A Structural Optimization Method and System for Suspension Line 3dB Bridges

Abstract

The present invention provides a method and system for optimizing the structure of a suspended line 3dB bridge, which relates to the field of radio frequency devices and is applied to a suspended line 3dB bridge. 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; determining a plurality of structure parameters to be optimized based on the structure information of the suspended line 3dB bridge 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; generating a plurality of structure optimization schemes based on the plurality of structure parameters to be optimized, and determining the optimal structure optimization scheme based on the plurality of structure optimization schemes through an improved genetic algorithm and finite element analysis, having the advantages of optimizing the structure of the suspended line 3dB bridge and improving the stability of the suspended line 3dB bridge.

Description

Technical Field

[0001] The present invention relates to the field of radio frequency devices, and particularly to a method and system for optimizing the structure of 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 in which a conductor strip line is suspended above a dielectric substrate and fixed by support posts or air bridges. Compared with traditional microstrip lines or strip lines, due to the 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 relies on experience or trial-and-error methods, 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 method and system for optimizing the structure of 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 arranged 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 indicators, where the plurality of structural optimization indicators 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 indicators includes: determining a plurality of structural optimization indicators 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 a plurality of structural optimization indicators; for each structural optimization indicator to be screened, based on the scores of each sample structural scheme in the plurality of structural optimization indicators, determining the score difference coefficient corresponding to the structural optimization indicator to be screened; based on the score difference coefficient corresponding to each structural optimization indicator to be screened, determining a plurality of structural optimization indicators and the weight corresponding to each structural optimization indicator.

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

[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 indices, determine the node centrality coefficient of the candidate structural optimization index.

[0009] Further, 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 score of each sample structural scheme in 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.

[0010] Further, 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.

[0011] Further, 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 already-optimized structural parameters, 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 an already-optimized structural parameter; S15. Determine whether all the to-be-optimized structural parameters have been optimized. If so, end the optimization. If not, execute S12.

[0012] Further, 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.

[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, calculating the fitness values of the multiple structural optimization schemes corresponding to the value segment through finite element analysis, 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; determining the fitness value of the structural optimization scheme corresponding to the current structural parameters to be optimized through finite element analysis; 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 indices, wherein the plurality of structural optimization indices 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:

[0016] 1. In the prior art, all the structural parameters of the 3dB bridge are adjusted one by one, resulting 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 indices 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.

[0017] 2. The score difference coefficient reflects the score fluctuation degree of the structural optimization index among different samples. The larger the score difference coefficient is, the stronger the discrimination ability of the index for different structural schemes is, that is, the higher the "discrimination degree" of the index is. By setting the first threshold, candidate indexes with larger score difference coefficients are screened out. These indexes 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 indexes are automatically identified to avoid the deviation of manual subjective judgment. Not only the discrimination degree of the index is considered, but also the correlation between indexes is combined to ensure that the screened indexes are both independent and representative. Through weight assignment, the indexes with the greatest impact on the optimization goal are highlighted, the interference of secondary indexes is reduced, and the optimization efficiency is improved. The index weights can be dynamically adjusted according to different optimization scenarios, with strong adaptability.

[0018] 3. 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 optimum. This method screens out the parameters with the greatest impact on the optimization goal through the global influence coefficient, reduces the number of optimization variables, and lowers the computational complexity.

[0019] 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 areas. 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. It balances local optimization and global exploration, enhancing the robustness of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] This specification will be further described in the form of exemplary embodiments, and these exemplary embodiments will be described in detail through the drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0021] Figure 1 is a schematic structural diagram of a suspended line 3dB bridge according to some embodiments of this specification;

[0022] Figure 2 is a schematic flowchart of a structural optimization method for a suspended line 3dB bridge according to some embodiments of this specification;

[0023] Figure 3 It is a schematic flow chart for determining multiple structure optimization indicators shown in some embodiments of this specification;

[0024] Figure 4 It is a schematic diagram of an index map shown in some embodiments of this specification;

[0025] Figure 5 It is a schematic flow chart for determining an optimal structure optimization scheme shown in some embodiments of this specification;

[0026] 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.

