CN111832247B - Method and device for determining size of via hole anti-pad based on BP neural network - Google Patents
Method and device for determining size of via hole anti-pad based on BP neural network Download PDFInfo
- Publication number
- CN111832247B CN111832247B CN202010589240.5A CN202010589240A CN111832247B CN 111832247 B CN111832247 B CN 111832247B CN 202010589240 A CN202010589240 A CN 202010589240A CN 111832247 B CN111832247 B CN 111832247B
- Authority
- CN
- China
- Prior art keywords
- neural network
- size
- pad
- parasitic capacitance
- output vector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 111
- 238000000034 method Methods 0.000 title claims abstract description 56
- 230000003071 parasitic effect Effects 0.000 claims abstract description 47
- 238000004513 sizing Methods 0.000 claims description 10
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 9
- 239000002648 laminated material Substances 0.000 claims description 9
- 239000011889 copper foil Substances 0.000 claims description 8
- 238000005553 drilling Methods 0.000 claims description 8
- 239000011800 void material Substances 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 4
- 229910052802 copper Inorganic materials 0.000 claims 1
- 239000010949 copper Substances 0.000 claims 1
- 238000005457 optimization Methods 0.000 abstract description 21
- 230000000694 effects Effects 0.000 abstract description 7
- 230000006870 function Effects 0.000 description 7
- 210000002569 neuron Anatomy 0.000 description 5
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000630 rising effect Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/398—Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/392—Floor-planning or layout, e.g. partitioning or placement
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Architecture (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measurement Of Resistance Or Impedance (AREA)
- Production Of Multi-Layered Print Wiring Board (AREA)
Abstract
The application discloses a via hole reverse pad size determining method based on a BP (back propagation) neural network, which can construct the BP neural network taking parameters influencing parasitic capacitance as input and taking indexes for measuring the parasitic capacitance as output, can adaptively optimize a via hole by utilizing the trained BP neural network, and can adaptively adjust the size of a reverse pad only by setting an expected value to obtain the expected optimal size. The variable priority is not required to be manually set, the experience requirement of a designer is low, and the optimization precision and efficiency can be improved. In addition, the application also provides a device, equipment and a readable storage medium for determining the size of the via hole anti-pad based on the BP neural network, and the technical effects of the device and the equipment correspond to the technical effects of the method.
Description
Technical Field
The present application relates to the field of integrated circuit technologies, and in particular, to a method, an apparatus, a device, and a readable storage medium for determining a size of a via anti-pad based on a BP neural network.
Background
Along with the continuous improvement of the signal rate, the rising edge of the signal is continuously reduced, and the influence of the via hole on the whole high-speed link is larger and larger. For the server industry, with the arrival of PCIE5.0, the signal rate reaches 32G, and the signal rising edge further decreases, so that optimization of the via hole is particularly important.
Many structural parameters of the via have a significant impact on the signal. For example, via stubs can cause severe degradation of high-speed signal quality, which can cause severe reflections and resonances. The parasitic parameters of the via hole are determined by parameters such as the aperture size of the via hole, the size of the pad, the position of the reflow ground hole, the size of the anti-pad and the like. At present, the problem that the via stub can be well solved in a mode of reasonably planning the layer surface or adding back drilling when the PCB is distributed is solved, the influence of a via on a signal is mainly caused by parasitic capacitance and inductance effect, wherein the influence of the parasitic capacitance on the signal is dominant, and the sizes of a via pad, a reflow ground hole and an anti-pad all influence the parasitic capacitance, so that the optimization of the via pad and the anti-pad becomes a main optimization means for reducing the capacitance of the via, which is also one of main directions of the optimization of the via in the current PCB design.
At present, when the via hole optimization is performed in the industry, the position coordinates of a bonding pad, a GND hole and the size of an anti-bonding pad are set as variables, a scanning range is set for the variables, and the optimal values are found by traversing the influence of each variable on impedance and return loss. When the number of variables is large, in order to simplify the optimization process, a designer needs to determine the priority of the variables by experience, and then needs to perform the optimization step by step according to the priority of the variables, which consumes a lot of time, and the obtained optimized value may not meet the actual requirement at the same time.
