A microwave filter debugging method based on multi-swarm particle swarm optimization method
By employing a multi-swarm particle swarm optimization method, particles with low target parameter values are discarded, while high-efficiency particles are retained for iteration. This solves the problem of high computational load in filter debugging, achieving a more efficient debugging process and shorter debugging time.
Patent Information
- Application Number
- CN202411188586.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-08-28
AI Technical Summary
In existing technologies, the filter debugging process involves a large amount of computation and is complex, making it difficult to reduce the amount of computation.
A multi-swarm particle swarm optimization method is adopted. By initializing the particle swarm and iterating, particles with target parameter values less than a preset threshold are discarded, and high-efficiency particles are retained for the next iteration to determine the target operating parameter values for debugging the microwave filter.
While ensuring debugging effectiveness, the number of particles participating in the iteration is reduced, the amount of computation is decreased, the debugging time is shortened, and the user experience is improved.
Smart Images

Figure CN118709573B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of specific computing models, and in particular to a microwave filter debugging method based on a multi-swarm particle swarm optimization method. Background Art
[0002] The filter includes a filter circuit composed of capacitors, inductors and resistors. The filter can filter out the frequency point of a specific frequency in the input signal or the frequency other than the frequency point to obtain the required output signal; in practical applications, the filter can remove noise in the signal, thereby improving the signal-to-noise ratio and making the signal clearer; or, in data communication, the filter can also limit the bandwidth of the transmission signal to avoid interference between signals.
[0003] In order to make the same filter adapt to different usage requirements, the filter can be debugged. In related technologies, such as the Chinese patent application document with publication number CN115902487A, a microwave filter debugging method based on knowledge-driven particle swarm optimization is provided, including: changing the adjustable variable u* of the adjustable component on the microwave filter for multiple times to obtain the S matrix S*, using a vector network analyzer to sample and measure the S parameter s, and constructing a debugging data set; inputting the performance index I*, amplitude-frequency response Ra and phase-frequency response that meet the requirements into the trained knowledge model, and obtaining the corresponding output to constitute the debugging knowledge; using the debugging knowledge to determine the optimization range of the particle swarm algorithm, designing an evaluation function according to the optimization range, performance index and its requirements, as well as the population size, inertia weight, acceleration weight, stop standard and debugging and adjustment parameters, and performing iterative optimization; converting the S matrix S* into the Y parameter yp, and using the Elman neural network to construct a debugging process model, combining the iterative optimization results, and obtaining the optimization results of the adjustable variables to reflect the performance index of the microwave filter; when the debugging standard is reached, the actual microwave filter is debugged according to the optimization results, and the overall performance of the microwave filter is comprehensively evaluated.
[0004] In the related art, there are many types of parameters used when debugging the filter, and the calculation process of the filter debugging process is relatively complicated, which is not convenient for reducing the calculation amount of the filter debugging process. Summary of the invention
[0005] In order to overcome the problem that it is not convenient to reduce the amount of calculation in the debugging process of the filter in the related art, the present application provides a microwave filter debugging method based on a multi-population particle swarm optimization method, comprising: initializing a particle swarm in a pre-constructed multi-population particle swarm algorithm, and iterating the initialized particle swarm; different positions in the solution space of the particle swarm correspond to different operating parameter values of the microwave filter; in the case of iterating the multi-population particle swarm algorithm, determining the target parameter value of the particle in the particle swarm; the target parameter value is used to characterize the ratio between the search efficiency of the particle for the fitness value and the aggregation degree of the population where the particle is located, and the fitness value is used to characterize the degree of difference between the first filter parameter and the target filter parameter of the microwave filter; discarding the particles whose target parameter value in the particle swarm is less than a preset parameter threshold, and performing the next iteration of the multi-population particle swarm algorithm according to the retained particles; when the number of iterations reaches the preset number threshold, determining the target particle with the smallest fitness value in the particle swarm, and determining the target operating parameter value corresponding to the spatial position of the target particle in the solution space, so as to operate the microwave filter according to the target operating parameter value.
[0006] In this way, when performing iterations of the multi-population particle swarm algorithm, particles whose target parameter values are less than the preset parameter threshold in the particle swarm can be discarded, and the next iteration of the multi-population particle swarm algorithm is performed based on the retained particles. The target parameter value is used to characterize the ratio between the search efficiency of the particle for the fitness value and the aggregation degree of the population to which the particle belongs. The operation parameter value can be searched through particles with higher search efficiency as much as possible. Therefore, the number of particles participating in the iteration can be reduced while ensuring the debugging effect of the microwave filter, thereby reducing the computational complexity of the debugging process of the microwave filter.
