature data and analyzing the feature data in this step have been explained in detail above, they will not be repeated here.

[0079] Step S104: input the characteristic vector into the electromagnetic compatibility evaluation model to obtain an evaluation result.

[0080] It is understandable that when the characteristic vector corresponding to the PCB board to be evaluated is obtained, it needs to be input into the electromagnetic compatibility evaluation model. The electromagnetic compatibility evaluation model will perform design evaluation according to the stacking design, layout design, wiring design, ground design, interface filter design and shielding cover design in turn, and judge whether the characteristic vector meets the preset conditions, wherein the preset conditions include design rules and corresponding scoring criteria; if the preset conditions are met, the evaluation result corresponding to the characteristic vector is output. Among them, (1) the design rules of stacking design are: the key signal layer must have a ground plane adjacent return reference; the key main power supply has a corresponding ground plane adjacent return reference; the secondary wiring layer needs to have a ground or power supply layer as a return reference; in general, the second layer and the second to last layer are complete ground planes; both the power supply and ground planes can be used as reference planes, and the ground plane should be preferably selected, and the power plane should be secondarily selected. The scoring criteria are: parameterize the relevant design rules, judge whether the characteristic vector meets the preset conditions, and score if the preset conditions are met, with each item scoring 20 points, and a full score of 100 points. (2) The design rules of layout design are: frequency angle analysis, whether the device layout follows the rule of gradually decreasing frequency from the inside to the outside; power angle analysis, whether the device layout separates high-power and low-power devices; sensitivity angle analysis, whether the device layout separates strong interference devices and sensitive devices. The scoring criteria are: parameterize the relevant design rules, judge whether the characteristic vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (3) The design rules of wiring design are: in order to reduce crosstalk between lines, the line spacing should be no less than 3 times the line width; the power layer should be shrunk to 20 times the thickness of the dielectric layer; key signals such as high-speed signals, clock signals, sensitive signals, etc., have a complete reference plane and a complete ground plane; whether the high-speed signal maintains the same reference plane when changing layers, and if not, whether the reference layer is connected through the nearest ground via. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (4) The design rules for ground design are: whether to set the reference ground plane area according to different power supply voltages and level signals, such as analog ground, digital ground, power ground, etc. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (5) The design rules for interface filter design are: whether the external input cable and the wiring inside the PCB board are isolated through the common mode inductor; whether the interface sets a separate shell ground for common mode capacitor bridging; whether there are other devices or parallel wiring in the projection area of ​​the interface filter circuit. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points.(6) The design rules for shielding cover design are: identify the operating frequency of chips, clocks, crystal oscillators and other devices, and whether to design or reserve shielding covers for devices above 30MHz; determine the device density of PCB, and if the PCB density is high, whether to design or reserve shielding covers for interface power filter circuits; identify circuit functions, and if there are high-speed circuits such as antennas and radio frequency, whether to design or reserve shielding covers for local circuits, etc. The scoring criteria are: parameterize the relevant design rules, and determine whether the feature vector meets the preset conditions. If the preset conditions are met, the score is scored, and each item is scored 20 points, with a full score of 100 points.

[0081] By evaluating the design in the order of stack-up design, layout design, wiring design, ground design, interface filtering design, and shielding cover design, the completion of the PCB design can be scored in multiple dimensions, thereby improving the quality and efficiency of the PCB design, which has wide applicability.

[0082] After obtaining the evaluation result, it is necessary to perform risk analysis on the evaluation result to obtain the risk probability corresponding to the feature vector; and determine whether the evaluation result is less than the preset scoring threshold or whether the risk probability is greater than the preset risk threshold. If the evaluation result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, a risk optimization solution corresponding to the feature vector is automatically generated. In this way, by performing risk analysis on the evaluation result, it is helpful to generate corresponding risk optimization solutions in a targeted manner, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0083] Table 2 EMC design evaluation results of a PCB

[0084]

[0085]

[0086] The embodiment of the present application performs multi-dimensional decomposition on the feature data corresponding to the extracted target PCB board, which can be accurately decomposed into specific dimensions, which is beneficial to the subsequent training of an electromagnetic compatibility evaluation model that can evaluate multiple dimensions; the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB; the second feature data corresponding to the PCB board to be evaluated is analyzed, which is convenient for subsequent input into the PCB electromagnetic compatibility evaluation model for evaluation; the electromagnetic compatibility evaluation model can accurately and efficiently evaluate the compatibility of the model to be evaluated, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0087] Embodiment 2

[0088] Please refer to Figure 3 , Figure 3It is a structural schematic diagram of an embodiment of an electromagnetic compatibility evaluation system for a PCB provided by the present application, comprising: an acquisition module 100, a training module 200, a parsing module 300 and an evaluation module 400;

[0089] The acquisition module 100 is used to acquire first feature data corresponding to the target PCB board, and parse the first feature data to obtain a first training data set of different dimensions, wherein the dimensions include stacking design, layout design, wiring design, ground design, interface filtering design and shielding cover design;

[0090] It is understandable that the target PCB board needs to be determined first. If there is a design file corresponding to the PCB board, the design file of the target PCB board can be opened using electronic design automation software to obtain the information corresponding to the PCB board. After obtaining the information of the PCB board, it is necessary to perform data preprocessing on the information, and extract features related to stack design, layout design, wiring design, ground layer design, interface filter design and shielding cover design from the preprocessed data to obtain the first feature data, wherein the stack design includes features such as the top layer, power layer, signal layer and bottom layer, the layout design includes features such as important components (such as chips and crystal oscillators) and ordinary components, the wiring design includes features such as line length and line width, the ground design includes features such as analog ground, digital ground and power ground, the interface filter design includes features such as interface filtering, and the shielding cover design includes features such as shielding cover.

[0091] After obtaining the first characteristic data, it is necessary to analyze the first characteristic data and divide it according to different dimensions to form multiple first training data sets, wherein each first training data set corresponds to a specific division dimension, such as stackup design, layout design, wiring design, formation design, interface filtering design and shielding cover design.

[0092] It should be noted that if the data volume of the first training data set is insufficient, data enhancement technology may be used to increase the diversity and quantity of the data, wherein the data enhancement technology may include data transformation, noise addition, data interpolation, etc.

[0093] The training module 200 is used to perform parameterization processing on the first training data set to obtain a second training data set, and train the second training data set through a fully connected network to obtain an electromagnetic compatibility assessment model;

[0094] It is understandable that after obtaining the first training sample data set containing PCB features, it is necessary to further analyze the first training sample to determine the parameters corresponding to each feature in the first training sample, and convert each sample in the first training data set into a parameterized representation, and integrate all parameterized samples into a new data set, namely the second training data set, to ensure that each sample in the data set contains all necessary parameter information.

[0095] The electromagnetic compatibility assessment system for PCB further includes: a rejection module; the rejection module is used to identify abnormal data in the second training data set and reject the abnormal data in the second training data set. Specifically, after obtaining the second training data set, it is necessary to identify the abnormal data in the second training data set and reject the abnormal data in the second training data set, wherein the abnormal data may be a null value or data beyond a normal parameter range.

[0096] Table 1 Parameters

[0097]

[0098]

[0099] The training module 200 includes: an input unit, a calculation unit and a training unit; the input unit is used to input the second training data set into the fully connected network to obtain an initial evaluation model; the calculation unit is used to calculate the loss value between the initial evaluation model and the real model through the MSE loss function; the training unit is used to calculate the parameter gradient in the fully connected network through the back propagation algorithm based on the loss value, and update the weight of the fully connected network based on the parameter gradient until the weight of the fully connected network meets the preset threshold, and obtain the electromagnetic compatibility evaluation model. Specifically, first, it is necessary to construct a PCB design structure that meets the electromagnetic compatibility design requirements as an objective function, and input the second training data set into the fully connected network to obtain an initial evaluation model; secondly, the loss value between the initial evaluation model and the real model is calculated through the MSE loss function; finally, based on the loss value, the parameter gradient in the fully connected network is calculated through the back propagation algorithm, and the weight of the fully connected network is updated based on the parameter gradient using the Adam optimizer until the weight of the fully connected network meets the preset threshold, and the optimal weight is obtained, and the parameters of the objective function are updated based on the optimal weight, and finally the electromagnetic compatibility evaluation model is obtained. The training process diagram of the electromagnetic compatibility assessment model is shown in Figure 2 As shown, the calculation formula of the MES loss function is:

[0100]

[0101] In the formula, n represents the number of samples, y represents the true value, Indicates the predicted value of the model. Generally speaking, the smaller the MSE, the more accurate the model's prediction.

[0102] In this way, the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB.

[0103] It should be noted that in order to verify the effectiveness of the electromagnetic compatibility evaluation model, multiple PCBs with good electromagnetic compatibility performance and PCBs with electromagnetic interference problems can be selected and input into the trained electromagnetic compatibility evaluation model. By judging whether they meet the requirements of electromagnetic compatibility, if they meet the requirements of electromagnetic compatibility, a good PCB is output; if they do not meet the requirements of electromagnetic compatibility, a PCB with electromagnetic interference problems is output.

[0104] The analysis module 300 is used to obtain second feature data corresponding to the PCB board to be evaluated, and analyze the second feature data to obtain a feature vector;

[0105] Since the steps of extracting feature data and analyzing the feature data in this step have been explained in detail above, they will not be repeated here.

[0106] The evaluation module 400 is used to input the feature vector into the electromagnetic compatibility evaluation model to obtain an evaluation result.

[0107] The evaluation module 400 includes: a judgment unit and an output unit; the judgment unit is used to judge whether the feature vector meets the preset conditions, wherein the preset conditions include design rules and corresponding scoring standards; if the preset conditions are met, the evaluation result corresponding to the feature vector is output. Specifically, it can be understood that when the feature vector corresponding to the PCB board to be evaluated is obtained, it needs to be input into the electromagnetic compatibility evaluation model, and the electromagnetic compatibility evaluation model will perform design evaluation according to stack design, layout design, wiring design, ground design, interface filter design and shielding cover design in sequence, by judging whether the feature vector meets the preset conditions, wherein the preset conditions include design rules and corresponding scoring standards; if the preset conditions are met, the evaluation result corresponding to the feature vector is output. Among them, (1) the design rules of stack design are: the key signal layer must have a ground plane adjacent return reference; the key main power supply has a corresponding ground plane adjacent return reference; the secondary wiring layer needs to have a ground or power supply layer as a return reference; in general, the second layer and the second to last layer are complete ground planes; both the power supply and ground planes can be used as reference planes, and the ground plane should be preferred, and the power plane should be secondary. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (2) The design rules for layout design are: frequency perspective analysis, whether the device layout follows the rule of gradually decreasing frequency from the inside to the outside; power perspective analysis, whether the device layout separates high-power and low-power devices; sensitivity perspective analysis, whether the device layout separates strong interference devices and sensitive devices. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (3) The design rules for wiring design are: in order to reduce crosstalk between lines, the line spacing should be no less than 3 times the line width; the power layer should be shrunk to 20 times the thickness of the dielectric layer; whether key signals such as high-speed signals, clock signals, sensitive signals, etc. have a complete reference plane and a complete ground plane; whether high-speed signals maintain the same reference plane when changing layers, and if not, whether the reference layer is connected through the nearest ground via. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (4) The design rules for ground design are: whether the reference ground plane area is set separately according to different power supply voltages and level signals, such as analog ground, digital ground, power ground, etc. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points.(5) The design rules for interface filter design are: whether the external input cable and the wiring inside the PCB board are isolated by common-mode inductors; whether the interface has a separate shell ground for common-mode capacitor bridging; whether there are other devices or parallel wiring in the projection area of ​​the interface filter circuit. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points. (6) The design rules for shielding cover design are: identify the operating frequency of chips, clocks, crystal oscillators and other devices, and whether to design or reserve shielding covers for devices above 30MHz; determine the device density of the PCB, and if the PCB density is high, whether to design or reserve shielding covers for the interface power filter circuit; identify the circuit function, and if there are high-speed circuits such as antennas and radio frequencies, whether to design or reserve shielding covers for local circuits, etc. The scoring criteria are: parameterize the relevant design rules, determine whether the feature vector meets the preset conditions, and score if it meets the preset conditions. Each item is scored 20 points, with a total score of 100 points.

[0108] By evaluating the design in the order of stack-up design, layout design, wiring design, ground design, interface filtering design, and shielding cover design, the completion of the PCB design can be scored in multiple dimensions, thereby improving the quality and efficiency of the PCB design, which has wide applicability.

[0109] Among them, the electromagnetic compatibility assessment system of the PCB also includes: an analysis module, the analysis module includes an analysis unit and a generation unit: the analysis unit is used to perform risk analysis on the assessment result to obtain the risk probability corresponding to the feature vector; the generation unit is used to determine whether the assessment result is less than a preset scoring threshold or whether the risk probability is greater than a preset risk threshold, if the assessment result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, then automatically generate a risk optimization scheme relative to the feature vector. Specifically, after obtaining the assessment result, it is necessary to perform risk analysis on the assessment result to obtain the risk probability corresponding to the feature vector; and determine whether the assessment result is less than the preset scoring threshold or whether the risk probability is greater than the preset risk threshold, if the assessment result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, then automatically generate a risk optimization scheme relative to the feature vector. In this way, by performing risk analysis on the assessment result, it is helpful to generate corresponding risk optimization schemes in a targeted manner in the subsequent process, thereby improving the quality and efficiency of PCB design, and the applicability is wide.

[0110] Table 2 EMC design evaluation results of a PCB

[0111]

[0112]

[0113] The embodiment of the present application performs multi-dimensional decomposition on the feature data corresponding to the extracted target PCB board, which can be accurately decomposed into specific dimensions, which is beneficial to the subsequent training of an electromagnetic compatibility evaluation model that can evaluate multiple dimensions; the electromagnetic compatibility evaluation model is trained by the second training data set, which is helpful for the subsequent accurate evaluation of the electromagnetic compatibility of the PCB; the second feature data corresponding to the PCB board to be evaluated is analyzed, which is convenient for subsequent input into the PCB electromagnetic compatibility evaluation model for evaluation; the electromagnetic compatibility evaluation model can accurately and efficiently evaluate the compatibility of the model to be evaluated, thereby improving the quality and efficiency of PCB design, and has wide applicability.

[0114] The device embodiments described above are merely illustrative, wherein the modules described as separate components may or may not be physically separated, i.e., may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the method of this embodiment.

[0115] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-monitorable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0116] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application.

[0117] It is particularly pointed out that for those skilled in the art, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A method for evaluating electromagnetic compatibility of a PCB, characterized in that: include: Acquire first feature data corresponding to the target PCB board, and parse the first feature data to obtain first training data sets of different dimensions, wherein the dimensions include stack-up design, layout design, wiring design, ground design, interface filtering design, and shielding cover design; Performing parameterization processing on the first training data set to obtain a second training data set, and training the second training data set through a fully connected network to obtain an electromagnetic compatibility assessment model; Acquire second feature data corresponding to the PCB board to be evaluated, and parse the second feature data to obtain a feature vector; The characteristic vector is input into the electromagnetic compatibility evaluation model to obtain an evaluation result.

2. The electromagnetic compatibility evaluation method of PCB according to claim 1, characterized in that: Before training the second training data set through the fully connected network, the method further includes: Identify abnormal data in the second training data set, and remove the abnormal data in the second training data set.

3. The electromagnetic compatibility evaluation method of PCB according to claim 1, characterized in that: The second training data set is trained by a fully connected network to obtain an electromagnetic compatibility assessment model, specifically: Inputting the second training data set into the fully connected network to obtain an initial evaluation model; Calculate the loss value between the initial evaluation model and the true model by using the MSE loss function; Based on the loss value, the parameter gradient in the fully connected network is calculated by a back propagation algorithm, and the weight of the fully connected network is updated based on the parameter gradient until the weight of the fully connected network meets a preset threshold, thereby obtaining an electromagnetic compatibility assessment model.

4. The electromagnetic compatibility evaluation method of PCB according to claim 1, characterized in that: The inputting the characteristic vector into the electromagnetic compatibility evaluation model to obtain the evaluation result is specifically: Determining whether the feature vector meets a preset condition, wherein the preset condition includes a design rule and a corresponding scoring standard; If the preset conditions are met, the evaluation result corresponding to the feature vector is output.

5. The electromagnetic compatibility evaluation method of PCB according to claim 1, characterized in that: After obtaining the evaluation results, the method further includes: Performing risk analysis on the assessment result to obtain the risk probability corresponding to the feature vector; Determine whether the assessment result is less than a preset scoring threshold or whether the risk probability is greater than a preset risk threshold; if the assessment result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, automatically generate a risk optimization plan corresponding to the feature vector.

6. A PCB electromagnetic compatibility evaluation system, characterized in that: include: Acquisition module, training module, parsing module and evaluation module; The acquisition module is used to acquire first feature data corresponding to the target PCB board, and parse the first feature data to obtain first training data sets of different dimensions, wherein the dimensions include stacking design, layout design, wiring design, ground design, interface filtering design and shielding cover design; The training module is used to perform parameterization processing on the first training data set to obtain a second training data set, and train the second training data set through a fully connected network to obtain an electromagnetic compatibility assessment model; The analysis module is used to obtain second feature data corresponding to the PCB board to be evaluated, and analyze the second feature data to obtain a feature vector; The evaluation module is used to input the feature vector into the electromagnetic compatibility evaluation model to obtain an evaluation result.

7. The electromagnetic compatibility evaluation system for PCB according to claim 6, characterized in that: Also includes: Eliminate module; The elimination module is used to identify abnormal data in the second training data set and eliminate the abnormal data in the second training data set.

8. The electromagnetic compatibility evaluation system for PCB according to claim 6, characterized in that: The training module includes: an input unit, a calculation unit and a training unit; The input unit is used to input the second training data set into the fully connected network to obtain an initial evaluation model; The calculation unit is used to calculate the loss value between the initial evaluation model and the true model through the MSE loss function; The training unit is used to calculate the parameter gradient in the fully connected network through a back propagation algorithm based on the loss value, and update the weight of the fully connected network based on the parameter gradient until the weight of the fully connected network meets a preset threshold, thereby obtaining an electromagnetic compatibility assessment model.

9. The electromagnetic compatibility evaluation system for PCB according to claim 6, characterized in that: The evaluation module includes: a judgment unit and an output unit; The judging unit is used to judge whether the feature vector meets a preset condition, wherein the preset condition includes a design rule and a corresponding scoring standard; If the preset conditions are met, the evaluation result corresponding to the feature vector is output.

10. The electromagnetic compatibility evaluation system for PCB according to claim 6, characterized in that: Also includes: An analysis module, the analysis module comprising an analysis unit and a generation unit: The analysis unit is used to perform risk analysis on the assessment result to obtain the risk probability corresponding to the feature vector; The generation unit is used to determine whether the evaluation result is less than a preset scoring threshold or whether the risk probability is greater than a preset risk threshold. If the evaluation result is less than the preset scoring threshold or the risk probability is greater than the preset risk threshold, a risk optimization plan corresponding to the feature vector is automatically generated.

Images