Therefore, the existing via hole optimization scheme needs designers to have rich experience, has poor optimization effect and consumes long time.
Disclosure of Invention
The application aims to provide a method, a device, equipment and a readable storage medium for determining the size of a via hole anti-pad based on a BP (back propagation) neural network, which are used for solving the problems that the prior via hole optimization scheme needs to set variable priority according to experience, the optimization effect is poor, and the consumed time is long.
The specific scheme is as follows:
in a first aspect, the present application provides a via antipad size determination method based on a BP neural network, including:
obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance;
constructing a BP neural network according to the input vector and the output vector;
training the BP neural network by using the training sample, wherein the size of the anti-bonding pad is a variable in the training process, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants;
and obtaining the optimal size of the anti-bonding pad by utilizing the trained BP neural network through back propagation according to the target value of the index for measuring the parasitic capacitance.
Preferably, the input vector includes an anti-pad size, and further includes any one or more of: drilling size, via pad size, laminated material dielectric constant, copper foil roughness, and loss factor.
Preferably, the output vector comprises impedance and/or return loss.
Preferably, the constructing a BP neural network according to the input vector and the output vector includes:
and constructing a BP neural network according to a target formula according to the input vector and the output vector, wherein the target formula is as follows:
RL＝f1(v1*Df+v2*d_hole+v3*d_pad_sigal+v4*d_void_via)，
Z＝f2(ω1*DK+ω2*roughf+ω3*d_hole+ω4*d_pad_sigal+ω5*d_void_via)；
wherein Z is impedance and RL is return loss; d _ void _ via is the size of an anti-bonding pad, d _ hole is the size of a drilled hole, d _ pad _ signal is the size of a via hole bonding pad, DK is the dielectric constant of the laminated material, rough is the roughness of the copper foil, and Df is a loss factor; f. of1() And f2() Representing two calculation relationships, v1,v2,v3,v4And ω1，ω2，ω3，ω4，ω5Are network weights.
Preferably, the constructing a BP neural network according to the input vector and the output vector includes:
and constructing a BP neural network according to the input vector and the output vector, and setting the number of hidden layer nodes of the BP neural network to be 3.
Preferably, the training the BP neural network by using the training sample includes:
s1, inputting the input vector in the training sample into the BP neural network to obtain an actual output vector;
s2, calculating the expected error between the actual output vector and the output vector in the training sample;
s3, judging whether the expected error is smaller than a preset threshold value; if so, judging that the training is finished, otherwise, adjusting the network weight of the BP neural network, and entering S1.
Preferably, the adjusting the network weight of the BP neural network, entering S1, includes:
judging whether the current iteration times reach the maximum iteration times or not; if so, judging that the training is finished, otherwise, adjusting the network weight of the BP neural network, and entering S1.
In a second aspect, the present application provides an apparatus for determining a via antipad size based on a BP neural network, including:
a training sample acquisition module: the method comprises the steps of obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance;
a neural network construction module: the BP neural network is constructed according to the input vector and the output vector;
a neural network training module: the BP neural network is trained by utilizing the training sample, the size of the anti-bonding pad is a variable in the training process, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants;
a size determination module: and obtaining the optimal size of the anti-bonding pad by utilizing the trained BP neural network through back propagation according to the target value of the index for measuring the parasitic capacitance.
In a third aspect, the present application provides a via antipad sizing device based on a BP neural network, including:
a memory: for storing a computer program;
a processor: for executing the computer program to implement the steps of the BP neural network-based via antipad sizing method described above.
In a fourth aspect, the present application provides a readable storage medium having stored thereon a computer program for implementing the steps of the BP neural network-based via antipad sizing method as described above when executed by a processor.