[0007] At the same time, since the particles whose target parameter values in the particle swarm are less than the preset parameter threshold are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed according to the retained particles, the overall search efficiency of the operating parameter values through the retained particles is higher. Therefore, it is also possible to shorten the time required for debugging the microwave filter while ensuring the debugging effect of the microwave filter.
[0008] Furthermore, by searching the operating parameter values of the microwave filter through the particle swarm, compared with manually adjusting the operating parameter values of the filter, the number of user operations can be reduced, thereby improving the user's operating experience in debugging the microwave filter.
[0009] Optionally, determining a target parameter value of a particle in a particle swarm includes: determining a fitness value of the particle in the particle swarm according to a first filter parameter of a microwave filter under a first operating parameter value and a target filter parameter; the first operating parameter is an operating parameter value corresponding to a spatial position of the particle in a solution space; determining a search efficiency value of the particle for the fitness value according to multiple fitness values of the particle in multiple iterations; determining a degree of aggregation value of a population to which the particle in the particle swarm belongs, and taking a ratio of the search efficiency value of the particle to the degree of aggregation value as the target parameter value of the particle.
[0010] Optionally, according to multiple fitness values of the particle in multiple iterations, a search efficiency value of the particle for the fitness value is determined, including: ;in, is the particle's search efficiency for the fitness value at the Mth iteration, is the fitness value of the particle at the kth iteration, is the fitness value of the particle at the k-1th iteration, M is the number of iterations of the multi-population particle swarm algorithm, exp is an exponential function with a natural constant as the base, ReLU is a linear rectification function, and norm is a normalization function.
[0011] In this way, if the fitness value of the particle is always in the rising stage during multiple iterations, compared with the fitness value of the particle not always in the rising stage during multiple iterations, the value of the linear rectification function is larger, which makes the value of the search efficiency value smaller. Therefore, the search efficiency value can better characterize the search efficiency of the particle for the fitness value.
[0012] Optionally, the aggregation degree value of the population to which the particles belong in the particle swarm is determined by: ,in, is the aggregation degree value of the particle population at the Mth iteration, is the distance between the i-th particle and the j-th particle in the population of the particle in the solution space at the M-th iteration, is the number of particles in the population where the particle is located at the Mth iteration, and norm is the normalization function.
[0013] Optionally, the first filter parameters include a first reflection coefficient and a first transmission coefficient, and the target filter parameters include a target reflection coefficient and a target transmission coefficient; determining the fitness value of a particle in the particle swarm according to the first filter parameters and the target filter parameters of the microwave filter under the first operating parameter value includes: ; Where Y is the fitness value of the particle, b is the upper limit of the target frequency range, and a is the lower limit of the target frequency range. is the first reflection coefficient of the microwave filter at the frequency f when the microwave filter operates according to the first operating parameter value, is the target reflection coefficient at frequency f, is the first transmission coefficient of the microwave filter at the frequency f when the microwave filter operates according to the first operating parameter value, is the target transmission coefficient at frequency f, To take the absolute value operation.
[0014] In this way, the first reflection coefficient of the microwave filter is compared with the target reflection coefficient, and the first transmission coefficient of the microwave filter is compared with the target transmission coefficient. The obtained fitness value can better characterize the degree of difference between the first filter parameter of the microwave filter and the target filter parameter, so as to obtain a smaller fitness value.
[0015] Optionally, the target filter parameters are obtained in the following manner: obtaining a simulation model for a microwave filter created by electromagnetic simulation software, the simulation model being used to output a target output signal based on an input target input signal; inputting the target input signal into the simulation model so that the simulation model outputs the target output signal; and when the simulation model outputs the target output signal, using the filter parameters of the simulation model as the target filter parameters.
[0016] In this way, by inputting the target input signal into the simulation model so that the simulation model outputs the target output signal, when the simulation model outputs the target output signal, the filter parameters of the simulation model can be used as the target filter parameters, which can provide a better reference for the debugging of the microwave filter.
[0017] Optionally, the microwave filter includes a plurality of tuning rods, and when the plurality of tuning rods are in different positions, the operating parameter values corresponding to the microwave filter are different.