[0027] 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

[0028] 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 structure or operation.

[0029] 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 1As shown in the figure, the suspended line 3dB bridge includes four ports 6, a dielectric substrate 1, 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 disposed 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 disposed 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 signal transmission between the microstrip lines and the double-sided coupled transmission line group 7 is realized 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.

[0030] Among them, the dielectric substrate 1 includes two dielectric layers, and a copper layer is added between the two dielectric layers. For each double-sided coupled transmission line, an oval groove is left in the 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 the metallized vias, that is, the metallized vias in the middle of the groove, 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.

[0031] 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 shown in Figure 2 As shown in the figure, a method for optimizing the structure of a suspended line 3dB bridge may include the following steps:

[0032] Step 210, determining a plurality of structure optimization indicators.

[0033] Among them, the plurality of structure optimization indicators 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.

[0034] Figure 3 is a schematic flow chart of determining a plurality of structure optimization indicators according to some embodiments of the present specification, as shown in Figure 3 As shown in the figure, preferably, step 210 specifically includes:

[0035] Determining a plurality of structure optimization indicators to be screened and a plurality of structure parameters to be optimized to be screened;

[0036] Based on multiple structural parameters to be screened and parameter constraint sets for optimization, multiple sample structure schemes are generated, where the values of at least one structural parameter to be screened and optimized are different between any two sample structure schemes.

[0037] For each sample structure scheme, through finite element analysis, the scores of the sample structure scheme in multiple structural optimization metrics are determined.

[0038] For each structural optimization metric to be screened, based on the scores of each sample structure scheme in multiple structural optimization metrics, the score difference coefficient corresponding to the structural optimization metric to be screened is determined.

[0039] Based on the score difference coefficient corresponding to each structural optimization metric to be screened, multiple structural optimization metrics and the weight corresponding to each structural optimization metric are determined.

[0040] Specifically, the multiple structural optimization metrics to be screened may include electromagnetic field related metrics:

[0041] 1. Bandwidth

[0042] Definition: The frequency range in which the bridge satisfies 3dB power distribution within the target frequency band.

[0043] 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%).

[0044] 2. Isolation

[0045] Definition: The degree of signal leakage between ports, usually expressed in dB (such as ≥20dB).

[0046] Optimization goal: By adjusting the symmetry of the side double-sided coupled transmission line, reduce the crosstalk between ports.

[0047] 3. Insertion loss

[0048] Definition: The power loss when the signal passes through the bridge (such as ≤0.5dB).

[0049] Optimization goal: By optimizing the microstrip line impedance matching, reduce the conductor loss and dielectric loss.

[0050] 4. Amplitude / phase imbalance

[0051] Definition: The amplitude or phase difference between the output ports (such as amplitude imbalance ≤0.3dB, phase imbalance ≤3°).

[0052] Optimization goal: By adjusting the length and coupling coefficient of the double-sided coupled transmission line, ensure symmetry.

[0053] The multiple structural optimization metrics to be screened may include thermal effect related metrics:

[0054] 1. Thermal stability

[0055] Definition: The performance change of the bridge in a high - temperature environment (e.g., temperature coefficient ≤ 5 ppm / ℃).

[0056] Optimization goal: Select low - loss dielectric materials and optimize the heat - dissipation structure.

[0057] 2. Power capacity

[0058] Definition: The maximum power that the bridge can withstand (e.g., 10 W).

[0059] Optimization goal: Reduce the current density by increasing the width or thickness of the microstrip line.

[0060] Among multiple structural optimization indicators to be screened, mechanical stress - related indicators can be included:

[0061] 1. Structural reliability

[0062] Definition: The failure probability of the bridge in a vibration or shock environment (e.g., the performance change ≤ 5% after vibration testing).

[0063] Optimization goal: Optimize the fixing method of the double - sided coupled transmission line to reduce stress concentration.

[0064] 2. Size

[0065] Definition: The physical size limit of the bridge (e.g., ≤ 20 mm × 10 mm).

[0066] Optimization goal: Meet the miniaturization requirements through compact layout and multi - layer design.