The application provides a via hole anti-pad size determination method based on a BP neural network, which comprises the following steps: obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance; constructing a BP neural network according to the input vector and the output vector; training the BP neural network by using a training sample, wherein the size of an anti-bonding pad is a variable in the training process, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants; and obtaining the optimal size of the anti-bonding pad by utilizing the trained BP neural network through back propagation according to the target value of the index for measuring the parasitic capacitance.
Therefore, the method can construct the BP neural network which takes the parameters influencing the parasitic capacitance as input and takes the index for measuring the parasitic capacitance as output, the trained BP neural network can be used for adaptively optimizing the via hole, and the size of the anti-bonding pad can be adaptively adjusted only by setting an expected value, so that the expected optimal size is obtained. The variable priority is not required to be manually set, the experience requirement of a designer is low, and the optimization precision and efficiency can be improved.
In addition, the application also provides a device, equipment and a readable storage medium for determining the size of the via hole anti-pad based on the BP neural network, and the technical effect of the device and the equipment corresponds to the technical effect of the method, and the details are not repeated here.
Drawings
For a clearer explanation of the embodiments or technical solutions of the prior art of the present application, the drawings needed for the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart illustrating a first implementation of a via antipad size determination method based on a BP neural network according to an embodiment of the present disclosure;
fig. 2 is a flowchart illustrating an implementation of a second method for determining a size of a via anti-pad based on a BP neural network according to an embodiment of the present disclosure;
FIG. 3 is a functional block diagram of an embodiment of a via antipad sizing apparatus based on a BP neural network according to the present application;
fig. 4 is a schematic structural diagram of an embodiment of a via antipad sizing device based on a BP neural network provided in the present application.
Detailed Description
The core of the application is to provide a method, a device, equipment and a readable storage medium for determining the size of a via hole anti-pad based on a BP neural network. The variable priority is not required to be manually set, the experience requirement of a designer is low, and the optimization precision and efficiency can be improved.
In order that those skilled in the art will better understand the disclosure, the following detailed description will be given with reference to the accompanying drawings. It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without making any creative effort belong to the protection scope of the present application.
Referring to fig. 1, a first embodiment of a method for determining a via anti-pad size based on a BP neural network provided by the present application is described below, where the first embodiment includes:
s101, obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance;
s102, constructing a BP neural network according to the input vector and the output vector;
s103, training the BP neural network by using the training sample, wherein in the training process, the size of the anti-bonding pad is variable, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants;
and S104, obtaining the optimal size of the anti-bonding pad by utilizing the trained BP neural network through back propagation according to the target value of the index for measuring the parasitic capacitance.
Specifically, the input vector includes an anti-pad size, and further includes any one or more of: drilling size, via pad size, laminated material dielectric constant, copper foil roughness, and loss factor. The output vector includes impedance and/or return loss.
In the process of constructing the BP neural network, the number of neurons in each layer of the BP neural network is determined according to the input vector output vector, and a transfer function and a training function of the neurons in each layer are set.
The BP neural network is a multi-layer feedforward neural network trained according to an error back propagation algorithm. Before training the feedforward neural network, initializing the weight and the threshold, and setting the maximum iteration times and the training target. Inputting the input vector in the training sample into a BP neural network to obtain an actual output vector, calculating the error between the actual output vector and the output vector in the training sample, and stopping training when the error reaches a training target or the iteration number reaches the maximum iteration number. In the process of network simulation, when the error signal is reversely transmitted, the BP neural network reversely transmits the error signal according to the original forward transmission path, and adjusts the connection weight system of each neuron of each hidden layer so as to lead the expected error signal to tend to be minimum, thereby realizing the self-adaptive adjustment.
It should be noted that, in an actual PCB design, the via hole drilling, the pad size, and the like are fixed values, and the only variable to be adjusted is the anti-pad size. Therefore, in the training process of the BP neural network, only the size of the anti-bonding pad is variable, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constant.
