Structure optimization method and system for suspension line 3dB bridge

Abstract

The invention provides a structure optimization method and system for a suspension line 3dB bridge, relates to the field of radio frequency devices, is applied to the suspension line 3dB bridge, and comprises the steps that a plurality of structure optimization indexes are determined, and the structure optimization indexes are related to a plurality of physical fields; obtaining the structural information of the 3dB bridge of the suspension line to be optimized; based on the structure information of the suspension line 3dB bridge to be optimized, a plurality of structure parameters to be optimized are determined, and the structure parameters to be optimized comprise the structure parameters of the center double-sided coupled transmission line and the structure parameters of the side double-sided coupled transmission line; according to the method, multiple structure optimization schemes are generated based on multiple to-be-optimized structure parameters, an optimal structure optimization scheme is determined based on the multiple structure optimization schemes through an improved genetic algorithm and finite element analysis, and the method has the advantages that the structure of the suspension line 3dB bridge is optimized, and the stability of the suspension line 3dB bridge is improved.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency devices, and particularly to a structural optimization method and system for a suspended stripline 3dB bridge. Background Art

[0002] A 3dB bridge is a microwave passive device and belongs to a type of hybrid coupler. Its core function is to equally distribute an input signal to two output ports, and there is a 90° phase difference between the two output signals. In radio frequency and microwave systems, 3dB bridges are widely used in scenarios such as signal synthesis, distribution, balanced amplification, and phase modulation. A suspended stripline is a microwave transmission line structure where the conductor strip line is suspended above the dielectric substrate and fixed by support posts or air bridges. Compared with traditional microstrip lines or strip lines, since there is an air gap between the conductor strip line of the suspended stripline and the dielectric substrate, the dielectric loss is reduced.

[0003] In the prior art, adjusting the structural parameters of the suspended stripline 3dB bridge depends on experience or the trial-and-error method, resulting in low optimization efficiency and difficulty in finding the global optimal solution. Moreover, the structural optimization of the suspended stripline 3dB bridge only focuses on the electrical performance (such as S parameters) of the suspended stripline 3dB bridge, while ignoring multi-physical field coupling problems such as electromagnetic field distribution, dielectric loss, and thermal effects.

[0004] Therefore, there is a need to provide a structural optimization method and system for a suspended stripline 3dB bridge to optimize the structure of the suspended stripline 3dB bridge and improve the stability of the suspended stripline 3dB bridge. Summary of the Invention

[0005] The present invention provides a structural optimization method for a suspended line 3dB bridge, which is applied to a suspended line 3dB bridge. The suspended line 3dB bridge includes four ports, a dielectric substrate, a first microstrip line, a second microstrip line, a third microstrip line, a fourth microstrip line, and a double-sided coupled transmission line group disposed on the dielectric substrate. The four ports are respectively connected to the first microstrip line, the second microstrip line, the third microstrip line, and the fourth microstrip line. The first microstrip line and the second microstrip line are located on one side of the double-sided coupled transmission line group, and the third microstrip line and the fourth microstrip line are located on the other side of the double-sided coupled transmission line group. The double-sided coupled transmission line group includes a central double-sided coupled transmission line and at least two side double-sided coupled transmission lines, and the at least two side double-sided coupled transmission lines are symmetrically arranged around the central double-sided coupled transmission line. The method includes: determining a plurality of structural optimization indexes, where the plurality of structural optimization indexes are related to a plurality of physical fields; obtaining the structural information of the suspended line 3dB bridge to be optimized; based on the structural information of the suspended line 3dB bridge to be optimized, determining a plurality of structural parameters to be optimized, where the structural parameters to be optimized include the structural parameters of the central double-sided coupled transmission line and the structural parameters of the side double-sided coupled transmission lines; based on the plurality of structural parameters to be optimized, generating a variety of structural optimization schemes, and through an improved genetic algorithm and finite element analysis, determining the optimal structural optimization scheme based on the variety of structural optimization schemes.

[0006] Further, determining a plurality of structural optimization indexes includes: determining a plurality of structural optimization indexes to be screened and a plurality of structural parameters to be optimized to be screened; based on the plurality of structural parameters to be optimized to be screened and a parameter constraint set, generating a variety of sample structural schemes; for each sample structural scheme, through finite element analysis, determining the scores of the sample structural scheme in the plurality of structural optimization indexes; for each structural optimization index to be screened, based on the scores of each sample structural scheme in the plurality of structural optimization indexes, determining the score difference coefficient corresponding to the structural optimization index to be screened; based on the score difference coefficient corresponding to each structural optimization index to be screened, determining a plurality of structural optimization indexes and the weight corresponding to each structural optimization index.

[0007] Further, based on the score difference coefficient corresponding to each structural optimization index to be screened, determining a plurality of structural optimization indexes and the weight corresponding to each structural optimization index includes: based on the score difference coefficient corresponding to each structural optimization index to be screened, determining a plurality of candidate structural optimization indexes; for any two candidate structural optimization indexes, according to the scores of each sample structural scheme in the two candidate structural optimization indexes, calculating the score correlation coefficient of the two candidate structural optimization indexes; based on the score correlation coefficient of the two candidate structural optimization indexes, establishing an index map; based on the index map, determining the node center coefficient of each candidate structural optimization index; based on the node center coefficient and the score difference coefficient of each candidate structural optimization index, determining a plurality of structural optimization indexes and the weight corresponding to each structural optimization index.

[0008] Further, based on the index graph, determine the node centrality coefficient of each candidate structural optimization index, including: based on the index graph, determine the initial node centrality coefficient of each candidate structural optimization index; for each candidate structural optimization index, based on the initial node centrality coefficient of the candidate structural optimization index and the initial node centrality coefficient of the first-order adjacent candidate structural optimization index, determine the node centrality coefficient of the candidate structural optimization index.