[0067] Among multiple structural optimization indicators to be screened, multi - physical - field co - optimization indicators can be included:

[0068] 1. Electromagnetic - thermal coupling

[0069] Definition: The sensitivity of the electromagnetic performance of the bridge to temperature change (e.g., when the temperature rises by 10℃, the bandwidth change ≤ 1%).

[0070] Optimization goal: Through thermal simulation analysis, adjust the microstrip line layout to reduce the hot - spot temperature.

[0071] 2. Mechanical - electromagnetic coupling

[0072] Definition: The influence of mechanical stress on the electromagnetic performance of the bridge (e.g., the isolation degradation ≤ 3 dB after vibration).

[0073] Optimization goal: Through finite - element analysis, optimize the structure to reduce the influence of stress on the coupling coefficient.

[0074] Multiple structural parameters to be optimized and screened can include structural parameters related to microstrip lines. For example, the microstrip line widths, lengths, spacings of the first, second, third, and fourth microstrip lines, the interlayer coupling distance of the double-sided coupled transmission line group, the sizes of metallized vias, etc.

[0075] As Figure 1 shown, both the central double-sided coupled transmission line and at least two side double-sided coupled transmission lines are elliptical. Multiple structural parameters to be optimized and screened can include structural parameters related to the double-sided coupled transmission line group. 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 length and minor axis length of the elliptical cutout.

[0076] Multiple structural parameters to be optimized and screened can also include structural parameters related to other components of the suspended line 3dB bridge. For example, the dielectric layer thickness, dielectric material, sealed cavity size, etc.

[0077] The scores of the sample structure scheme in multiple structural optimization indicators can be determined through finite element analysis according to the following process:

[0078] S21. Establish a finite element model:

[0079] Geometric modeling: Create a finite element mesh based on the 3D CAD model of the sample structure scheme.

[0080] Material definition: Set material parameters (such as dielectric constant, conductivity, elastic modulus, etc.).

[0081] Boundary conditions: Apply excitation sources (such as voltage, current) and constraints (such as fixed ends, matched loads).

[0082] Mesh division: Use hexahedron, tetrahedron or hybrid meshes to ensure the mesh density in key areas (such as the coupling area).

[0083] S22. Solve the physical field distribution:

[0084] Electromagnetic field analysis:

[0085] Calculate S parameters to evaluate port reflection and transmission performance.

[0086] Extract the field distribution (such as electric field intensity, current density) to analyze the coupling efficiency.

[0087] Mechanical analysis:

[0088] Calculate stress, strain and deformation to evaluate mechanical reliability.

[0089] Thermal analysis:

[0090] Calculate the temperature distribution and thermal stress, and evaluate the thermal stability.

[0091] S23. Extract the structure optimization indicators:

[0092] Electromagnetic field performance indicators:

[0093] Bandwidth: Determine the 3dB bandwidth through the frequency response curve of the S-parameters.

[0094] Insertion loss: Calculate the amplitude attenuation of S21.

[0095] Isolation: Calculate the power ratio between the non-excited port and the excited port.

[0096] Power capacity: Evaluate based on thermal analysis and material heat resistance.

[0097] Mechanical performance indicators:

[0098] Maximum stress: Compare with the material yield strength to evaluate safety.

[0099] Deformation amount: Evaluate mechanical stability.

[0100] Thermal performance indicators:

[0101] Highest temperature: Ensure that it does not exceed the material operating temperature.

[0102] Thermal stress: Evaluate the structural deformation caused by thermal expansion.

[0103] S24. Calculate the score:

[0104] Normalize each indicator value to a unified scale (such as 0 - 1 or -1 to 1).

[0105] For example, bandwidth score = (actual bandwidth - lower limit of target bandwidth) / (upper limit of target bandwidth - lower limit of target bandwidth).

[0106] Only for example, assume the optimization goals are:

[0107] Performance indicators: Bandwidth (≥5GHz), insertion loss (≤0.5dB), isolation (≥25dB).

[0108] Mechanical indicators: Maximum stress (≤100MPa).