[0018] Optionally, a particle swarm in a pre-constructed multi-population particle swarm algorithm is initialized, including: setting a plurality of populations of particle swarms, wherein the total number of the plurality of populations of particle swarms is a target number; obtaining an adjustment amplitude of a tuning rod of the microwave filter, and determining a maximum distance in a solution space of the multi-population particle swarm algorithm according to the adjustment amplitude; and initializing a velocity of the particle swarm in the solution space to complete the initialization of the particle swarm in the pre-constructed multi-population particle swarm algorithm.
[0019] In this way, the maximum distance of the solution space of the multi-swarm particle swarm algorithm is determined according to the adjustment amplitude, so that the search range of the particles can correspond to the adjustment amplitude of the tuning rod, so that the particles converge at a spatial position that better matches the corresponding operating parameter value.
[0020] The technical solution provided by the embodiments of the present application may include the following beneficial effects: when performing iterations of a multi-population particle swarm algorithm, particles whose target parameter values in the particle swarm are less than a preset parameter threshold value may be discarded, and the next iteration of the multi-population particle swarm algorithm may be performed based on the retained particles. The target parameter value is used to characterize the ratio between the search efficiency of the particle for the fitness value and the aggregation degree of the population to which the particle belongs. The search for the operating parameter value may be performed through particles with higher search efficiency as much as possible. Therefore, the number of particles participating in the iteration may be reduced while ensuring the debugging effect of the microwave filter, thereby reducing the computational complexity of the debugging process of the microwave filter.
[0021] At the same time, since the particles whose target parameter values in the particle swarm are less than the preset parameter threshold are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed according to the retained particles, the overall search efficiency of the operating parameter values through the retained particles is higher. Therefore, it is also possible to shorten the time required for debugging the microwave filter while ensuring the debugging effect of the microwave filter.
[0022] Furthermore, by searching the operating parameter values of the microwave filter through the particle swarm, compared with manually adjusting the operating parameter values of the filter, the number of user operations can be reduced, thereby improving the user's operating experience in debugging the microwave filter.
[0023] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0025] Figure 1 It is a flowchart schematic diagram of a microwave filter debugging method based on a multi-swarm particle swarm optimization method according to an exemplary embodiment;
[0026] Figure 2 The figure is a flowchart of a process of determining target parameter values of particles in a particle group according to an exemplary embodiment. DETAILED DESCRIPTION
[0027] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.
[0028] First, the application scenario of the embodiment of the present application is briefly introduced. In the application scenario of the present application, in order to make the same filter adapt to different usage requirements, for example, in order to enable the filter to output the target output signal according to the input signal, the filter can be debugged.
[0029] When debugging the filter, it is necessary to manually observe the waveform of the output signal displayed by the oscilloscope and repeatedly debug the filter to make the waveform of the output signal displayed by the oscilloscope consistent with the waveform of the target output signal; manual debugging of the filter depends on the level of the debugger, and the debugging process requires a lot of time of the debugger.
[0030] In the related art, a debugging process model is constructed using a trained knowledge model and an Elman neural network to implement filter debugging. However, the parameters used in the related art are of various types, which is not conducive to reducing the amount of calculation in the filter debugging process.
[0031] In view of the above technical problems, an embodiment of the present application provides a microwave filter debugging method based on a multi-population particle swarm optimization method. When iterating the multi-population particle swarm algorithm, particles whose target parameter values in the particle swarm are less than a preset parameter threshold value can be discarded, and the next iteration of the multi-population particle swarm algorithm is performed based on the retained particles. The target parameter value is used to characterize the ratio between the search efficiency of the particle for the fitness value and the aggregation degree of the population to which the particle belongs. The operation parameter value can be searched through particles with higher search efficiency as much as possible. Therefore, the number of particles participating in the iteration can be reduced while ensuring the debugging effect of the microwave filter, thereby reducing the computational complexity of the debugging process of the microwave filter.
[0032] At the same time, since the particles whose target parameter values in the particle swarm are less than the preset parameter threshold are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed according to the retained particles, the overall search efficiency of the operating parameter values through the retained particles is higher. Therefore, it is also possible to shorten the time required for debugging the microwave filter while ensuring the debugging effect of the microwave filter.
[0033] Furthermore, by searching the operating parameter values of the microwave filter through the particle swarm, compared with manually adjusting the operating parameter values of the filter, the number of user operations can be reduced, thereby improving the user's operating experience in debugging the microwave filter.