The method for determining the size of the via hole anti-pad based on the BP neural network can construct the BP neural network which takes parameters influencing parasitic capacitance as input and indexes for measuring the parasitic capacitance as output, the via hole can be adaptively optimized by utilizing the trained BP neural network, and the size of the anti-pad can be adaptively adjusted only by setting an expected value, so that the expected optimal size is obtained. The priority of the variable is not required to be manually set, the requirement on experience of a designer is low, and the optimization precision and efficiency can be improved.
The second embodiment of the method for determining the size of the via anti-pad based on the BP neural network provided by the present application is described in detail below, and the second embodiment is implemented based on the first embodiment and is expanded to a certain extent on the basis of the first embodiment.
Specifically, in the second embodiment, the input vector includes an anti-pad size, a drill hole size, a via pad size, a dielectric constant of a laminated material, a roughness of a copper foil, and a loss factor, and the output vector includes an impedance and a return loss.
Referring to fig. 2, the second embodiment specifically includes:
s201, obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises an anti-pad size, a drilling size, a via pad size, a laminated material dielectric constant, copper foil roughness and a loss factor, and the output vector comprises impedance and return loss;
s202, constructing a BP neural network according to the input vector and the output vector, initializing network weight, and setting maximum iteration times and a preset threshold;
s203, inputting the input vector in the training sample into the BP neural network to obtain an actual output vector;
s204, calculating an expected error between the actual output vector and an output vector in the training sample;
s205, judging whether the expected error is smaller than a preset threshold value; if yes, judging that the training is finished, and entering S207; otherwise, go to S206;
s206, judging whether the current iteration times reach the maximum iteration times; if so, judging that the training is finished, and entering S207; otherwise, adjusting the network weight of the BP neural network, and entering S203;
specifically, during the training process, the size of the anti-pad is variable, and parameters influencing parasitic capacitance except the size of the anti-pad are constant.
And S207, obtaining the optimal size of the anti-pad by utilizing the trained BP neural network through back propagation according to the target value of the impedance and the target value of the return loss.
Specifically, first, an input variable and an output variable are determined, and for the output variable impedance Z, the correlated input variables include: the drilling dimension d _ hole, the via pad dimension d _ pad _ signal, the reverse pad dimension d _ void _ via, the dielectric constant DK of the laminated material, and the roughness of the copper foil rough, therefore, the relationship is as follows:
RL＝f1(v1*Df+v2*d_hole+v3*d_pad_sigal+v4*d_void_via)；
for the output variable return loss RL, its dependent variables include: the via hole size, via pad size, anti-pad size, and dissipation factor Df, therefore, the relationship is as follows:
Z＝f2(ω1*DK+ω2*roughf+ω3*d_hole+ω4*d_pad_sigal+ω5*d_void_via)。
in the above two relations, f1() And f2() Representing two calculation relationships, v1,v2,v3,v4And ω1，ω2，ω3，ω4，ω5All are network weights, which are obtained by training the network. In general, MATLAB software is used without assigning values by itself, and net (minmax (p), [3,2 ]) is used]After { 'tansig', 'logsig' }, 'train lm') it is automatically assigned, ranging from 0,1]In the meantime. In the training process, the network weight and the size of the anti-pad d _ void _ via are variables, and the rest parameters are constants.
An example procedure is as follows, taking adaptive optimization of via impedance as an example:
% constructing input vector p in training sample
p＝[3.18 3.5 10 20 30；
3.18 3.5 10 20 32；
3.18 3.5 10 20 34；
……
3.18 3.5 10 20 46]
% creates a BP network, sets the number of hidden layer neurons to 3, and the transfer function to tansig
% set middle layer neuron number 3, transfer function logsig, training function rainlm
net＝newff(minmax(p),[3,2],{'tansig','logsig'},'trainlm')；
% set maximum number of iterations to 500
net.trainParam.epochs＝500；
% setting training target, i.e. default threshold value is 0.01 by default
net.trainParam.goal＝0.01；
% BP neural network training
net＝train(net,p)；
y＝sim(net,p)；
It can be seen that, in order to solve the problem that the adaptive degree of optimization of the via hole is not high at present, the method for determining the size of the via hole anti-pad based on the BP neural network provided in this embodiment adopts a reverse thinking, and the BP neural network is used to implement adaptive optimization of the via hole in a result-oriented manner. In practical application, only expected impedance and return loss need to be set, and the size of the anti-pad can be adaptively adjusted to meet the expected standard, so that the phenomenon of discontinuous impedance of the interconnection line is improved. The optimization mode taking the target as the guide has lower requirements on the experience of an optimization designer, and can improve the optimization precision and efficiency.