[0109] Thermal indicators: Highest temperature (≤120°C).

[0110] Finite element analysis results of sample structure scheme A:

[0111] Bandwidth = 5.2GHz, insertion loss = 0.45dB, isolation = 26dB.

[0112] Maximum stress = 95 MPa, maximum temperature = 115 °C.

[0113] Scores of Sample Structure Scheme A in multiple structural optimization indicators:

[0114] Bandwidth score = (5.2 - 5) / (6 - 5) = 0.2;

[0115] Insertion loss score = (0.5 - 0.45) / (0.5 - 0.3) = 0.25;

[0116] Isolation score = (26 - 25) / (30 - 25) = 0.2;

[0117] Maximum stress score = (100 - 95) / (100 - 80) = 0.25;

[0118] Maximum temperature score = (120 - 115) / (120 - 100) = 0.25.

[0119] For each structural optimization indicator to be screened, the standard deviation of the scores of each sample structure scheme in multiple structural optimization indicators can be calculated as the score difference coefficient corresponding to the structural optimization indicator to be screened.

[0120] Preferably, based on the score difference coefficient corresponding to each structural optimization indicator to be screened, multiple structural optimization indicators and the weight corresponding to each structural optimization indicator are determined, including:

[0121] Based on the score difference coefficient corresponding to each structural optimization indicator to be screened, multiple candidate structural optimization indicators are determined; for example, the structural optimization indicators to be screened with a score difference coefficient greater than the first threshold can be used as candidate structural optimization indicators;

[0122] For any two candidate structural optimization indicators, according to the scores of each sample structure scheme in the two candidate structural optimization indicators, the score correlation coefficient of the two candidate structural 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 in the two candidate structural optimization indicators, the score correlation coefficient of the two candidate structural optimization indicators is calculated;

[0123] Based on the score correlation coefficient of the two candidate structural optimization indicators, an index map is established;

[0124] Based on the index map, the node central coefficient of each candidate structural optimization indicator is determined;

[0125] Based on the node centrality coefficient and score difference coefficient of each candidate's structural optimization index, determine multiple structural optimization indices and the weights corresponding to each structural optimization index. For example, a candidate's 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.

[0126] Specifically, Figure 4 is a schematic diagram of an index map according to some embodiments of this specification. As Figure 4 shown, the index map may include nodes representing candidates' structural optimization indices. For any two candidates' structural optimization indices, when the score correlation coefficient between the two candidates' structural optimization indices is greater than a second threshold, it is determined that there is a correlation between the two candidates' structural optimization indices, and the nodes representing the two candidates' structural optimization indices can be connected by an edge. The greater the score correlation coefficient between the two candidates' structural optimization indices, the shorter the edge.

[0127] Preferably, based on the index map, determining the node centrality coefficient of each candidate's structural optimization index includes:

[0128] Based on the index map, determine the initial node centrality coefficient of each candidate's structural optimization index;

[0129] For each candidate's structural optimization index, based on the initial node centrality coefficient of the candidate's structural optimization index and the initial node centrality coefficients of the first-order adjacent candidates' structural optimization indices, determine the node centrality coefficient of the candidate's structural optimization index, where the first-order adjacent candidates' structural optimization indices are the candidates' structural optimization indices directly connected by an edge.

[0130] Specifically, the initial node centrality coefficient of a candidate's structural optimization index can be calculated according to the following formula:

[0131] ,

[0132] where, is the initial node centrality coefficient of the i-th candidate's structural optimization index, is the total number of edges of the node corresponding to the i-th candidate's structural optimization index, is the total number of candidates' structural optimization indices, is the total number of edges of the node corresponding to the m-th candidate's structural optimization index.

[0133] The node centrality coefficient of a candidate's structural optimization index can be calculated according to the following formula:

[0134] ,

[0135] where, is the node center coefficient of the structural optimization index for the i-th candidate. is the score correlation coefficient between the structural optimization index of the i-th candidate and the structural optimization index of the g-th first-order adjacent candidate. is the initial node center coefficient of the g-th first-order adjacent candidate's structural optimization index for the i-th candidate's structural optimization index. is the total number of first-order adjacent candidate's structural optimization indexes for the i-th candidate's structural optimization index.

[0136] The weight corresponding to each structural optimization index can be calculated according to the following formula:

[0137] ,

[0138] where, is the weight corresponding to the i-th structural optimization index, is the score difference coefficient of the i-th structural optimization index, is the node center coefficient of the h-th structural optimization index, is the score difference coefficient of the h-th structural optimization index, is the total number of structural optimization indexes.

[0139] 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 indexes with larger score difference coefficients are screened out. These indexes 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 indexes are automatically identified to avoid the deviation of manual subjective judgment. Not only the discrimination degree of the index is considered, but also the correlation between indexes is combined to ensure that the selected indexes are both independent and representative. Through weight assignment, the indexes that have the greatest impact on the optimization goal are highlighted, the interference of secondary indexes is reduced, and the optimization efficiency is improved. The index weights can be dynamically adjusted according to different optimization scenarios, with strong adaptability.

[0140] Step 220, obtain the structural information of the suspension line 3dB bridge to be optimized.

[0141] Step 230, based on the structural information of the suspension line 3dB bridge to be optimized, determine multiple structural parameters to be optimized.

[0142] Among them, the structural parameters to be optimized include the structural parameters of the center double-sided coupled transmission line and the structural parameters of the side double-sided coupled transmission line.

[0143] Preferably, step 230 specifically includes:

[0144] For each structural parameter to be screened for optimization and each structural optimization metric, based on the scores of each sample structural scheme in the structural optimization metric, calculate the influence coefficient of the structural parameter to be screened for optimization on the structural optimization metric. Specifically, according to the calculation formula of the non-linear correlation coefficient (for example, 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 metric to calculate the influence coefficient of the structural parameter to be screened for optimization on the structural optimization metric;

[0145] Based on the influence coefficient of each structural parameter to be screened for optimization on each structural optimization metric, determine multiple structural parameters to be optimized.

[0146] Preferably, based on the influence coefficient of each structural parameter to be screened for optimization on each structural optimization metric, determine multiple structural parameters to be optimized, including:

[0147] 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 metric and the weight corresponding to each structural optimization metric, determine the global influence coefficient of the structural parameter to be screened for optimization;

[0148] 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 parameter to be screened for optimization with a global influence coefficient greater than the fifth threshold can be used as the structural parameter to be optimized.

[0149] Specifically, the global influence coefficient of the structural parameter to be screened for optimization can be calculated according to the following formula:

[0150] ,

[0151] 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 metric.

[0152] It can be understood that in the prior art, relying on the experience of engineers or the trial-and-error method, it is easy to overlook 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 metric, the contribution of the parameter to the optimization target 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 target through the global influence coefficient, reduces the number of optimization variables, and reduces the calculation complexity.

[0153] Step 240: Generate multiple structure optimization schemes based on multiple structure parameters to be optimized, and determine the optimal structure optimization scheme based on multiple structure optimization schemes through an improved genetic algorithm and finite element analysis.

[0154] Figure 5 It is a schematic flowchart of determining the optimal structure optimization scheme shown in some embodiments of this specification. As shown in Figure 5 As shown, preferably, step 240 specifically includes:

[0155] S11: Determine the parameter priority of each structure parameter to be optimized;

[0156] S12: Based on the parameter priority of each structure parameter to be optimized, determine the current structure parameter to be optimized. For example, use the structure parameter to be optimized with the highest parameter priority and not yet completed optimization as the current structure parameter to be optimized;

[0157] S13: Generate a structure optimization scheme corresponding to the current structure parameter to be optimized according to the constraint conditions of the current structure parameter to be optimized and the optimal values of the structure parameters that have been optimized;

[0158] S14: Through an improved genetic algorithm and finite element analysis, based on the structure optimization scheme corresponding to the current structure parameter to be optimized, determine the optimal value of the current structure parameter to be optimized, and mark the current structure parameter to be optimized as a structure parameter that has been optimized;

[0159] S15: Determine whether all structure parameters to be optimized have been optimized. If so, end the optimization. If not, execute S12.