[0034] Figure 1 is a flowchart schematic diagram of a microwave filter debugging method based on a multi-swarm particle swarm optimization method according to an exemplary embodiment, such as Figure 1 As shown, the method includes the following steps.
[0035] In step S101, a particle swarm in a pre-constructed multi-swarm particle swarm algorithm is initialized, and the initialized particle swarm is iterated.
[0036] Different positions in the solution space of the particle swarm correspond to different operating parameter values of the microwave filter.
[0037] In one embodiment, the microwave filter includes a plurality of tuning rods, and when the plurality of tuning rods are in different positions, the operating parameter values corresponding to the microwave filter are different.
[0038] By adjusting the adjustment rod of the microwave filter, the operating parameter value of the microwave filter can be changed. Here, different operating parameter values correspond to different positions in the solution space of the particle swarm. Therefore, when multiple tuning rods are in different positions, they correspond to different positions in the solution space of the particle swarm.
[0039] For example, a preset correspondence relationship can be determined, which is used to characterize the correspondence between different positions of the tuning rod and different positions in the solution space of the particle swarm. In order to determine the target position of the different positions of the tuning rod that can enable the filter to output the target output signal, particles can be used to search in the solution space of the particle swarm.
[0040] The multi-population particle swarm algorithm is an improved algorithm based on the particle swarm optimization algorithm. The particle swarm optimization algorithm is an optimization algorithm that simulates the foraging behavior of bird flocks. It finds the optimal solution to the problem through information sharing and mutual cooperation among individuals in the group.
[0041] The multi-population particle swarm algorithm can divide the entire particle swarm into multiple populations based on the particle swarm optimization algorithm. Each population can have an independent search solution space, and the particles can communicate information to improve the global search ability and convergence speed of the algorithm.
[0042] In one embodiment, a particle swarm in a pre-constructed multi-population particle swarm algorithm is initialized, including: setting a plurality of particle swarms of populations, wherein the total number of the particle swarms of the plurality of populations is a target number; obtaining an adjustment amplitude of a tuning rod of the microwave filter, and determining a maximum distance in a solution space of the multi-population particle swarm algorithm according to the adjustment amplitude; and initializing a velocity of the particle swarm in the solution space to complete the initialization of the particle swarm in the pre-constructed multi-population particle swarm algorithm.
[0043] The target number can be set according to actual needs, for example, the target number can be between 500 and 600, and the number of populations in the particle swarm can be between 5 and 10, for example.
[0044] The maximum distance of the solution space of the multi-swarm particle swarm algorithm is determined according to the adjustment amplitude, so that the search range of the particles can correspond to the adjustment amplitude of the tuning rod, so that the particles converge at a spatial position that better matches the corresponding operating parameter value.
[0045] In step S102, when performing multiple iterations of the particle swarm algorithm, target parameter values of particles in the particle swarm are determined.
[0046] The target parameter value is used to characterize the ratio between the particle's search efficiency for the fitness value and the aggregation degree of the particle's population, and the fitness value is used to characterize the difference between the first filter parameter of the microwave filter and the target filter parameter.
[0047] Since different positions in the solution space of the particle swarm correspond to different operating parameter values of the microwave filter, the operating parameter values of the microwave filter can be searched by searching the positions of the particles of the particle swarm in the solution space.
[0048] The smaller the fitness value corresponding to the particle is, the smaller the difference between the first filter parameter and the target filter parameter of the microwave filter is, and the more the output signal of the microwave filter when operating according to the first filter parameter is consistent with the target output signal.
[0049] On the contrary, the larger the fitness value corresponding to the particle is, the greater the difference between the first filter parameter and the target filter parameter of the microwave filter is, and the more the output signal of the microwave filter when operating according to the first filter parameter is inconsistent with the target output signal.
[0050] The target parameter value is used to characterize the ratio between the particle search efficiency for fitness value and the aggregation degree of the particle population. In the early stage of the multi-population particle swarm algorithm iteration, the search degree of particles in the particle swarm is shallow, the particles may be more dispersed and the fitness value is small; with the increase of the number of iterations, the particles are usually concentrated near the corresponding local optimal solution, which increases the aggregation degree of the particle population.
[0051] In the process of particle swarm iteration, as the particles' search for fitness values deepens, the aggregation of the particle population increases. There may still be some particles with low search efficiency for fitness values. The probability of these particles finding the minimum fitness value after iteration is lower. Continuing to use these particles to search for fitness values will cause a waste of computing power. Therefore, the target parameter value can help screen out particles in the particle swarm that may cause a waste of computing power.
