CN116542135A - Microwave filter shape coupling relation hybrid modeling method, device and storage device - Google Patents
Microwave filter shape coupling relation hybrid modeling method, device and storage device Download PDFInfo
- Publication number
- CN116542135A CN116542135A CN202310394727.1A CN202310394727A CN116542135A CN 116542135 A CN116542135 A CN 116542135A CN 202310394727 A CN202310394727 A CN 202310394727A CN 116542135 A CN116542135 A CN 116542135A
- Authority
- CN
- China
- Prior art keywords
- parameter
- microwave filter
- shape coupling
- coupling relation
- representing
- 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.)
- Pending
Links
- 230000008878 coupling Effects 0.000 title claims abstract description 97
- 238000010168 coupling process Methods 0.000 title claims abstract description 97
- 238000005859 coupling reaction Methods 0.000 title claims abstract description 97
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 239000011159 matrix material Substances 0.000 claims description 35
- 238000013528 artificial neural network Methods 0.000 claims description 27
- 238000005070 sampling Methods 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 230000005540 biological transmission Effects 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000002310 reflectometry Methods 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 4
- 230000009286 beneficial effect Effects 0.000 abstract description 4
- 238000013507 mapping Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000000903 blocking effect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000003786 synthesis reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000011031 large-scale manufacturing process Methods 0.000 description 1
- 238000003754 machining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000007493 shaping process Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0499—Feedforward networks
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medical Informatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Control Of Motors That Do Not Use Commutators (AREA)
Abstract
The invention provides a microwave filter shape coupling relation hybrid modeling method, equipment and storage equipment, which construct an original data set D 1 And D 2 Extracting D by vector fitting method 1 Y parameter Y in (a) 1 Build up of resonant screw length x 1 And Y parameter Y 1 Data set Y of (2) 1 Extracting D 2 Y parameter Y in (a) 2 Build up of coupling screw length x 2 And Y parameter Y 2 Data set Y of (2) 2 The method comprises the steps of carrying out a first treatment on the surface of the Designing a microwave filter shape coupling hybrid model: the shape coupling relation model under the serious detuning state and the shape coupling relation model under the slight detuning state realize the secondary resonanceScrew length x 1 Length of coupling screw x 2 To Y parameter Y 1 、y 2 Is mapped to the mapping of (a). The beneficial effects of the invention are as follows: the microwave filter is divided into two states of slight detuning and serious detuning, different structural geometric parameters are selected in a targeted mode to establish a shape coupling relation model, the complexity of establishing the shape coupling relation model is reduced, and the accuracy of the shape coupling relation model under various states is improved.
Description
Technical Field
The invention relates to the field of microwave filters, in particular to a method, equipment and storage equipment for modeling a shape coupling relation of a microwave filter in a mixed mode.
Background
As the international 5G competition becomes more and more aggressive, the 5G base station construction accelerates. The microwave filter is a core passive component for controlling the frequency response of a transmission signal in the 5G base station, transmits useful signal frequency components, blocks useless signal frequency components, and is important for optimizing spectrum resource allocation and improving the quality of a communication system. However, in the production process of the microwave filter, machining tolerance cannot be avoided, the microwave filter obtained by batch processing of the production line cannot meet the requirements of factory filtering performance indexes, and the performance debugging process is indispensable.
The debugging workers measure scattering parameters (S parameters) in real time based on the vector network analyzer, and filter performance (performance) is calculated. When the filtering performance does not meet the factory requirement, structural geometric parameters (shapes) influencing the filtering performance are selected according to experience, the adjusting quantity is determined, and the technical index is changed until the filtering performance meets the requirement. But the manual debugging has low efficiency, high cost and high rejection rate, prevents the construction process of the 5G base station, and intelligent debugging becomes a bottleneck for high-quality and large-scale production of microwave filters to be broken through.