[0160] Preferably, determining the parameter priority of each structure parameter to be optimized includes:

[0161] Based on the global influence coefficient of each structure parameter to be screened, determine the global influence coefficient of each structure parameter to be optimized. For example, if the global influence coefficient of the structure parameter to be screened A is 0.5, then the corresponding global influence coefficient of the structure parameter to be optimized A is 0.5;

[0162] Based on the global influence coefficient of each structure parameter to be optimized, determine the parameter priority of each structure parameter to be optimized, where the greater the global influence coefficient, the higher the parameter priority of the structure parameter to be optimized.

[0163] Preferably, through an improved genetic algorithm and finite element analysis, based on the structure optimization scheme corresponding to the current structure parameter to be optimized, determining the optimal value of the current structure parameter to be optimized includes:

[0164] For each value segment corresponding to the current structural parameter to be optimized, sample multiple structural optimization schemes corresponding to the value segment. Through finite element analysis, calculate the fitness values of the multiple structural optimization schemes corresponding to the value segment. According to the fitness values of the multiple structural optimization schemes 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;

[0165] According to the fitness value characteristics corresponding to each value segment, determine the priority value corresponding to each value segment;

[0166] Through finite element analysis, determine the fitness value of the structural optimization scheme corresponding to the current structural parameter to be optimized;

[0167] Based on the fitness value of the structural optimization scheme corresponding to the current structural parameter to be optimized and the priority value corresponding to each value segment, use the genetic algorithm to determine the optimal value of the current structural parameter to be optimized.

[0168] For example, if the value range corresponding to the current structural parameter 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 [20 - 30] segment: 22, 24, 26, 28, 30.

[0169] The priority value corresponding to the value segment can be calculated according to the following formula:

[0170] ,

[0171] 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.

[0172] The optimal value of the current structural parameter to be optimized can be determined according to the following process:

[0173] Combining the global fitness value and the value segment priority value, use an improved genetic algorithm to determine the optimal value of the structural parameter to be optimized. The specific steps include:

[0174] S31. Population initialization: Randomly generate an initial population, where each individual contains the values of all parameters to be optimized.

[0175] S32. Fitness evaluation: Perform FEA on each individual and calculate its fitness value.

[0176] S33. Selection operation: Select excellent individuals to enter the next generation according to the fitness value and priority value. For example, individuals within the value range with a higher priority value have a higher probability of being selected; individuals with a higher fitness value are preferentially retained.

[0177] 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 range with a higher priority value. For example, if the priority value of the priority value segment [20 - 30] is 0.8 and the priority value of [40 - 50] is 0.2, then the mutation operation is more likely to perturb the parameters within the [20 - 30] segment.

[0178] S35. Update the priority value: Dynamically adjust the priority value corresponding to each value range according to the fitness distribution of each generation of the population. For example, if the fitness of individuals within a certain value range is continuously high, then increase its priority value; if there are no excellent individuals in a certain value range for a long time, then decrease its priority value

[0179] S36. Iterative update: Repeat the above steps until the convergence condition is met (such as the change in the fitness value is less than the threshold).

[0180] 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 ranges. This method divides continuous parameters into discrete value ranges, reducing the search dimension. Dynamically allocate priorities based on the fitness value characteristics of each value range, 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 scheme level and the priority value at the parameter segment level. The value range with a higher priority value is preferentially crossed and mutated, dynamically adjusting the search direction. Balances local optimization and global exploration, and improves the robustness of the algorithm.

[0181] Figure 6 is a schematic diagram of the modules of a structure optimization system for a suspended line 3dB bridge shown 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.

[0182] The index determination module is used to determine multiple structure optimization indexes, where the multiple structure optimization indexes are related to multiple physical fields;

[0183] The information acquisition module is used to acquire the structure information of the suspended line 3dB bridge to be optimized;

[0184] A parameter determination module, configured to determine a plurality of structural parameters to be optimized based on the structural information of the suspension line 3dB bridge, 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;

[0185] A structure optimization module, configured to generate a variety of structure optimization schemes based on the plurality of structural parameters to be optimized, and determine the optimal structure optimization scheme based on the variety of structure optimization schemes through an improved genetic algorithm and finite element analysis.

[0186] A structure optimization system for a suspension line 3dB bridge can apply the above-mentioned structure optimization method for a suspension line 3dB bridge, which will not be elaborated here.

[0187] 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 deformations 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 introduced 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, where 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 arranged around the central double-sided coupled transmission line; The method includes: Determining a plurality of structure optimization indexes, where 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, 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 lines; 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, based on the variety of structure optimization schemes, determining the optimal structure optimization scheme; Based on the structure information of the suspended line 3dB bridge to be optimized, determining a plurality of structure parameters to be optimized, including: Determining 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 structure parameter to be optimized to be screened and each structure optimization index, based on the scores of each sample structure scheme in the structure optimization index, calculating the influence coefficient of the structure parameter to be optimized to be screened on the structure optimization index; For each structure parameter to be optimized to be screened, based on the influence coefficient of the structure parameter to be optimized to be screened on each structure optimization index and the weight corresponding to each structure optimization index, determining the global influence coefficient of the structure parameter to be optimized to be screened; Based on the global influence coefficient of each structure parameter to be optimized to be screened, determining a plurality of structure parameters to be optimized; 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, based on the variety of structure optimization schemes, determining the optimal structure optimization scheme, including: For each value range corresponding to the structure parameter to be optimized, sampling a variety of structure optimization schemes corresponding to the value range, through finite element analysis, calculating the fitness values of the variety of structure optimization schemes corresponding to the value range, and based on the fitness values of the variety of structure optimization schemes corresponding to the value range, determining the fitness value characteristics corresponding to the value range; Based on the fitness value characteristics corresponding to each value range, determining the priority value corresponding to each value range; Through finite element analysis, determining the fitness value of the structure optimization scheme corresponding to the structure parameter to be optimized; Through the genetic algorithm, based on the fitness value of the structure optimization scheme corresponding to the structure parameter to be optimized and the priority value corresponding to each value range, determining the optimal value of the structure parameter to be optimized, where the optimal structure optimization scheme includes the optimal value of the structure parameter to be optimized.

2. The structural optimization method for a suspended line 3dB bridge according to claim 1, characterized in that Determine multiple structural optimization indicators, including: Determine multiple structural optimization indicators to be screened; For each sample structure scheme, through finite element analysis, determine the scores of the sample structure scheme in multiple structural optimization indicators; For each structural optimization indicator to be screened, based on the scores of each sample structure scheme in multiple structural optimization indicators, determine the score difference coefficient corresponding to the structural optimization indicator to be screened; Based on the score difference coefficient corresponding to each structural optimization indicator to be screened, determine multiple structural optimization indicators and the weight corresponding to each structural optimization indicator.

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 structural optimization indicator to be screened, determine multiple structural optimization indicators and the weight corresponding to each structural optimization indicator, including: Based on the score difference coefficient corresponding to each structural optimization indicator to be screened, determine multiple candidate structural optimization indicators; For any two candidate structural optimization indicators, according to the scores of each sample structure scheme in the two candidate structural optimization indicators, calculate the score correlation coefficient of the two candidate structural optimization indicators; Based on the score correlation coefficient of the two candidate structural optimization indicators, establish an index map; Based on the index map, determine the node central coefficient of each candidate structural optimization indicator; Based on the node central coefficient and score difference coefficient of each candidate structural optimization indicator, determine multiple structural optimization indicators and the weight corresponding to each structural optimization indicator.

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

5. A structural optimization method for a suspended line 3dB bridge according to claim 1, characterized in that 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 the 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.

6. The structural optimization method for a suspended line 3dB bridge according to claim 5, characterized in that, 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 to be optimized, 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.

7. A structural optimization system for a suspended line 3dB bridge, characterized in that, Applying the structural optimization method for a suspended line 3dB bridge according to any one of claims 1-6, comprising: An index determination module for determining a plurality of structural optimization indices, wherein the plurality of structural optimization indices 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