[0052] In step S103, particles in the particle swarm whose target parameter values are less than a preset parameter threshold are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed based on the retained particles.
[0053] Since the target parameter value represents the ratio between the particle's search efficiency for the fitness value and the aggregation degree of the particle's population, the particles whose target parameter value in the particle swarm is less than the preset parameter threshold are particles with lower search efficiency for the fitness value in the particle swarm and higher aggregation degree of the population.
[0054] The probability of finding a smaller fitness value after continuing iteration for particles whose target parameter value is less than the preset parameter threshold is lower. Continuing to use particles whose target parameter value is less than the preset parameter threshold will cause a waste of computing power. Therefore, the particles whose target parameter value is less than the preset parameter threshold in the particle swarm are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed based on the retained particles. This can not only ensure the search effect for the fitness value, but also reduce the amount of calculation in the fitness value search process, thereby reducing the amount of calculation required for debugging the microwave filter and shortening the time required for debugging the microwave filter.
[0055] The preset parameter threshold may be determined according to the value range of the target parameter value. For example, when the value range of the target parameter value is between 0 and 1, the preset parameter threshold may be between 0.1 and 0.2.
[0056] In step S104, when the number of iterations reaches a preset number threshold, the target particle with the smallest fitness value in the particle swarm is determined, and the target operating parameter value corresponding to the spatial position of the target particle in the solution space is determined to operate the microwave filter according to the target operating parameter value.
[0057] The particles participating in the next iteration are the particles whose target parameter values in the particle swarm of the previous iteration process are greater than or equal to the preset parameter threshold. After multiple iterations of the multi-swarm particle swarm algorithm and the abandonment of particles whose target parameter values are less than the preset parameter threshold, the particles in the particle swarm can be gathered near a smaller fitness value.
[0058] In order to take into account the search effect for a smaller fitness value and the amount of computation required for the particle swarm algorithm to iterate, a preset number threshold can be set in advance. When the number of iterations reaches the preset number threshold, the iteration of the particle swarm algorithm is stopped; on the contrary, when the number of iterations does not reach the preset number threshold, the iteration of the particle swarm algorithm can continue to be executed and the particles whose target parameter values are less than the preset parameter threshold can be discarded in order to obtain a smaller fitness value. The preset number threshold can be set according to the actual accuracy requirements, for example, the preset number threshold can be between 300 and 500.
[0059] When the number of iterations reaches a preset threshold, since the fitness value is used to characterize the degree of difference between the first filter parameter and the target filter parameter of the microwave filter, the target particle with the smallest fitness value in the particle swarm can be determined so that the filter outputs a target output signal that better meets actual needs.
[0060] Furthermore, since the spatial position of the target particle in the solution space corresponds to the operating parameter value of the microwave filter, the target operating parameter value corresponding to the spatial position of the target particle in the solution space can be determined to operate the microwave filter according to the target operating parameter value.
[0061] According to the microwave filter debugging method based on the multi-population particle swarm optimization method provided by the embodiment of the present application, when iterating the multi-population particle swarm algorithm, particles whose target parameter values in the particle swarm are less than the preset parameter threshold value can be discarded, and the next iteration of the multi-population particle swarm algorithm is performed based on the retained particles. The target parameter value is used to characterize the ratio between the search efficiency of the particle for the fitness value and the aggregation degree of the population to which the particle belongs. The operation parameter value can be searched through particles with higher search efficiency as much as possible. Therefore, the number of particles participating in the iteration can be reduced while ensuring the debugging effect of the microwave filter, thereby reducing the calculation amount of the debugging process of the microwave filter.
[0062] At the same time, since the particles whose target parameter values in the particle swarm are less than the preset parameter threshold are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed according to the retained particles, the overall search efficiency of the operating parameter values through the retained particles is higher. Therefore, it is also possible to shorten the time required for debugging the microwave filter while ensuring the debugging effect of the microwave filter.
[0063] Furthermore, by searching the operating parameter values of the microwave filter through the particle swarm, compared with manually adjusting the operating parameter values of the filter, the number of user operations can be reduced, thereby improving the user's operating experience in debugging the microwave filter.
[0064] In one embodiment, the process of determining the target parameter value of a particle in a particle group is as follows: Figure 2 The example shown may include step S201, step S202 and step S203.