The accurate prediction of the filtering performance of the microwave filter is the basis of intelligent debugging, so that the frequent blind adjustment of the structural geometric parameters of the actual microwave filter can be avoided, and the debugging time cost and the rejection rate are greatly reduced. The structural geometric parameters of the shape and the product performance are in strong coupling, strong nonlinearity and strong dynamic characteristics, a shape coupling relation model is built, the product performance is accurately predicted, and the intelligent debugging foundation is achieved.
The filtering performance of the microwave filter is frequently changed in the debugging process, and the mixed modeling of the microwave filter based on the resonance state is beneficial to the improvement of modeling accuracy. However, in the existing modeling method, a geometric coupling relation model between a certain characteristic parameter and a structural geometric parameter of the microwave filter is established by data-driven modeling methods such as support vector regression and a neural network, the influence of a resonance state is not considered, the dynamic characteristic of the whole debugging process of the microwave filter is difficult to comprehensively reflect, and the modeling precision is low.
Disclosure of Invention
In order to solve the problems, the invention provides a microwave filter shape coupling relation hybrid modeling method, a device and a storage device, wherein the debugging process is divided into two resonance states of serious detuning and good tuning according to the filtering performance; and the geometric parameters of different structures are selected to establish the geometric coupling relation model for different resonance states, so that the accuracy of the geometric coupling relation model under various states is improved.
A microwave filter shape coupling relation hybrid modeling method mainly comprises the following steps:
s1: construction of the raw dataset D 1 And D 2 : the structural geometrical parameters of the microwave filter comprise the length x of the resonant screw 1 And coupling screw length x 2 The method comprises the steps of carrying out a first treatment on the surface of the Multiple x-changes on microwave filter 1 Sampling and measuring S parameter S 1 Build comprising x 1 、s 1 Raw data set D at sampling frequency f 1 The method comprises the steps of carrying out a first treatment on the surface of the Multiple x-changes on microwave filter 2 Sampling and measuring S parameter S 2 Build comprising x 2 、s 2 Raw data set D at sampling frequency f 2 ；
S2: extracting D by vector fitting method 1 Middle s 1 Y parameter Y of (2) 1 Build up of resonant screw length x 1 And Y parameter Y 1 Data set Y of (2) 1 The method comprises the steps of carrying out a first treatment on the surface of the Extracting D by vector fitting method 2 Middle s 2 Y parameter Y of (2) 2 Build up of coupling screw length x 2 And Y parameter Y 2 Data set Y of (2) 2 ；
S3: according to the debugging characteristics of the microwave filter and the Y parameter Y 1 And y 2 Is characterized by designing microwave filterA wave-shaping coupling hybrid model;
s4: construction of a shape coupling relation model M in a severely detuned state using a BP neural network 1 Realizing the length x of the resonant screw 1 To Y parameter Y 1 Is mapped to;
s5: building a shape coupling relation model M under good tuning state by using BP neural network 2 Realize the length x of the coupling screw 2 To Y parameter Y 2 Is mapped to;
s6: according to the Y parameter Y 1 And y 2 And converting into an S parameter to predict the filtering performance index.
Further, in step S2, the Y matrix is obtained by converting the S parameterIs a complex matrix, S 11 Representing the reflectivity of the input signal energy, S 21 Representing the transmission rate of the input signal energy, S 12 Representing the transmission rate of the output signal energy, S 21 The reflectivity of the output signal energy is represented, and the S parameter is converted into a Y matrix by the following formula (1) to formula (4):
(1)
(2)
(3)
(4)
wherein Y is 0 Is a unitary matrix, Y 11 Representing the relationship between the input signal voltage and the input signal current, Y 12 Representing the relationship between the output signal voltage and the input signal current, Y 21 Representing the relationship between the input signal voltage and the output signal current, Y 22 Representing output signal voltage and input signalRelationship of number current.
Further, based on a vector fitting method, extracting Y parameters in a Y matrix:
(5)
where s=jω, ω represents the angular frequency corresponding to the sampling frequency f, λ k Is the kth pole of the Y matrix, r ijk For the remainders corresponding to the kth pole, i, j=1, 2, k=1, 2, … …, N represents the order of the microwave filter, the poles of the Y matrix are denoted as l= [ lambda ] 1 ,λ 2 , …, λ N ,]The remainder of the Y matrix is denoted r ij =[r ij1,…, r ijN ]The poles and remainders of the Y matrix are called the Y parameter.