[0009] Further, based on the structural information of the suspension line 3dB bridge to be optimized, determine multiple structural parameters to be optimized, including: for each structural parameter to be screened for optimization and each structural optimization index, calculate the influence coefficient of the structural parameter to be screened for optimization on the structural optimization index based on the score of each sample structural scheme in the structural optimization index; based on the influence coefficient of each structural parameter to be screened for optimization on each structural optimization index, determine multiple structural parameters to be optimized.

[0010] Further, based on the influence coefficient of each structural parameter to be screened for optimization on each structural optimization index, determine multiple structural parameters to be optimized, including: for each structural parameter to be screened for optimization, determine the global influence coefficient of the structural parameter to be screened for optimization based on the influence coefficient of the structural parameter to be screened for optimization on each structural optimization index and the weight corresponding to each structural optimization index; based on the global influence coefficient of each structural parameter to be screened for optimization, determine multiple structural parameters to be optimized.

[0011] Further, based on multiple structural parameters to be optimized, generate multiple structural optimization schemes, and through an improved genetic algorithm and finite element analysis, based on multiple structural optimization schemes, determine the optimal structural optimization scheme, including: S11. Determine the parameter priority of each structural parameter to be optimized; S12. Based on the parameter priority of each structural parameter to be optimized, determine the current structural parameter to be optimized; S13. According to the constraint conditions of the current structural parameter to be optimized and the optimal values of the structural parameters that have been optimized, generate a structural optimization scheme corresponding to the current structural parameter to be optimized; S14. Through an improved genetic algorithm and finite element analysis, based on the structural optimization scheme corresponding to the current structural parameter to be optimized, determine the optimal value of the current structural parameter to be optimized, and mark the current structural parameter to be optimized as a structural parameter that has been optimized; S15. Determine whether all the structural parameters to be optimized have been optimized. If so, end the optimization. If not, execute S12.

[0012] Further, determine the parameter priority of each structural parameter to be optimized, including: based on the global influence coefficient of each structural parameter to be screened for optimization, determine the global influence coefficient of each structural parameter to be optimized; based on the global influence coefficient of each structural parameter to be optimized, determine the parameter priority of each structural parameter to be optimized.

[0013] Further, through an improved genetic algorithm and finite element analysis, based on the structural optimization scheme corresponding to the current structural parameters to be optimized, the optimal values of the current structural parameters to be optimized are determined, including: for each value segment corresponding to the current structural parameters to be optimized, sampling multiple structural optimization schemes corresponding to the value segment, through finite element analysis, calculating the fitness values of the multiple structural optimization schemes corresponding to the value segment, determining the fitness value characteristics corresponding to the value segment according to the fitness values of the multiple structural optimization schemes corresponding to the value segment; determining the priority value corresponding to each value segment according to the fitness value characteristics corresponding to each value segment; through finite element analysis, determining the fitness value of the structural optimization scheme corresponding to the current structural parameters to be optimized; and determining the optimal values of the current structural parameters to be optimized through the genetic algorithm based on the fitness value of the structural optimization scheme corresponding to the current structural parameters to be optimized and the priority value corresponding to each value segment.

[0014] The present invention provides a structural optimization system for a suspended line 3dB bridge, applying the above-mentioned structural optimization method for a suspended line 3dB bridge, including: an index determination module for determining a plurality of structural optimization indexes, wherein the plurality of structural optimization indexes are related to a plurality of physical fields; an information acquisition module for acquiring the structural information of the suspended line 3dB bridge to be optimized; a parameter determination module for determining a plurality of structural parameters to be optimized based on the structural information of the suspended line 3dB bridge to be optimized, wherein the structural parameters to be optimized include the structural parameters of the central double-sided coupled transmission line and the structural parameters of the side double-sided coupled transmission line; and a structural optimization module for generating a plurality of structural optimization schemes based on the plurality of structural parameters to be optimized, and determining the optimal structural optimization scheme based on the plurality of structural optimization schemes through an improved genetic algorithm and finite element analysis.

[0015] Compared with the prior art, the structural optimization method and system for a suspended line 3dB bridge provided by the present invention at least have the following beneficial effects: 1. In the prior art, adjusting all the structural parameters of the 3dB bridge one by one results in a huge amount of calculation and low efficiency. There is a lack of quantitative evaluation of the importance of parameters, which easily ignores key parameters or over-optimizes secondary parameters. The present invention analyzes key variables through structural information, avoids ineffective optimization of non-key parameters, and significantly reduces the computational complexity. Moreover, in the prior art, only a single physical field (such as the electromagnetic field) is optimized, ignoring the influence of other physical fields (such as the thermal field, mechanical stress) on the performance, resulting in the failure of the optimization result in practical applications due to the multi-physical field coupling effect. In the present invention, the structural optimization indexes are related to a plurality of physical fields (such as electromagnetic, thermal, mechanical), ensuring the comprehensive performance of the bridge in a complex environment. Through multi-physical field collaborative optimization, the performance bottleneck caused by single-physical field optimization is reduced.

[0016] 2. The score difference coefficient reflects the score fluctuation degree of the structure optimization index among different samples. The larger the score difference coefficient is, the stronger the discrimination ability of the index for different structure schemes, that is, the higher the "discrimination degree" of the index. By setting the first threshold, candidate indicators with larger score difference coefficients are screened out. These indicators are more likely to contain key factors that have a significant impact on structure optimization. Through the analysis of the score difference coefficient and the correlation coefficient, key indicators are automatically identified to avoid the deviation of manual subjective judgment. Not only considering the discrimination degree of the indicators, but also combining the correlation between the indicators to ensure that the selected indicators are both independent and representative. Through weight assignment, the indicators that have the greatest impact on the optimization goal are highlighted, reducing the interference of secondary indicators and improving the optimization efficiency. The indicator weights can be dynamically adjusted according to different optimization scenarios, with strong adaptability.