[0065] In step S201, the fitness values of particles in the particle swarm are determined according to the first filter parameters of the microwave filter under the first operating parameter value and the target filter parameters.
[0066] The fitness value is used to characterize the difference between the first filter parameter and the target filter parameter of the microwave filter under the first operating parameter value, and the first operating parameter is the operating parameter value corresponding to the spatial position of the particle in the solution space.
[0067] Different spatial positions of particles in the solution space correspond to different operating parameter values of the microwave filter. By comparing the first filter parameter of the microwave filter at the first operating parameter value with the target filter parameter, the fitness value of the particle in the particle swarm can be determined, so that the degree of difference between the first filter parameter of the microwave filter at the first operating parameter value and the target filter parameter can be characterized by the fitness value.
[0068] In one embodiment, the first filter parameters include a first reflection coefficient and a first transmission coefficient, and the target filter parameters include a target reflection coefficient and a target transmission coefficient; determining the fitness value of a particle in a particle swarm according to the first filter parameters and the target filter parameters of the microwave filter under the first operating parameter value includes: ; Where Y is the fitness value of the particle, b is the upper limit of the target frequency range, and a is the lower limit of the target frequency range. is the first reflection coefficient of the microwave filter at the frequency f when the microwave filter operates according to the first operating parameter value, is the target reflection coefficient at frequency f, is the first transmission coefficient of the microwave filter at the frequency f when the microwave filter operates according to the first operating parameter value, is the target transmission coefficient at frequency f, To take the absolute value operation.
[0069] The scattering coefficient (Scattering Parameters, also commonly referred to as S parameters) of a filter is a parameter that describes the measurement results of a network analyzer. The scattering coefficient can characterize the voltage wave amplitude and phase relationship between the input port and the output port of a microwave component (such as a filter, amplifier, and antenna). Alternatively, the scattering coefficient can characterize the amplitude and phase relationship between the current wave at the input port and the output port of a microwave component.
[0070] The scattering coefficient can generally include the reflection coefficient, transmission coefficient, reverse transmission coefficient and output reflection coefficient; among which, the reflection coefficient is used to describe the ratio of the reflected wave to the incident wave at the input port; the transmission coefficient is used to describe the ratio of the transmitted wave from the input port to the output port; the reverse transmission coefficient is used to describe the ratio of the reverse transmitted wave from the output port to the input port; the output reflection coefficient is used to describe the ratio of the reflected wave to the incident wave at the output port.
[0071] In the embodiment of the present application, the reflection coefficient and the transmission coefficient are taken as examples to implement the supervision of the debugging process of the microwave filter. The first filter parameter includes a first reflection coefficient and a first transmission coefficient. The first reflection coefficient refers to the reflection coefficient when the microwave filter operates according to the first operating parameter value. The first transmission coefficient refers to the transmission coefficient when the microwave filter operates according to the first operating parameter value. The target reflection coefficient is the desired reflection coefficient, and the target transmission coefficient is the desired transmission coefficient.
[0072] In this way, the first reflection coefficient of the microwave filter is compared with the target reflection coefficient, and the first transmission coefficient of the microwave filter is compared with the target transmission coefficient. The obtained fitness value can better characterize the degree of difference between the first filter parameter of the microwave filter and the target filter parameter, so as to obtain a smaller fitness value.
[0073] In one embodiment, the target filter parameters are obtained in the following manner: obtaining a simulation model for a microwave filter created by electromagnetic simulation software, the simulation model being used to output a target output signal based on an input target input signal; inputting the target input signal into the simulation model so that the simulation model outputs the target output signal; and when the simulation model outputs the target output signal, using the filter parameters of the simulation model as the target filter parameters.
[0074] The filter can be simulated through electromagnetic simulation software such as HFSS (High Frequency Structure Simulator), COMSOL Multiphysics and NI Multisim. Since the electromagnetic simulation software is made by integrating the data of multiple filters, the virtual simulation model created by the electromagnetic simulation software can provide a reference for the debugging of the actual microwave filter, so as to better realize the debugging of the microwave filter.
[0075] In this way, by inputting the target input signal into the simulation model so that the simulation model outputs the target output signal, when the simulation model outputs the target output signal, the filter parameters of the simulation model can be used as the target filter parameters, which can provide a better reference for the debugging of the microwave filter.
[0076] In step S202, according to multiple fitness values of particles in multiple iterations, a search efficiency value of the particle for the fitness value is determined.