In the following, a device for determining a size of a via antipad based on a BP neural network according to an embodiment of the present application is introduced, and a device for determining a size of a via antipad based on a BP neural network described below and a method for determining a size of a via antipad based on a BP neural network described above may be referred to correspondingly.
As shown in fig. 3, the apparatus for determining a via antipad size based on a BP neural network of this embodiment includes:
the training sample acquisition module 301: the method comprises the steps of obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance;
the neural network construction module 302: the BP neural network is constructed according to the input vector and the output vector;
the neural network training module 303: the BP neural network is trained by utilizing the training sample, the size of the anti-bonding pad is a variable in the training process, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants;
the size determination module 304: and obtaining the optimal size of the anti-bonding pad by utilizing the trained BP neural network through back propagation according to the target value of the index for measuring the parasitic capacitance.
The device for determining the size of the via anti-pad based on the BP neural network of this embodiment is used to implement the method for determining the size of the via anti-pad based on the BP neural network, and therefore a specific implementation manner in the device can be seen in the embodiment parts of the method for determining the size of the via anti-pad based on the BP neural network of the foregoing, for example, the training sample obtaining module 301, the neural network constructing module 302, the neural network training module 303, and the size determining module 304 are respectively used to implement steps S101, S102, S103, and S104 in the method for determining the size of the via anti-pad based on the BP neural network. Therefore, specific embodiments thereof may be referred to in the description of the corresponding respective partial embodiments, and will not be described herein.
In addition, since the device for determining the size of the via anti-pad based on the BP neural network of the present embodiment is used for implementing the method for determining the size of the via anti-pad based on the BP neural network, the function corresponds to the function of the method, and details are not described here.
In addition, the present application also provides a via antipad sizing device based on a BP neural network, as shown in fig. 4, including:
the memory 401: for storing a computer program;
the processor 402: for executing the computer program for implementing the steps of the BP neural network based via antipad sizing method as described above.
Finally, the present application provides a readable storage medium having stored thereon a computer program for implementing the steps of the BP neural network based via antipad sizing method as described above when executed by a processor.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The above detailed descriptions of the solutions provided in the present application, and the specific examples applied herein are set forth to explain the principles and implementations of the present application, and the above descriptions of the examples are only used to help understand the method and its core ideas of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (9)
1. A via hole anti-pad size determination method based on a BP neural network is characterized by comprising the following steps:
obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance;
constructing a BP neural network according to the input vector and the output vector;
training the BP neural network by using the training sample, wherein the size of the anti-bonding pad is a variable in the training process, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants;
obtaining the optimal size of the anti-bonding pad by utilizing the trained BP neural network through back propagation according to the target value of the index for measuring the parasitic capacitance;
wherein the constructing a BP neural network according to the input vector and the output vector comprises:
and constructing a BP neural network according to a target formula according to the input vector and the output vector, wherein the target formula is as follows:
RL=f1(v 1 *Df+v 2 *d_hole+v 3 *d_pad_sigal+v 4 *d_void_via)，
Z=f2(ω 1 *DK+ω 2 *rough+ω 3 *d_hole+ω 4 *d_pad_sigal+ω 5 *d_void_via)；
wherein Z is impedance and RL is return loss; d _ void _ via is the anti-pad size, d _ hole is the drilling size, d _ pad _ signal is the via pad size, DK is the dielectric constant of the laminated material, rough is the roughness of the copper foilDf is a loss factor; f. of1() And f2() Two kinds of calculation relations are represented,v 1 ,v 2 ,v 3 ,v 4andω 1 ,ω 2 ,ω 3 ,ω 4 ,ω 5are network weights.