[0017] 3. In the prior art, relying on engineers' experience or the trial-and-error method, it is easy to ignore key parameters or overemphasize secondary parameters. Based on the integration of the non-linear correlation coefficient and the weight corresponding to the structure optimization index, the contribution of parameters to the optimization goal is quantified, reducing subjective judgment. In the prior art, all parameters need to be optimized one by one, with a large amount of calculation and easy to fall into local optimum. This method screens out the parameters that have the greatest impact on the optimization goal through the global influence coefficient, reducing the number of optimization variables and the computational complexity.

[0018] 4. Existing methods usually perform global search or uniform sampling on parameters without considering the inherent characteristic differences of parameter value segments. This method divides continuous parameters into discrete value segments, reducing the search dimension. Based on the fitness value characteristics of each value segment, priorities are dynamically assigned to guide the algorithm to preferentially explore high-potential regions. Through segmented focusing and priority guidance, inefficient calculations are significantly reduced, accelerating convergence. Traditional genetic algorithms only rely on fitness values for selection, crossover, and mutation, and are easy to fall into local optimum. This method simultaneously considers the fitness value at the scheme level and the priority value at the parameter segment level. The value segments with higher priority values are preferentially crossed and mutated, dynamically adjusting the search direction. Balancing local optimization and global exploration improves the robustness of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described by way of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where: Figure 1 is a schematic structural diagram of a suspended line 3dB bridge according to some embodiments of this specification; Figure 2 is a schematic flowchart of a structure optimization method for a suspended line 3dB bridge according to some embodiments of this specification; Figure 3 is a schematic flowchart of determining multiple structure optimization indicators according to some embodiments of this specification; Figure 4 It is a schematic diagram of an index spectrum shown in some embodiments of this specification; Figure 5 It is a schematic flowchart of determining an optimal structure optimization scheme shown in some embodiments of this specification; Figure 6 It is a schematic diagram of modules of a structure optimization system for a suspended line 3dB bridge shown in some embodiments of this specification.

[0020] In the figure, 1 is a dielectric substrate; 2 is a first microstrip line; 3 is a second microstrip line; 4 is a third microstrip line; 5 is a fourth microstrip line; 6 is a port; 7 is a double-sided coupled transmission line group; 71 is a central double-sided coupled transmission line; 72 is a side double-sided coupled transmission line. Detailed implementation manners

[0021] To more clearly illustrate the technical solutions of the embodiments of this specification, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, this specification can also be applied to other similar scenarios based on these drawings. Unless obvious from the language context or otherwise stated, the same reference numerals in the figures represent the same structures or operations.

[0022] Figure 1 It is a schematic diagram of the structure of a suspended line 3dB bridge shown in some embodiments of this specification. As Figure 1 shown, the suspended line 3dB bridge includes four ports 6, a dielectric substrate 1, and a first microstrip line 2, a second microstrip line 3, a third microstrip line 4, a fourth microstrip line 5, and a double-sided coupled transmission line group 7 provided on the dielectric substrate 1. The four ports 6 are respectively connected to the first microstrip line 2, the second microstrip line 3, the third microstrip line 4, and the fourth microstrip line 5. The first microstrip line 2 and the second microstrip line 3 are located on one side of the double-sided coupled transmission line group 7, and the third microstrip line 4 and the fourth microstrip line 5 are located on the other side of the double-sided coupled transmission line group 7. The double-sided coupled transmission line group 7 includes a central double-sided coupled transmission line 71 and at least two side double-sided coupled transmission lines 72. The at least two side double-sided coupled transmission lines 72 are symmetrically arranged around the central double-sided coupled transmission line 71, that is, the number of double-sided coupled transmission lines included in the double-sided coupled transmission line group 7 is an odd number. The microstrip line and the double-sided coupled transmission line group 7 achieve signal transmission through interlayer coupling. The four ports 6 are respectively connected to the first microstrip line 2, the second microstrip line 3, the third microstrip line 4, and the fourth microstrip line 5 through vias.

[0023] Among them, the dielectric substrate 1 includes two dielectric layers, and a metal copper layer is added between the two dielectric layers. For each double-sided coupled transmission line, an elliptical cutout is left in the metal copper layer to form the double-sided coupled transmission line. The surrounding is a sealed cavity, and the upper and lower sides are air. On this basis, through the vertical transition structure of metallized vias, that is, the metallized vias in the cutout, the microstrip lines on both sides of the dielectric substrate 1 are connected to realize the transition of the microstrip lines, so as to achieve the purpose of coplanarity of the microstrip lines.

[0024] Figure 2 is a schematic flow chart of a method for optimizing the structure of a suspended line 3dB bridge according to some embodiments of the present specification, as Figure 2 shown, a method for optimizing the structure of a suspended line 3dB bridge may include the following steps: Step 210, determine a plurality of structure optimization indexes.

[0025] Among them, the plurality of structure optimization indexes are related to a plurality of physical fields, and the plurality of physical fields may include multiple physical fields such as electromagnetic fields, thermal effects, and mechanical stresses.

[0026] Figure 3 is a schematic flow chart of determining a plurality of structure optimization indexes according to some embodiments of the present specification, as Figure 3 shown, preferably, step 210 specifically includes: Determine a plurality of structure optimization indexes to be screened and a plurality of structure parameters to be optimized to be screened; Based on the plurality of structure parameters to be optimized to be screened and the parameter constraint set, generate a variety of sample structure schemes, wherein the values of at least one structure parameter to be optimized to be screened in any two sample structure schemes are different; For each sample structure scheme, through finite element analysis, determine the score of the sample structure scheme in a plurality of structure optimization indexes; For each structure optimization index to be screened, based on the scores of each sample structure scheme in the plurality of structure optimization indexes, determine the score difference coefficient corresponding to the structure optimization index to be screened; Based on the score difference coefficient corresponding to each structure optimization index to be screened, determine a plurality of structure optimization indexes and the weight corresponding to each structure optimization index.

[0027] Specifically, the plurality of structure optimization indexes to be screened may include electromagnetic field-related indexes: 1. Bandwidth Definition: The frequency range in which the bridge satisfies 3dB power distribution within the target frequency band.