[0077] The search efficiency value of the fitness value of the particles can be used to characterize the search efficiency of the fitness value of the particles in the particle swarm. The higher the search efficiency value of the fitness value of the particles is, the higher the search efficiency of the fitness value of the particles is; on the contrary, the lower the search efficiency value of the fitness value of the particles is, the lower the search efficiency of the fitness value of the particles is.
[0078] In one embodiment, according to multiple fitness values of particles in multiple iterations, determining the search efficiency value of the fitness value of the particle includes:
[0079] ;in, is the particle's search efficiency for the fitness value at the Mth iteration, is the fitness value of the particle at the kth iteration, is the fitness value of the particle at the k-1th iteration, M is the number of iterations of the multi-population particle swarm algorithm, exp is an exponential function with a natural constant as the base, ReLU is a linear rectification function, and norm is a normalization function.
[0080] When the variable of the linear rectification function is less than 0, the value of the linear rectification function is 0; when the variable of the linear rectification function is greater than or equal to 0, the value of the linear rectification function is equal to the value of the variable; by processing the difference between the fitness value of the current iteration process and the fitness value of the previous iteration process through the linear rectification function, when the fitness value of the particle decreases, the value of the linear rectification function is 0.
[0081] When the fitness value of the particle tends to converge, the fitness value of the particle should gradually decrease. If the fitness value of the particle is always in the rising stage during multiple iterations, it means that the particle's search efficiency for the fitness value is low.
[0082] In the calculation formula for the search efficiency value of the particle in the embodiment of the present application, if the fitness value of the particle is always in an increasing stage during multiple iterations, compared with the fitness value of the particle not always in an increasing stage during multiple iterations, the value of the linear rectification function is larger, which makes the value of the search efficiency value smaller. Therefore, the search efficiency value of the embodiment of the present application can better characterize the search efficiency of the particle for the fitness value.
[0083] In step S203, the aggregation degree value of the population to which the particle belongs in the particle swarm is determined, and the ratio of the search efficiency value of the particle to the aggregation degree value is used as the target parameter value of the particle.
[0084] The aggregation degree value of the particle population in the particle swarm can characterize the aggregation degree of the particle population. As the iterative process of the multi-population particle swarm algorithm proceeds, the same particle in the particle swarm gradually approaches the minimum target parameter value, and the aggregation degree of the particle population gradually increases. Therefore, the aggregation degree value of the particle population in the particle swarm can also better reflect the progress of the iteration.
[0085] In one embodiment, the aggregation degree value of the population to which the particles belong in the particle group is determined by: ,in, is the aggregation degree value of the particle population at the Mth iteration, is the distance between the i-th particle and the j-th particle in the population of the particle in the solution space at the M-th iteration, is the number of particles in the population where the particle is located at the Mth iteration, and norm is the normalization function.
[0086] The degree of aggregation of particles in the population to which the particles belong can be effectively determined by the distance between different particles in the population to which the particles belong. The greater the distance between different particles in the population to which the particles belong, the lower the degree of aggregation of particles in the population to which the particles belong. On the contrary, the smaller the distance between different particles in the population to which the particles belong, the higher the degree of aggregation of particles in the population to which the particles belong.
[0087] It should be understood that the features of some embodiments of the various applications described herein may be combined with each other, unless specifically stated otherwise.
[0088] Although terms such as "first", "second", and "third" may be used herein to describe various components, parts, regions, layers, or sections, these components, parts, regions, layers, or sections are not limited to these terms. Instead, these terms are only used to distinguish one component, part, region, layer, or section from another component, part, region, layer, or section. Therefore, without departing from the teachings of the examples described herein, the first component, part, region, layer, or section mentioned in the examples may also be referred to as the second component, part, region, layer, or section.
[0089] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description herein, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0090] In addition, the word "exemplary" is used herein to mean serving as an example, instance, or diagram. Any aspect or design described herein as "exemplary" is not necessarily to be understood as being advantageous over other aspects or designs. Instead, the use of the word exemplary is intended to present concepts in a concrete manner. As used herein, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or".
[0091] Likewise, although the present application has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. With particular regard to the various functions performed by the components (e.g., elements, resources, etc.) described above, unless otherwise noted, the terms used to describe such components are intended to correspond to any component (functionally equivalent) that performs the specific functions of the described components, even if not structurally equivalent to the disclosed structures.
[0092] Additionally, while particular features of the present application may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations as may be desired and advantageous for any given or particular application.
[0093] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application, and the specification and embodiments are only considered as exemplary.
[0094] It should be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.
Claims
1. A microwave filter debugging method based on a multi-swarm particle swarm optimization method, characterized in that: include: Initializing a particle swarm in a pre-constructed multi-swarm particle swarm algorithm, and iterating the initialized particle swarm; different positions in the solution space of the particle swarm correspond to different operating parameter values of the microwave filter; In the case of multiple iterations of the particle swarm algorithm, the target parameter values of the particles in the particle swarm are determined, including: Determine the fitness value of a particle in the particle swarm according to a first filter parameter and a target filter parameter of the microwave filter under a first operating parameter value; the first operating parameter is an operating parameter value corresponding to a spatial position of the particle in the solution space; According to the multiple fitness values of particles in multiple iterations, the search efficiency value of the particle for the fitness value is determined, including: ;in, is the particle's search efficiency for the fitness value at the Mth iteration, is the fitness value of the particle at the kth iteration, is the fitness value of the particle at the k-1th iteration, M is the number of iterations of the multi-population particle swarm algorithm, exp is an exponential function with a natural constant as the base, ReLU is a linear rectification function, and norm is a normalization function; Determine the aggregation degree value of the particle population in the particle swarm by the following method: ,in, is the aggregation degree value of the particle population at the Mth iteration, is the distance between the i-th particle and the j-th particle in the population of the particle in the solution space at the M-th iteration, is the number of particles in the population where the particle is located at the Mth iteration, and the ratio of the particle's search efficiency value to the aggregation degree value is used as the particle's target parameter value; The target parameter value is used to characterize the ratio between the particle's search efficiency for the fitness value and the aggregation degree of the particle's population, and the fitness value is used to characterize the degree of difference between the first filter parameter of the microwave filter and the target filter parameter; The particles whose target parameter values are less than the preset parameter threshold in the particle swarm are discarded, and the next iteration of the multi-swarm particle swarm algorithm is performed based on the retained particles; When the number of iterations reaches a preset number threshold, a target particle with the smallest fitness value in the particle swarm is determined, and a target operating parameter value corresponding to the spatial position of the target particle in the solution space is determined, so as to operate the microwave filter according to the target operating parameter value.
2. The microwave filter debugging method based on multi-swarm particle swarm optimization method according to claim 1, characterized in that: The first filter parameters include a first reflection coefficient and a first transmission coefficient, and the target filter parameters include a target reflection coefficient and a target transmission coefficient; Determining the fitness value of a particle in a particle swarm according to a first filter parameter of the microwave filter under a first operating parameter value and a target filter parameter includes: ; Where Y is the fitness value of the particle, b is the upper limit of the target frequency range, and a is the lower limit of the target frequency range. is the first reflection coefficient of the microwave filter at the frequency f when the microwave filter operates according to the first operating parameter value, is the target reflection coefficient at frequency f, is the first transmission coefficient of the microwave filter at the frequency f when the microwave filter operates according to the first operating parameter value, is the target transmission coefficient at frequency f, To take the absolute value operation.
3. The microwave filter debugging method based on multi-swarm particle swarm optimization method according to claim 2, characterized in that: The target filter parameters are obtained by: Acquire a simulation model for a microwave filter created by electromagnetic simulation software, wherein the simulation model is used to output a target output signal according to an input target input signal; Inputting a target input signal into the simulation model so that the simulation model outputs a target output signal; When the simulation model outputs a target output signal, the filter parameters of the simulation model are used as target filter parameters.
4. The microwave filter debugging method based on multi-swarm particle swarm optimization method according to claim 1, characterized in that: The microwave filter comprises a plurality of tuning rods, and when the plurality of tuning rods are in different positions, the operating parameter values corresponding to the microwave filter are different.
5. The microwave filter debugging method based on multi-swarm particle swarm optimization method according to claim 4, characterized in that: Initialize the particle swarm in the pre-built multi-swarm particle swarm algorithm, including: Set multiple populations of particle swarms, and the total number of particle swarms of multiple populations is the target number; Acquire the adjustment amplitude of the tuning rod of the microwave filter, and determine the maximum distance of the solution space of the multi-swarm particle swarm algorithm according to the adjustment amplitude; The velocity of the particle swarm is initialized in the solution space to complete the initialization of the particle swarm in the pre-built multi-swarm particle swarm algorithm.