2. The method of claim 1, wherein the input vector includes an anti-pad size, further comprising any one or more of: drilling size, via pad size, laminated material dielectric constant, copper foil roughness, and loss factor.
3. The method of claim 2, wherein the output vector comprises impedance and/or return loss.
4. The method of claim 1, wherein constructing a BP neural network from the input vector and the output vector comprises:
and constructing a BP neural network according to the input vector and the output vector, and setting the number of hidden layer nodes of the BP neural network to be 3.
5. The method of any one of claims 1-4, wherein the training the BP neural network with the training samples comprises:
s1, inputting the input vector in the training sample into the BP neural network to obtain an actual output vector;
s2, calculating the expected error between the actual output vector and the output vector in the training sample;
s3, judging whether the expected error is smaller than a preset threshold value; if so, judging that the training is finished, otherwise, adjusting the network weight of the BP neural network, and entering S1.
6. The method of claim 5, wherein the adjusting the network weights of the BP neural network, proceeding to S1, comprises:
judging whether the current iteration times reach the maximum iteration times or not; if so, judging that the training is finished, otherwise, adjusting the network weight of the BP neural network, and entering S1.
7. A via antipad size determination device based on a BP neural network is characterized by comprising:
a training sample acquisition module: the method comprises the steps of obtaining a training sample, wherein the training sample comprises an input vector and an output vector, the input vector comprises more than two parameters influencing parasitic capacitance, the parameters influencing the parasitic capacitance comprise the size of an anti-bonding pad, and the output vector comprises an index for measuring the parasitic capacitance;
a neural network construction module: the BP neural network is constructed according to the input vector and the output vector;
the neural network training module: the BP neural network is trained by utilizing the training sample, the size of the anti-bonding pad is a variable in the training process, and parameters influencing parasitic capacitance except the size of the anti-bonding pad are constants;
a size determination module: the BP neural network is used for obtaining the optimal size of the anti-bonding pad through back propagation according to the target value of the index for measuring the parasitic capacitance;
the neural network construction module is specifically used for:
and constructing a BP neural network according to a target formula according to the input vector and the output vector, wherein the target formula is as follows:
RL=f1(v 1 *Df+v 2 *d_hole+v 3 *d_pad_sigal+v 4 *d_void_via)，
Z=f2(ω 1 *DK+ω 2 *rough+ω 3 *d_hole+ω 4 *d_pad_sigal+ω 5 *d_void_via)；
wherein Z is impedance and RL is return loss; d _ void _ via is the size of an anti-bonding pad, d _ hole is the size of a drilled hole, d _ pad _ signal is the size of a via hole bonding pad, DK is the dielectric constant of the laminated material, rough is the roughness of the copper foil, and Df is a loss factor; f. of1() And f2() Two kinds of calculation relations are represented,v 1 ,v 2 ,v 3 ,v 4andω 1 ,ω 2 ,ω 3 ,ω 4 ,ω 5are network weights.
8. A via antipad sizing device based on a BP neural network, comprising:
a memory: for storing a computer program;
a processor: for executing the computer program for implementing the steps of the method for determining via antipad size based on a BP neural network according to any one of claims 1 to 6.