Further, in step S2, the dimension of the Y parameter is reduced, and after the dimension reduction, Y is selected 11 Pole imaginary parts Im (l), Y 11 The remainder of the real part Re (r 11 )、Y 21 The remainder of the real part Re (r 21 ) To reflect the Y parameter, i.e. the reduced dimension Y parameter y= [ Im (l), re (r) 11 ), Re(r 21 )]，r 11 Represents Y 11 Leave the number r 21 Represents Y 21 And (5) reserving.
Further, in step S3, the modeling process is mainly divided into two steps: based on the BP neural network, a shape coupling relation model of the microwave filter in a severe detuning state and a shape coupling relation model of the microwave filter in a good tuning state are established, and the shape coupling relation model of the microwave filter in the severe detuning state and the shape coupling relation model of the microwave filter in the good tuning state are combined to obtain the shape coupling mixed model of the microwave filter.
The storage device stores instructions and data for implementing a microwave filter shape coupling relation hybrid modeling method.
A microwave filter shape coupling relationship hybrid modeling apparatus, comprising: a processor and the storage device; the processor loads and executes the instructions and data in the storage device to realize a microwave filter shape coupling relation hybrid modeling method.
The technical scheme provided by the invention has the beneficial effects that: the microwave filter is divided into two states of slight detuning and serious detuning, different structural geometric parameters are selected in a targeted mode to establish a shape coupling relation model, the complexity of establishing the shape coupling relation model is reduced, and the accuracy of the shape coupling relation model under various states is improved.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a microwave filter shape coupling relation hybrid modeling method in an embodiment of the invention.
FIG. 2 is a schematic diagram of a microwave filter shape coupling hybrid model designed in an embodiment of this invention.
Fig. 3 is a schematic diagram of an electromagnetic model of a sixth-order microwave filter built in the HFSS according to an embodiment of the present invention.
FIG. 4 is a graph of the comparative results of a model of the shape coupling relationship in severely detuned micro states in an embodiment of the invention.
Fig. 5 is a graph of the comparison of the model of the shape coupling relationship in the well-tuned state in the embodiment of the present invention.
FIG. 6 is a schematic diagram of the operation of a hardware device in an embodiment of the invention.
Detailed Description
For a clearer understanding of technical features, objects and effects of the present invention, a detailed description of embodiments of the present invention will be made with reference to the accompanying drawings.
The embodiment of the invention provides a microwave filter shape coupling relation hybrid modeling method, a device and a storage device.
Referring to fig. 1, fig. 1 is a flowchart of a microwave filter shape coupling relation hybrid modeling method in an embodiment of the invention, which specifically includes:
s1: in the data acquisition phase, the original data set D is constructed 1 And D 2 . The structural geometrical parameters of the microwave filter comprise the length x of the resonant screw 1 And coupling screw length x 2 . Multiple changes on microwave filterVariable x 1 Sampling and measuring S parameter S 1 Build comprising x 1 、s 1 Raw data set D at sampling frequency f 1 The method comprises the steps of carrying out a first treatment on the surface of the Multiple x-changes on microwave filter 2 Sampling and measuring S parameter S 2 Build comprising x 2 、s 2 Raw data set D at sampling frequency f 2 。
S2: extracting D by vector fitting method 1 Middle s 1 Y parameter Y of (2) 1 Build up of resonant screw length x 1 And Y parameter Y 1 Data set Y of (2) 1 The method comprises the steps of carrying out a first treatment on the surface of the Extracting D by vector fitting method 2 Middle s 2 Y parameter Y of (2) 2 Build up of coupling screw length x 2 And Y parameter Y 2 Data set Y of (2) 2 。
First, by S parameter (including S 1 Sum s 2 ) Converting to obtain Y matrix (Y parameter is pole and residue of Y matrix, and Y parameter is Y after dimension reduction 11 Pole imaginary parts Im (lambda), Y 11 The remainder of the real part Re (r 11 ) And Y 21 The remainder of the real part Re (r 21 )). S parameterIs a complex matrix, S 11 Representing the reflectivity of the input signal energy, S 21 Representing the transmission rate of the input signal energy, S 12 Representing the transmission rate of the output signal energy, S 22 The reflectivity, i, representing the output signal energy is the imaginary unit. S parameters are converted into a Y matrix through the formulas (1) to (4):
(1)
(2)
(3)
(4)
wherein Y is 0 Is Y with 11 Identity matrix of the same dimension, Y 11 Representing the relationship between the input signal voltage and the input signal current, Y 12 Representing the relationship between the output signal voltage and the input signal current, Y 21 Representing the relationship between the input signal voltage and the output signal current, Y 22 The relationship between the output signal voltage and the input signal current is shown.
And secondly, extracting Y parameters in the Y matrix based on a vector fitting method. The microwave filter Y matrix may be expressed in the form of a polynomial as follows,
(5)
in the formula (5), s=jω, ω represents an angular frequency corresponding to the sampling frequency f, λ k Is the kth pole of the Y matrix, r ijk For the remainder corresponding to the kth pole, i, j=1, 2, the poles of the y matrix are denoted as l= [ λ ] 1 ,λ 2 , …, λ N ,]K=1, 2, … …, N represents the order of the microwave filter, and the remainder of the Y matrix is denoted r ij =[r ij1,…, r ijN ]The poles and remainders of the Y matrix are called Y parameters, and are complex. In this embodiment, the Y parameter is obtained by solving the most commonly used vector fitting method.
And finally, reducing the dimension of the Y parameter. The extraction process of the Y parameter is based on the principle that poles of all elements in the Y matrix are equal, so that only the poles of one element in the Y parameter matrix are needed in the modeling process, and in the embodiment, Y is selected 11 The poles reflect the pole characteristics of the Y matrix; y is Y 11 And Y 22 、Y 21 And Y 12 The properties are substantially consistent, the invention selects Y 11 And Y 21 Reflecting the remainder characteristics of the Y matrix; the poles and the remainder of the Y matrix are both imaginary numbers, the input and output of the neural network are both real numbers, certain processing is needed, and the real parts of the poles and the imaginary parts of the remainder of the Y matrix are both 0. Through dimension reduction, the embodiment selects Y 11 Pole imaginary parts Im (l), Y 11 The remainder of the real part Re (r 11 )、Y 21 The remainder of the real part Re (r 21 ) To reflect the Y parameter, i.e. Y parameter y= [ Im (l), re (r) 11 ), Re(r 21 )]，r 11 Represents Y 11 Leave the number r 21 Represents Y 21 And (5) reserving.
S3: overall design of microwave filter shape coupling hybrid model: and (3) carrying out overall design on the microwave filter shape coupling hybrid model according to the debugging characteristics and the characteristics of the Y parameters. Because the filter is divided into two states, the two states are required to be modeled respectively, the overall design structure diagram of the shape coupling hybrid model is shown in fig. 2, and the modeling process is mainly divided into two steps: establishing a shape coupling relation model in a severe detuning state of the microwave filter and a shape coupling relation model in a good tuning state of the microwave filter, and finally, combining the established neural networks, namely, the shape coupling relation model M in the severe detuning state of the microwave filter 1 Shape coupling relation model M under good tuning state with microwave filter 2 And combining to obtain the microwave filter shape coupling mixed model.
S4: building a shape coupling relation model under a serious detuning state: the data set Y is processed according to the most common ten-fold crossing method 1 The method is divided into a training set and a testing set, and adopts a blocking modeling method in patent CN109783905B to construct a shape coupling relation model M under a serious detuning state by using a BP neural network 1 Realizing the length x of the resonant screw 1 To Y parameter Y 1 Is mapped to the mapping of (a).
S5: building a shape coupling relation model under a slight detuning state: the data set Y is processed according to the most common ten-fold crossing method 2 The method is divided into a training set and a testing set, and adopts a blocking modeling method in patent CN109783905B to construct a shape coupling relation model M in a micro-detuning state by using a BP neural network 2 Realize the length x of the coupling screw 2 To Y parameter Y 2 Is mapped to the mapping of (a).
And finally, outputting to obtain a complete Y parameter by utilizing the microwave filter shape coupling mixed model, and converting the Y parameter into an S parameter to predict the filtering performance index.
And a simulation debugging platform is built based on the three-dimensional electromagnetic software HFSS and MATLAB. A six-order cross-coupling-free microwave filter simulation model shown in fig. 3 is established in three-dimensional electromagnetic software HFSS, S parameters are solved, and a formal coupling relation model is established in MATLAB. The designed microwave filter has 6 resonant screws and 5 coupling screws.
As shown in FIG. 2, in severe detuning, the pole imaginary part data set, Y, of the resonant screw length, Y parameter is utilized 11 And Y 21 The neural network 1, the neural network 2 and the neural network 3 are trained by the left real part data set, after all training of the three neural networks (namely the BP neural network 1, the BP neural network 2 and the BP neural network 3) is completed, the modeling process of the severe detuning filter is measured by the test set, the length of the resonance screw is given, the Y parameter is obtained by the output synthesis of the three neural networks (namely the output of the shape coupling relation model under the severe detuning state), and then the Y parameter is converted into the S parameter for display. Amplitude-phase response curve of ideal S parameter and S output by model m The relationship between the amplitude-phase response curves is shown in fig. 4. Therefore, the amplitude-frequency curve and the phase-frequency curve are well fitted, and the geometric coupling relation model constructed by aiming at the serious detuning state is effective. When the microwave filter is seriously detuned, the resonant screw needs to be regulated, and the shape coupling relation model M in the seriously detuned state is used 1 The filtering performance can be predicted.
With slight detuning, using pole imaginary data sets, Y, of coupling screw length and Y parameters 11 And Y 21 The neural network 4, the neural network 5 and the neural network 6 are trained by the left real part data set, after all training of the three neural networks (namely the BP neural network 4, the BP neural network 5 and the BP neural network 6) is completed, the modeling process of the tight slight harmonic filter is measured by the test set, the length of the coupling screw rod is given, the Y parameter is obtained by the output synthesis of the three neural networks (namely the output of the shape coupling relation model in the slight detuning state), and then the Y parameter is converted into the S parameter for display. Amplitude-phase response curve of ideal S parameter and S output by model m The relationship between the amplitude-phase response curves is shown in fig. 5. It can be seen that both amplitude-frequency curve and phase-frequency curve are simulatedWell-suited, the present invention is effective against a slightly detuned state-built shape coupling relation model. When the microwave filter is slightly detuned, the coupling screw rod needs to be adjusted, and the shape coupling relation model M in the slightly detuned state is used 2 The filtering performance can be predicted.
Referring to fig. 6, fig. 6 is a schematic working diagram of a hardware device according to an embodiment of the present invention, where the hardware device specifically includes: a microwave filter shape coupling relation hybrid modeling device 401, a processor 402 and a storage device 403.
A microwave filter shape coupling relation hybrid modeling device 401: the microwave filter shape coupling relation hybrid modeling device 401 implements the microwave filter shape coupling relation hybrid modeling method.
Processor 402: the processor 402 loads and executes instructions and data in the memory device 403 for implementing the one microwave filter shape coupling relationship hybrid modeling method.
Storage device 403: the storage device 403 stores instructions and data; the storage device 403 is configured to implement the microwave filter shape coupling relation hybrid modeling method.
The beneficial effects of the invention are as follows: the microwave filter is divided into two states of slight detuning and serious detuning, different structural geometric parameters are selected in a targeted mode to establish a shape coupling relation model, the complexity of establishing the shape coupling relation model is reduced, and the accuracy of the shape coupling relation model under various states is improved.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (7)
1. A microwave filter shape coupling relation hybrid modeling method is characterized in that: comprising the following steps:
s1: construction of the raw dataset D 1 And D 2 : the structural geometrical parameters of the microwave filter comprise the length of the resonant screwx 1 And coupling screw length x 2 The method comprises the steps of carrying out a first treatment on the surface of the Multiple x-changes on microwave filter 1 Sampling and measuring S parameter S 1 Build comprising x 1 、s 1 Raw data set D at sampling frequency f 1 The method comprises the steps of carrying out a first treatment on the surface of the Multiple x-changes on microwave filter 2 Sampling and measuring S parameter S 2 Build comprising x 2 、s 2 Raw data set D at sampling frequency f 2 ；
S2: extracting D by vector fitting method 1 Middle s 1 Y parameter Y of (2) 1 Build up of resonant screw length x 1 And Y parameter Y 1 Data set Y of (2) 1 The method comprises the steps of carrying out a first treatment on the surface of the Extracting D by vector fitting method 2 Middle s 2 Y parameter Y of (2) 2 Build up of coupling screw length x 2 And Y parameter Y 2 Data set Y of (2) 2 ；
S3: according to the debugging characteristics of the microwave filter and the Y parameter Y 1 And y 2 The characteristics of the model are that a microwave filter shape coupling mixed model is designed;
s4: construction of a shape coupling relation model M in a severely detuned state using a BP neural network 1 Realizing the length x of the resonant screw 1 To Y parameter Y 1 Is mapped to;
s5: building a shape coupling relation model M under good tuning state by using BP neural network 2 Realize the length x of the coupling screw 2 To Y parameter Y 2 Is mapped to;
s6: according to the Y parameter Y 1 And y 2 And converting into an S parameter to predict the filtering performance index.
2. The method for modeling the microwave filter shape coupling relation in a hybrid manner according to claim 1, wherein the method comprises the following steps: in step S2, the Y matrix is obtained by converting the S parameterIs a complex matrix, S 11 Representing the reflectivity of the input signal energy, S 21 Representing the transmission rate of the input signal energy, S 12 Representing the inputThe transmission rate of the output signal energy S 22 The reflectivity of the output signal energy is represented, and the S parameter is converted into a Y matrix by the following formula (1) to formula (4):
(1)
(2)
(3)
(4)
wherein Y is 0 Is a unitary matrix, Y 11 Representing the relationship between the input signal voltage and the input signal current, Y 12 Representing the relationship between the output signal voltage and the input signal current, Y 21 Representing the relationship between the input signal voltage and the output signal current, Y 22 The relationship between the output signal voltage and the input signal current is shown.
3. The method for modeling the microwave filter shape coupling relation in a hybrid manner according to claim 1, wherein the method comprises the following steps: in step S2, based on the vector fitting method, the Y parameter in the Y matrix is extracted:
(5)
wherein Y is 11 Representing the relationship between the input signal voltage and the input signal current, Y 12 Representing the relationship between the output signal voltage and the input signal current, Y 21 Representing the relationship between the input signal voltage and the output signal current, Y 22 Representing the relationship between the output signal voltage and the input signal current, s=jω, ω representing the angular frequency corresponding to the sampling frequency f, λ k Is the kth pole of the Y matrix, r ijk For the remainders corresponding to the kth pole, i, j=1, 2, k=1, 2, … …, N represents the order of the microwave filter, the poles of the Y matrix are denoted as l= [ lambda ] 1 ,λ 2 , …, λ N ,]The remainder of the Y matrix is denoted r ij =[r ij1,…, r ijN ]The poles and remainders of the Y matrix are called the Y parameter.
4. A method for modeling a microwave filter shape coupling relationship as defined in claim 3, wherein: in step S2, the dimension of the Y parameter is reduced, and Y is selected after the dimension reduction 11 Pole imaginary parts Im (l), Y 11 The remainder of the real part Re (r 11 )、Y 21 The remainder of the real part Re (r 21 ) To reflect the Y parameter, i.e. the reduced dimension Y parameter y= [ Im (l), re (r) 11 ), Re(r 21 )]，r 11 Represents Y 11 Leave the number r 21 Represents Y 21 And (5) reserving.
5. The method for modeling the microwave filter shape coupling relation in a hybrid manner according to claim 1, wherein the method comprises the following steps: in step S3, the modeling process is mainly divided into two steps: based on the BP neural network, a shape coupling relation model of the microwave filter in a severe detuning state and a shape coupling relation model of the microwave filter in a good tuning state are established, and the shape coupling relation model of the microwave filter in the severe detuning state and the shape coupling relation model of the microwave filter in the good tuning state are combined to obtain the shape coupling mixed model of the microwave filter.
6. A memory device, characterized by: the storage device stores instructions and data for implementing the microwave filter shape coupling relation hybrid modeling method according to any one of claims 1 to 5.
7. A microwave filter shape coupling relation mixed modeling device is characterized in that: comprising the following steps: a processor and a storage device; the processor loads and executes the instructions and data in the storage device to implement the microwave filter shape coupling relation hybrid modeling method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310394727.1A CN116542135A (en) | 2023-04-13 | 2023-04-13 | Microwave filter shape coupling relation hybrid modeling method, device and storage device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310394727.1A CN116542135A (en) | 2023-04-13 | 2023-04-13 | Microwave filter shape coupling relation hybrid modeling method, device and storage device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116542135A true CN116542135A (en) | 2023-08-04 |
Family
ID=87444410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310394727.1A Pending CN116542135A (en) | 2023-04-13 | 2023-04-13 | Microwave filter shape coupling relation hybrid modeling method, device and storage device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116542135A (en) |
-
2023
- 2023-04-13 CN CN202310394727.1A patent/CN116542135A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109783905B (en) | Intelligent microwave cavity filter debugging method based on particle swarm optimization algorithm | |
| CN108170922B (en) | Auxiliary debugging method and device for microwave filter and storage device | |
| CN105789812B (en) | A kind of cavity body filter Automatic adjustment method and system | |
| CN109063374A (en) | A kind of coupling matrix extracting method based on parameter optimization, equipment and storage equipment | |
| CN103888104B (en) | Method and system for designing FIR digital filter | |
| Zhang et al. | Yield-constrained optimization design using polynomial chaos for microwave filters | |
| CN104008248B (en) | The injection forming process based Robust Design and Tolerance Design Method based on Gaussian process | |
| CN105893669A (en) | Global simulation performance predication method based on data digging | |
| CN107562990A (en) | BAW filter design methods, device and equipment based on BVD models | |
| CN112305485B (en) | Method and device for correcting harmonic voltage measurement error of capacitor voltage transformer | |
| CN109301428A (en) | A kind of optimization method of the coupling matrix element of filter, equipment and storage equipment | |
| CN105024131A (en) | Design method for microwave cavity band-pass filter | |
| Caenepeel et al. | Efficient and automated generation of multidimensional design curves for coupled-resonator filters using system identification and metamodels | |
| WO2024099061A1 (en) | Frequency sweeping method and system for adaptive frequency point sampling, and related device | |
| CN116542135A (en) | Microwave filter shape coupling relation hybrid modeling method, device and storage device | |
| CN111159881B (en) | Rapid optimization design method applied to millimeter wave extension interaction oscillator | |
| CN115329974B (en) | Simulation method, simulation device, simulation equipment and storage medium | |
| CN115906303A (en) | Planar microwave filter design method and device based on machine learning | |
| CN102236727B (en) | Method for computer aided design of microwave filter | |
| CN101976294A (en) | Space mapping algorithm-based automatic computer-aid tuning design method for microwave device | |
| CN115329973A (en) | Simulation method, device, equipment and storage medium | |
| CN104217090A (en) | Method for analyzing and calculating voltage stabilization critical point of electric power system through Taylor series | |
| CN103474737A (en) | Millimeter wave E-plane filter for supporting vector machine to model for diaphragm and diaphragm modeling method | |
| CN114114117A (en) | Magnetic resonance spectrum phase correction method based on deep learning | |
| Xu et al. | Optimization of coaxial dielectric resonator filter with aggressive space mapping |
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 |