[0028] Optimization goal: By adjusting the width and spacing of the double-sided coupled transmission line, expand the bandwidth to the target range (such as 10%-20%).

[0029] 2. Isolation Definition: The degree of signal leakage between ports, usually expressed in dB (e.g., ≥20 dB).

[0030] Optimization objective: By adjusting the symmetry of the side double-sided coupled transmission lines, reduce crosstalk between ports.

[0031] 3. Insertion Loss Definition: The power loss when the signal passes through the bridge (e.g., ≤0.5 dB).

[0032] Optimization objective: By optimizing the impedance matching of the microstrip line, reduce conductor loss and dielectric loss.

[0033] 4. Amplitude / Phase Imbalance Definition: The amplitude or phase difference between the output ports (e.g., amplitude imbalance ≤0.3 dB, phase imbalance ≤3°).

[0034] Optimization objective: By adjusting the length and coupling coefficient of the double-sided coupled transmission lines, ensure symmetry.

[0035] Among multiple structural optimization indicators to be screened, thermal effect-related indicators can be included: 1. Thermal Stability Definition: The performance change of the bridge in a high-temperature environment (e.g., temperature coefficient ≤5 ppm / °C).

[0036] Optimization objective: Select low-loss dielectric materials and optimize the heat dissipation structure.

[0037] 2. Power Capacity Definition: The maximum power that the bridge can withstand (e.g., 10 W).

[0038] Optimization objective: By increasing the width or thickness of the microstrip line, reduce the current density.

[0039] Among multiple structural optimization indicators to be screened, mechanical stress-related indicators can be included: 1. Structural Reliability Definition: The failure probability of the bridge in a vibration or shock environment (e.g., the performance change after vibration test ≤5%).

[0040] Optimization objective: Optimize the fixing method of the double-sided coupled transmission lines to reduce stress concentration.

[0041] 2. Dimensions Definition: The physical size limit of the bridge (e.g., ≤20 mm × 10 mm).

[0042] Optimization objective: By means of compact layout and multi-layer design, meet the miniaturization requirements.

[0043] Among multiple structural optimization metrics to be screened, multi-physics field collaborative optimization metrics can be included: 1. Electromagnetic-thermal coupling Definition: The sensitivity of the electromagnetic performance of the bridge to temperature changes (e.g., when the temperature rises by 10°C, the bandwidth change ≤ 1%).

[0044] Optimization goal: Through thermal simulation analysis, adjust the microstrip line layout to reduce the hot spot temperature.

[0045] 2. Mechanical-electromagnetic coupling Definition: The influence of mechanical stress on the electromagnetic performance of the bridge (e.g., when vibrating, the isolation degradation ≤ 3 dB).

[0046] Optimization goal: Through finite element analysis, optimize the structure to reduce the influence of stress on the coupling coefficient.

[0047] Among multiple structural parameters of the structure to be optimized to be screened, structural parameters related to the microstrip line can be included. For example, the microstrip line widths, microstrip line lengths, microstrip line spacings, the interlayer coupling distances with the double-sided coupled transmission line group, the sizes of metallized vias, etc. of the first microstrip line, the second microstrip line, the third microstrip line, and the fourth microstrip line.

[0048] Such as Figure 1 As shown, both the central double-sided coupled transmission line and at least two side double-sided coupled transmission lines are elliptical. Among multiple structural parameters of the structure to be optimized to be screened, structural parameters related to the double-sided coupled transmission line group can be included. For example, the major axis lengths and minor axis lengths of the ellipses of the central double-sided coupled transmission line and the side double-sided coupled transmission lines, the spacing between the central double-sided coupled transmission line and the side double-sided coupled transmission lines, the spacing between two adjacent side double-sided coupled transmission lines, the major axis lengths and minor axis lengths of the elliptical slots.

[0049] Among multiple structural parameters of the structure to be optimized to be screened, structural parameters related to other components of the suspended line 3 dB bridge can also be included. For example, the dielectric layer thickness, dielectric material, sealed cavity size, etc.

[0050] The score of the sample structure scheme in multiple structural optimization metrics can be determined through finite element analysis according to the following process: S21. Establish a finite element model: Geometric modeling: Create a finite element mesh based on the 3D CAD model of the sample structure scheme.

[0051] Material definition: Set material parameters (such as dielectric constant, conductivity, elastic modulus, etc.).

[0052] Boundary conditions: Apply excitation sources (such as voltage, current) and constraints (such as fixed ends, matched loads).

[0053] Mesh Generation: Use hexahedral, tetrahedral or hybrid meshes to ensure the mesh density in critical regions (such as coupling regions).

[0054] S22. Solve the Physical Field Distribution: Electromagnetic Field Analysis: Calculate S-parameters to evaluate port reflection and transmission performance.

[0055] Extract field distributions (such as electric field intensity, current density) to analyze coupling efficiency.

[0056] Mechanical Analysis: Calculate stress, strain and deformation to evaluate mechanical reliability.

[0057] Thermal Analysis: Calculate temperature distribution and thermal stress to evaluate thermal stability.

[0058] S23. Extract Structure Optimization Metrics: Electromagnetic Field Performance Metrics: Bandwidth: Determine the 3dB bandwidth from the frequency response curve of S-parameters.

[0059] Insertion Loss: Calculate the amplitude attenuation of S21.

[0060] Isolation: Calculate the power ratio between the non-excited port and the excited port.

[0061] Power Capacity: Evaluate based on thermal analysis and material heat resistance.

[0062] Mechanical Performance Metrics: Maximum Stress: Compare with the material yield strength to evaluate safety.

[0063] Deformation: Evaluate mechanical stability.

[0064] Thermal Performance Metrics: Highest Temperature: Ensure it does not exceed the material operating temperature.

[0065] Thermal Stress: Evaluate the structural deformation caused by thermal expansion.

[0066] S24. Calculate the Score: Normalize each metric value to a unified scale (such as 0 - 1 or -1 to 1).

[0067] For example, Bandwidth Score = (Actual Bandwidth - Lower Limit of Target Bandwidth) / (Upper Limit of Target Bandwidth - Lower Limit of Target Bandwidth).

[0068] For example only, assume the optimization goals are: Performance Metrics: Bandwidth (≥5GHz), Insertion Loss (≤0.5dB), Isolation (≥25dB).

[0069] Mechanical index: maximum stress (≤100 MPa).

[0070] Thermal index: maximum temperature (≤120 °C).

[0071] Finite element analysis results of sample structure scheme A: Bandwidth = 5.2 GHz, insertion loss = 0.45 dB, isolation = 26 dB.

[0072] Maximum stress = 95 MPa, maximum temperature = 115 °C.

[0073] Scores of sample structure scheme A on multiple structure optimization indicators: Bandwidth score = (5.2 - 5) / (6 - 5) = 0.2; Insertion loss score = (0.5 - 0.45) / (0.5 - 0.3) = 0.25; Isolation score = (26 - 25) / (30 - 25) = 0.2; Maximum stress score = (100 - 95) / (100 - 80) = 0.25; Maximum temperature score = (120 - 115) / (120 - 100) = 0.25.

[0074] For each structure optimization indicator to be screened, the standard deviation of the scores of each sample structure scheme on multiple structure optimization indicators can be calculated as the score difference coefficient corresponding to the structure optimization indicator to be screened.

[0075] Preferably, based on the score difference coefficient corresponding to each structure optimization indicator to be screened, multiple structure optimization indicators and the weight corresponding to each structure optimization indicator are determined, including: Based on the score difference coefficient corresponding to each structure optimization indicator to be screened, multiple candidate structure optimization indicators are determined; for example, the structure optimization indicators to be screened with a score difference coefficient greater than the first threshold can be used as candidate structure optimization indicators; For any two candidate structure optimization indicators, according to the scores of each sample structure scheme on the two candidate structure optimization indicators, the score correlation coefficient of the two candidate structure optimization indicators is calculated. Specifically, according to the calculation formula of the non - linear correlation coefficient (such as the maximum information coefficient, distance correlation coefficient, mutual information, etc.), substituting the scores of each sample structure scheme on the two candidate structure optimization indicators, the score correlation coefficient of the two candidate structure optimization indicators is calculated; Based on the score correlation coefficient of the two candidate structure optimization indicators, an index map is established. Based on the index graph, determine the node centrality coefficient of each candidate structural optimization index; Based on the node centrality coefficient and the score difference coefficient of each candidate structural optimization index, determine multiple structural optimization indexes and the weight corresponding to each structural optimization index. For example, a candidate structural optimization index with a node centrality coefficient greater than a third threshold and a score difference coefficient greater than a fourth threshold can be used as a structural optimization index.

[0076] Specifically, Figure 4 is a schematic diagram of the index graph shown in some embodiments of this specification, as Figure 4 shown, the index graph may include nodes representing candidate structural optimization indexes. For any two candidate structural optimization indexes, when the score correlation coefficient between the two candidate structural optimization indexes is greater than a second threshold, it is determined that there is a correlation between the two candidate structural optimization indexes, and the nodes representing the two candidate structural optimization indexes can be connected by an edge. The greater the score correlation coefficient between the two candidate structural optimization indexes, the shorter the edge.

[0077] Preferably, based on the index graph, determining the node centrality coefficient of each candidate structural optimization index includes: Based on the index graph, determine the initial node centrality coefficient of each candidate structural optimization index; For each candidate structural optimization index, based on the initial node centrality coefficient of the candidate structural optimization index and the initial node centrality coefficient of the first-order adjacent candidate structural optimization indexes, determine the node centrality coefficient of the candidate structural optimization index, where the first-order adjacent candidate structural optimization indexes are the candidate structural optimization indexes directly connected by an edge.

[0078] Specifically, the initial node centrality coefficient of the candidate structural optimization index can be calculated according to the following formula: , where, is the initial node centrality coefficient of the i-th candidate structural optimization index, is the total number of edges of the node corresponding to the i-th candidate structural optimization index, is the total number of candidate structural optimization indexes, is the total number of edges of the node corresponding to the m-th candidate structural optimization index.

[0079] The node centrality coefficient of the candidate structural optimization index can be calculated according to the following formula: , where, is the node centrality coefficient of the i-th candidate structural optimization index, The correlation coefficient of the structural optimization index of the i-th candidate and the score of the structural optimization index of the g-th first-order adjacent candidate. The initial node center coefficient of the g-th first-order adjacent candidate of the structural optimization index of the i-th candidate. The total number of first-order adjacent candidates of the structural optimization index of the i-th candidate.

[0080] The weight corresponding to each structural optimization index can be calculated according to the following formula: , where, The weight corresponding to the i-th structural optimization index. The score difference coefficient of the i-th structural optimization index. The node center coefficient of the h-th structural optimization index. The score difference coefficient of the h-th structural optimization index. The total number of structural optimization indices.

[0081] It can be understood that the score difference coefficient reflects the score fluctuation degree of the structural optimization index among different samples. The larger the score difference coefficient, the stronger the discrimination ability of the index for different structural schemes, that is, the higher the "discrimination degree" of the index. By setting the first threshold, the candidate indices with larger score difference coefficients are screened out. These indices are more likely to contain key factors that have a significant impact on structural optimization. Through the analysis of the score difference coefficient and the correlation coefficient, key indices are automatically identified, avoiding the deviation of manual subjective judgment. Not only considering the discrimination degree of the indices, but also combining the correlation between the indices to ensure that the screened indices are both independent and representative. Through weight assignment, the indices that have the greatest impact on the optimization goal are highlighted, reducing the interference of secondary indices and improving the optimization efficiency. The index weights can be dynamically adjusted according to different optimization scenarios, with strong adaptability.

[0082] Step 220, obtain the structural information of the to-be-optimized suspended line 3dB bridge.

[0083] Step 230, based on the structural information of the to-be-optimized suspended line 3dB bridge, determine multiple to-be-optimized structural parameters.

[0084] Among them, the to-be-optimized structural parameters include the structural parameters of the central double-sided coupled transmission line and the structural parameters of the side double-sided coupled transmission line.

[0085] Preferably, step 230 specifically includes: For each structural parameter to be screened for optimization and each structural optimization index, based on the scores of each sample structural scheme in the structural optimization index, calculate the influence coefficient of the structural parameter to be screened for optimization on the structural optimization index. Specifically, according to the calculation formula of the non - linear correlation coefficient (such as the maximum information coefficient, distance correlation coefficient, mutual information, etc.), substitute the parameter values of the structural parameter to be screened for optimization in each sample structural scheme and the scores of each sample structural scheme in the structural optimization index to calculate the influence coefficient of the structural parameter to be screened for optimization on the structural optimization index; Based on the influence coefficient of each structural parameter to be screened for optimization on each structural optimization index, determine multiple structural parameters to be optimized.

[0086] Preferably, based on the influence coefficient of each structural parameter to be screened for optimization on each structural optimization index, determine multiple structural parameters to be optimized, including: For each structural parameter to be screened for optimization, based on the influence coefficient of the structural parameter to be screened for optimization on each structural optimization index and the weight corresponding to each structural optimization index, determine the global influence coefficient of the structural parameter to be screened for optimization; Based on the global influence coefficient of each structural parameter to be screened for optimization, determine multiple structural parameters to be optimized. For example, the structural parameters to be screened for optimization with a global influence coefficient greater than the fifth threshold can be used as the structural parameters to be optimized.

[0087] Specifically, the global influence coefficient of the structural parameter to be screened for optimization can be calculated according to the following formula: , where, is the global influence coefficient of the j - th structural parameter to be screened for optimization, is the influence coefficient of the j - th structural parameter to be screened for optimization on the h - th structural optimization index.

[0088] It can be understood that in the prior art, relying on engineers' experience or the trial - and - error method, it is easy to ignore key parameters or over - focus on secondary parameters. Based on the integration of the non - linear correlation coefficient and the weight corresponding to the structural optimization index, the contribution of parameters to the optimization goal is quantified, reducing subjective judgment. In the prior art, all parameters need to be optimized one by one, with a large amount of calculation and easy to fall into local optimality. This method screens out the parameters with the greatest influence on the optimization goal through the global influence coefficient, reduces the number of optimization variables, and reduces the computational complexity.

[0089] Step 240, based on multiple structural parameters to be optimized, generate multiple structural optimization schemes, and through the improved genetic algorithm and finite - element analysis, based on multiple structural optimization schemes, determine the optimal structural optimization scheme.

[0090] Figure 5It is a schematic flow chart for determining the optimal structure optimization solution shown in some embodiments of this specification. As shown in Figure 5 Preferably, step 240 specifically includes: S11. Determine the parameter priority of each structural parameter to be optimized; S12. Based on the parameter priority of each structural parameter to be optimized, determine the current structural parameter to be optimized. For example, take the structural parameter to be optimized with the highest parameter priority and not yet completed optimization as the current structural parameter to be optimized; S13. According to the constraint conditions of the current structural parameter to be optimized and the optimal values of the structural parameters that have been optimized, generate a structural optimization solution corresponding to the current structural parameter to be optimized; S14. Through an improved genetic algorithm and finite element analysis, based on the structural optimization solution corresponding to the current structural parameter to be optimized, determine the optimal value of the current structural parameter to be optimized, and mark the current structural parameter to be optimized as a structural parameter that has been optimized; S15. Determine whether all the structural parameters to be optimized have been optimized. If so, end the optimization. If not, execute S12.

[0091] Preferably, determining the parameter priority of each structural parameter to be optimized includes: Based on the global influence coefficient of each structural parameter to be screened, determine the global influence coefficient of each structural parameter to be optimized. For example, if the global influence coefficient of the structural parameter to be screened A is 0.5, then the corresponding global influence coefficient of the structural parameter to be optimized A is 0.5; Based on the global influence coefficient of each structural parameter to be optimized, determine the parameter priority of each structural parameter to be optimized, where the greater the global influence coefficient, the higher the parameter priority of the structural parameter to be optimized.

[0092] Preferably, through an improved genetic algorithm and finite element analysis, based on the structural optimization solution corresponding to the current structural parameter to be optimized, determining the optimal value of the current structural parameter to be optimized includes: For each value segment corresponding to the current structural parameter to be optimized, sample multiple structural optimization solutions corresponding to the value segment. Through finite element analysis, calculate the fitness values of the multiple structural optimization solutions corresponding to the value segment. According to the fitness values of the multiple structural optimization solutions corresponding to the value segment, determine the fitness value characteristics corresponding to the value segment, where the fitness value characteristics corresponding to the value segment may include the mean fitness value and the standard deviation of the fitness value; According to the fitness value characteristics corresponding to each value segment, determine the priority value corresponding to each value segment; Through finite element analysis, determine the fitness value of the structural optimization solution corresponding to the current structural parameter to be optimized; Based on the fitness value of the structural optimization scheme corresponding to the current structural parameters to be optimized and the priority value corresponding to each value segment, the optimal value of the current structural parameters to be optimized is determined by a genetic algorithm.

[0093] For example, if the value range corresponding to the current structural parameters to be optimized is [10, 50], it can be divided into 5 value segments: [10 - 20), [20 - 30), [30 - 40), [40 - 50], or a finer-grained division. Multiple candidate values are generated within each value segment through sampling methods (such as uniform sampling, Latin hypercube sampling) to form multiple structural optimization schemes. For example, 5 values are extracted within the segment [20 - 30]: 22, 24, 26, 28, 30.

[0094] The priority value corresponding to the value segment can be calculated according to the following formula: , where, is the priority value corresponding to the i-th value segment, is the mean fitness value corresponding to the i-th value segment, is the standard deviation of the fitness value corresponding to the i-th value segment, is a constant to avoid division by zero.

[0095] The optimal value of the current structural parameters to be optimized can be determined according to the following process: Combining the global fitness value and the value segment priority value, the optimal value of the structural parameters to be optimized is determined by an improved genetic algorithm. The specific steps include: S31. Population initialization: Randomly generate an initial population, and each individual contains the values of all parameters to be optimized.

[0096] S32. Fitness evaluation: Perform FEA on each individual and calculate its fitness value.

[0097] S33. Selection operation: According to the fitness value and the priority value, select excellent individuals to enter the next generation. For example: Individuals within the value segment with a higher priority value have a higher probability of being selected; individuals with a higher fitness value are preferentially retained.

[0098] S34. Crossover and mutation: Perform crossover (recombination) and mutation (random perturbation) on the selected individuals to generate new individuals. Among them, mutation is preferentially performed within the value segment with a higher priority value. For example, if the priority value of the priority value segment [20 - 30] is 0.8, while the priority value of [40 - 50] is 0.2, then the mutation operation is more likely to perturb the parameters within the segment [20 - 30].

[0099] S35. Update priority values: Dynamically adjust the priority values corresponding to each value segment according to the fitness distribution of each generation of population. For example, if the fitness of individuals within a certain value segment remains high, increase its priority value; if there are no excellent individuals in a certain value segment for a long time, decrease its priority value. S36. Iterative update: Repeat the above steps until the convergence condition is met (such as the change in fitness value is less than the threshold).

[0100] It can be understood that existing methods usually perform global search or uniform sampling on parameters without considering the inherent characteristic differences of parameter value segments. This method divides continuous parameters into discrete value segments, reducing the search dimension. Dynamically allocate priorities based on the fitness value characteristics of each value segment, guiding the algorithm to preferentially explore high-potential regions. Through segmented focusing and priority guidance, significantly reduce inefficient calculations and accelerate convergence. Traditional genetic algorithms only rely on fitness values for selection, crossover, and mutation, and are prone to falling into local optima. This method simultaneously considers the fitness value at the solution level and the priority value at the parameter segment level. The value segments with high priority values are preferentially crossed and mutated, dynamically adjusting the search direction. Balancing local optimization and global exploration improves the robustness of the algorithm.

[0101] Figure 6 is a schematic diagram of the modules of a structure optimization system for a suspended line 3dB bridge according to some embodiments of this specification, as Figure 6 shown, a structure optimization system for a suspended line 3dB bridge may include an index determination module, an information acquisition module, a parameter determination module, and a structure optimization module.

[0102] The index determination module is used to determine multiple structure optimization indexes, where the multiple structure optimization indexes are related to multiple physical fields; The information acquisition module is used to acquire the structure information of the suspended line 3dB bridge to be optimized; The parameter determination module is used to determine multiple structure parameters to be optimized based on the structure information of the suspended line 3dB bridge to be optimized, where the structure parameters to be optimized include the structure parameters of the central double-sided coupled transmission line and the structure parameters of the side double-sided coupled transmission line; The structure optimization module is used to generate multiple structure optimization schemes based on the multiple structure parameters to be optimized, and determine the optimal structure optimization scheme based on the multiple structure optimization schemes through an improved genetic algorithm and finite element analysis.

[0103] A structure optimization system for a suspended line 3dB bridge may apply the above-mentioned structure optimization method for a suspended line 3dB bridge, which will not be elaborated here.

[0104] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be regarded as consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly presented and described in this specification.

Claims

1. A structural optimization method for a suspended line 3dB bridge, characterized in that Applied to a suspended line 3dB bridge, wherein the suspended line 3dB bridge includes four ports, a dielectric substrate, and a first microstrip line, a second microstrip line, a third microstrip line, a fourth microstrip line, and a double-sided coupled transmission line group disposed on the dielectric substrate. The four ports are respectively connected to the first microstrip line, the second microstrip line, the third microstrip line, and the fourth microstrip line. The first microstrip line and the second microstrip line are located on one side of the double-sided coupled transmission line group, and the third microstrip line and the fourth microstrip line are located on the other side of the double-sided coupled transmission line group. The double-sided coupled transmission line group includes a central double-sided coupled transmission line and at least two side double-sided coupled transmission lines, and the at least two side double-sided coupled transmission lines are symmetrically disposed around the central double-sided coupled transmission line; The method includes: Determining a plurality of structure optimization indexes, wherein the plurality of structure optimization indexes are related to a plurality of physical fields; Obtaining the structure information of the suspended line 3dB bridge to be optimized; Based on the structure information of the suspended line 3dB bridge to be optimized, determining a plurality of structure parameters to be optimized, wherein the structure parameters to be optimized include the structure parameters of the central double-sided coupled transmission line and the structure parameters of the side double-sided coupled transmission line; Based on the plurality of structure parameters to be optimized, generating a variety of structure optimization schemes, and through an improved genetic algorithm and finite element analysis, determining the optimal structure optimization scheme based on the variety of structure optimization schemes.

2. The structural optimization method for a suspended line 3dB bridge according to claim 1, characterized in that, Determining a plurality of structure optimization indexes, including: Determining a plurality of structure optimization indexes to be screened and a plurality of structure parameters to be optimized to be screened; Based on the plurality of structure parameters to be optimized to be screened and the parameter constraint set, generating a variety of sample structure schemes; For each sample structure scheme, through finite element analysis, determining the scores of the sample structure scheme in a plurality of structure optimization indexes; For each structure optimization index to be screened, based on the scores of each sample structure scheme in the plurality of structure optimization indexes, determining the score difference coefficient corresponding to the structure optimization index to be screened; Based on the score difference coefficient corresponding to each structure optimization index to be screened, determining a plurality of structure optimization indexes and the weight corresponding to each structure optimization index.

3. The structural optimization method for a suspended line 3dB bridge according to claim 2, characterized in that, Based on the score difference coefficient corresponding to each structure optimization index to be screened, determining a plurality of structure optimization indexes and the weight corresponding to each structure optimization index, including: Based on the score difference coefficient corresponding to each structure optimization index to be screened, determining a plurality of candidate structure optimization indexes; For any two candidate structure optimization indexes, according to the scores of each sample structure scheme in the two candidate structure optimization indexes, calculating the score correlation coefficient of the two candidate structure optimization indexes; Based on the score correlation coefficient of the two candidate structure optimization indexes, establishing an index map; Based on the index map, determining the node center coefficient of each candidate structure optimization index; Based on the node center coefficient and the score difference coefficient of each candidate structure optimization index, determining a plurality of structure optimization indexes and the weight corresponding to each structure optimization index.

4. A structural optimization method for a suspended line 3dB bridge according to claim 3, characterized in that Based on the index map, determining the node center coefficient of each candidate structure optimization index, including: Based on the index map, determining the initial node center coefficient of each candidate structure optimization index; For each candidate structural optimization index, determine the node center coefficient of the candidate structural optimization index based on the initial node center coefficient of the candidate structural optimization index and the initial node center coefficient of the first-order adjacent candidate structural optimization index.

5. The structural optimization method for a suspended line 3dB bridge according to claim 4, wherein Based on the structural information of the to-be-optimized suspension line 3dB bridge, determine multiple to-be-optimized structural parameters, including: For each to-be-screened to-be-optimized structural parameter and each structural optimization index, calculate the influence coefficient of the to-be-screened to-be-optimized structural parameter on the structural optimization index based on the scores of each sample structural scheme on the structural optimization index; Based on the influence coefficient of each to-be-screened to-be-optimized structural parameter on each structural optimization index, determine multiple to-be-optimized structural parameters.

6. The structural optimization method for a suspended line 3dB bridge according to claim 5, characterized in that Based on the influence coefficient of each to-be-screened to-be-optimized structural parameter on each structural optimization index, determine multiple to-be-optimized structural parameters, including: For each to-be-screened to-be-optimized structural parameter, determine the global influence coefficient of the to-be-screened to-be-optimized structural parameter based on the influence coefficient of the to-be-screened to-be-optimized structural parameter on each structural optimization index and the weight corresponding to each structural optimization index; Based on the global influence coefficient of each to-be-screened to-be-optimized structural parameter, determine multiple to-be-optimized structural parameters.

7. A structural optimization method for a suspended line 3dB bridge according to claim 6, characterized in that, Based on multiple to-be-optimized structural parameters, generate multiple structural optimization schemes, and through an improved genetic algorithm and finite element analysis, based on multiple structural optimization schemes, determine the optimal structural optimization scheme, including: S11. Determine the parameter priority of each to-be-optimized structural parameter; S12. Based on the parameter priority of each to-be-optimized structural parameter, determine the current to-be-optimized structural parameter; S13. According to the constraint conditions of the current to-be-optimized structural parameter and the optimal values of the structural parameters that have been optimized, generate the structural optimization scheme corresponding to the current to-be-optimized structural parameter; S14. Through an improved genetic algorithm and finite element analysis, based on the structural optimization scheme corresponding to the current to-be-optimized structural parameter, determine the optimal value of the current to-be-optimized structural parameter, and mark the current to-be-optimized structural parameter as the optimized structural parameter; S15. Determine whether all to-be-optimized structural parameters have been optimized. If so, end the optimization. If not, execute S12.

8. A structural optimization method for a suspended line 3dB bridge according to claim 7, characterized in that Determine the parameter priority of each to-be-optimized structural parameter, including: Based on the global influence coefficient of each to-be-screened to-be-optimized structural parameter, determine the global influence coefficient of each to-be-optimized structural parameter; Based on the global influence coefficient of each to-be-optimized structural parameter, determine the parameter priority of each to-be-optimized structural parameter.

9. A structural optimization method for a suspended line 3dB bridge according to claim 7, characterized in that Through an improved genetic algorithm and finite element analysis, based on the structural optimization scheme corresponding to the current to-be-optimized structural parameter, determine the optimal value of the current to-be-optimized structural parameter, including: For each value segment corresponding to the current to-be-optimized structural parameter, sample multiple structural optimization schemes corresponding to the value segment, calculate the fitness values of the multiple structural optimization schemes corresponding to the value segment through finite element analysis, and determine the fitness value characteristics corresponding to the value segment according to the fitness values of the multiple structural optimization schemes corresponding to the value segment; According to the fitness value characteristics corresponding to each value segment, determine the priority value corresponding to each value segment; Determine the fitness value of the structural optimization scheme corresponding to the current structural parameters to be optimized through finite element analysis; Determine the optimal value of the current structural parameters to be optimized through a genetic algorithm based on the fitness value of the structural optimization scheme corresponding to the current structural parameters to be optimized and the priority value corresponding to each value segment.

10. An optimization system for the structure of a suspended line 3dB bridge, characterized in that, Apply a structural optimization method for a suspended line 3dB bridge according to any one of claims 1-9, including: An index determination module for determining a plurality of structural optimization indexes, wherein the plurality of structural optimization indexes are related to a plurality of physical fields; An information acquisition module for acquiring the structural information of the suspended line 3dB bridge to be optimized; A parameter determination module for determining a plurality of structural parameters to be optimized based on the structural information of the suspended line 3dB bridge to be optimized, wherein the structural parameters to be optimized include the structural parameters of the central double-sided coupled transmission line and the structural parameters of the side double-sided coupled transmission line; A structural optimization module for generating a plurality of structural optimization schemes based on the plurality of structural parameters to be optimized, and determining the optimal structural optimization scheme based on the plurality of structural optimization schemes through an improved genetic algorithm and finite element analysis.

Images