9. A readable storage medium having stored thereon a computer program for implementing the steps of the BP neural network based via antipad sizing method of any one of claims 1-6 when executed by a processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010589240.5A CN111832247B (en) | 2020-06-24 | 2020-06-24 | Method and device for determining size of via hole anti-pad based on BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010589240.5A CN111832247B (en) | 2020-06-24 | 2020-06-24 | Method and device for determining size of via hole anti-pad based on BP neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111832247A CN111832247A (en) | 2020-10-27 |
| CN111832247B true CN111832247B (en) | 2022-06-03 |
Family
ID=72898249
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010589240.5A Active CN111832247B (en) | 2020-06-24 | 2020-06-24 | Method and device for determining size of via hole anti-pad based on BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111832247B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395807B (en) * | 2020-11-12 | 2022-07-12 | 苏州浪潮智能科技有限公司 | Method and system for optimizing coupling of via hole and in-out wire after capacitance |
| CN112747663B (en) * | 2020-12-11 | 2023-06-02 | 浪潮电子信息产业股份有限公司 | Method, device and system for detecting hollowed-out size of differential via anti-bonding pad |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109240885A (en) * | 2018-08-30 | 2019-01-18 | 郑州云海信息技术有限公司 | A kind of method for monitoring performance, system and electronic equipment and storage medium |
| CN110765723A (en) * | 2019-11-15 | 2020-02-07 | 苏州浪潮智能科技有限公司 | Routing modeling optimization method and device based on BP neural network |
| CN111310400A (en) * | 2020-02-16 | 2020-06-19 | 苏州浪潮智能科技有限公司 | BP neural network-based capacitance anti-pad optimization method and system |
-
2020
- 2020-06-24 CN CN202010589240.5A patent/CN111832247B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109240885A (en) * | 2018-08-30 | 2019-01-18 | 郑州云海信息技术有限公司 | A kind of method for monitoring performance, system and electronic equipment and storage medium |
| CN110765723A (en) * | 2019-11-15 | 2020-02-07 | 苏州浪潮智能科技有限公司 | Routing modeling optimization method and device based on BP neural network |
| CN111310400A (en) * | 2020-02-16 | 2020-06-19 | 苏州浪潮智能科技有限公司 | BP neural network-based capacitance anti-pad optimization method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111832247A (en) | 2020-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111832247B (en) | Method and device for determining size of via hole anti-pad based on BP neural network | |
| CN111310400B (en) | BP neural network-based capacitance anti-pad optimization method and system | |
| CN113613388B (en) | Method, circuit board, equipment and storage medium for optimizing via hole anti-pad routing | |
| CN110765723B (en) | Routing modeling optimization method and device based on BP neural network | |
| CN112395807B (en) | Method and system for optimizing coupling of via hole and in-out wire after capacitance | |
| WO2020073857A1 (en) | Method for determining shape of anti-pad and printed circuit board | |
| CN111970835B (en) | Differential via hole and method, system, equipment and medium for manufacturing differential via hole | |
| CN115249005B (en) | Method, system and related equipment for automatic layout of radio frequency front-end module | |
| CN115496026A (en) | Method, device, equipment and storage medium for controlling isochronism of differential pair | |
| CN116056341A (en) | Wiring method and wiring system for signal wires of printed circuit board | |
| CN111200566A (en) | Network service flow information grooming method and electronic equipment | |
| US20050068800A1 (en) | High speed memory interface | |
| CN114564909A (en) | Decoupling capacitor optimization selection method and system | |
| CN113947056A (en) | Differential routing design method and differential line | |
| CN112004321A (en) | Design method for surface-mounted connector on circuit board and circuit board | |
| CN106446429B (en) | Method for dividing complex plane of printed circuit board | |
| CN112001139A (en) | PCB selection method, system and device | |
| EP4167147A1 (en) | Artificial intelligence planning method and artificial intelligence planning device | |
| CN104797078A (en) | Method for reducing impedance mismatching degree under condition of discontinuous returning path | |
| JPH1049568A (en) | Circuit board designing method and recording medium | |
| CN117725883A (en) | Differential via optimization method and device, electronic equipment and storage medium | |
| CN114423184B (en) | Special-shaped anti-bonding pad manufacturing method and device, electronic equipment and storage medium | |
| CN117990378B (en) | Engine sound quality evaluation method and device, electronic equipment and storage medium | |
| CN113609808A (en) | PCIExpress signal integrity improvement method for navigation display system | |
| CN115859901A (en) | PCB (printed Circuit Board) routing layer changing design method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |