An impedance adaptive matching method for radio frequency power supply

Abstract

The present application relates to the field of power management technology, and provides an impedance adaptive matching method for a radio frequency power supply. The method includes: constructing a three-dimensional model of a target radio frequency circuit based on predetermined circuit characteristics; obtaining a grid with point identification after gridding processing; screening out a target key grid set through a key point screening strategy; dynamically monitoring the first key grid, obtaining its grid feature timing, and extracting the characteristics under the first step time; using the impedance fitness evaluation function analysis to obtain the first impedance fitness; constructing the first impedance fitness timing; if the impedance fitness index is not met, adjusting the initialization output impedance of the radio frequency power supply. The present application solves the technical problem of low transmission efficiency caused by the mismatch between the output impedance of the radio frequency power supply and the impedance of the target radio frequency circuit, and realizes the improvement of the accuracy and efficiency of impedance matching through the impedance adaptive matching method, thereby ensuring the effect of the transmission efficiency of the radio frequency power supply.

Description

Technical Field

[0001] The present application relates to the field of power electronics technology, specifically to the field of power management technology, and in particular to an impedance adaptive matching method for a radio frequency power supply. Background Art

[0002] With the rapid development of wireless communications, radar, medical imaging, industrial heating and other fields, RF power supply is the core component of these systems, and its performance stability and efficiency are crucial to the operation of the overall system. RF power supply drives load devices such as antennas and induction coils by generating high-frequency electromagnetic waves to achieve functions such as signal transmission, energy conversion or object heating. However, in actual applications, the load impedance of RF power supply may change due to various factors such as environmental changes, equipment aging, process adjustments, etc. When the load impedance does not match the output impedance of the RF power supply, it will lead to problems such as reduced energy transmission efficiency and severe heat generation. The traditional RF power supply impedance matching method automatically adapts to changes in load impedance, resulting in disadvantages such as low matching accuracy and slow response speed, which is difficult to meet the modern system's requirements for efficient, stable and intelligent impedance matching. Summary of the invention

[0003] The present application provides an impedance adaptive matching method for a radio frequency power supply, aiming to solve the technical problem of low transmission efficiency caused by mismatch between the output impedance of the radio frequency power supply and the impedance of the target radio frequency circuit.

[0004] In view of the above problems, the present application provides an impedance adaptive matching method for a radio frequency power supply.

[0005] The present application provides an impedance adaptive matching method for a radio frequency power supply, the method comprising: constructing a target three-dimensional model according to target circuit feature information of a target radio frequency circuit collected based on predetermined circuit features, wherein the target radio frequency circuit refers to a circuit including a target radio frequency power supply in the circuit; performing gridding processing on the target three-dimensional model to obtain a target gridding result, the target gridding result including a plurality of grids with point identifiers; reading a key point screening strategy, and analyzing and screening the plurality of grids with point identifiers based on the key point screening strategy to obtain a target key grid set; dynamically monitoring a first key grid in the target key grid set to obtain a first grid feature time sequence, and extracting a first grid feature at a first step time in the first grid feature time sequence; analyzing the first grid feature in combination with an impedance fitness evaluation function to obtain a first impedance fitness at the first step time; constructing a first impedance fitness time sequence of the first key grid according to a first corresponding relationship between the first step time and the first impedance fitness; and adjusting the initialization output impedance of the target radio frequency power supply when the first impedance fitness time sequence does not meet the impedance fitness index constraint.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The above-mentioned impedance adaptive matching method of an RF power supply constructs a three-dimensional model based on the characteristic information of the target RF circuit, and meshes the model for easy analysis. Subsequently, by applying the key point screening strategy, a key grid set is screened out from the meshing result. After that, the first key grid among these key grids is dynamically monitored, its grid feature timing is recorded, and the grid features at a specific time are extracted. The extracted grid features are then analyzed using the impedance fitness evaluation function to evaluate the impedance fitness and construct the impedance fitness timing of the key grid. If the impedance fitness timing does not meet the preset impedance fitness index constraint, the initialization output impedance of the target RF power supply will be automatically adjusted to achieve adaptive impedance matching. This method improves the accuracy and response speed of impedance matching and ensures the transmission efficiency of the RF power supply.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A schematic diagram of a flow chart of an impedance adaptive matching method for a radio frequency power supply in one embodiment;

[0011] Figure 2 A schematic diagram of a process for constructing a target three-dimensional model of an impedance adaptive matching method for a radio frequency power source in an embodiment. DETAILED DESCRIPTION

[0012] The embodiment of the present application solves the technical problem of low transmission efficiency caused by mismatch between the output impedance of the RF power supply and the impedance of the target RF circuit by providing an impedance adaptive matching method for the RF power supply.

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0015] Examples, such as Figure 1 As shown, the present application provides an impedance adaptive matching method for a radio frequency power supply, the method comprising:

[0016] A target three-dimensional model is constructed according to target circuit characteristic information of a target radio frequency circuit collected based on predetermined circuit characteristics, wherein the target radio frequency circuit refers to a circuit including a target radio frequency power supply.

[0017] In an embodiment of the present application, in order to more accurately analyze and optimize the performance of the target RF circuit, the system terminal collects the characteristics of the target RF circuit based on the predetermined characteristic information of the target RF circuit to obtain the target circuit characteristic information. Among them, the target RF circuit refers to a circuit equipped with a target RF power supply. Subsequently, the collected target circuit characteristic information is modeled and optimized using a three-dimensional modeling component and an initialization layout strategy to generate a target three-dimensional model. This target three-dimensional model can intuitively reflect the layout of the circuit, the location of the components, and the relationship between them. The system terminal uses this target three-dimensional model to further analyze and optimize the impedance matching problem of the circuit, thereby improving the transmission efficiency of the RF power supply and the overall performance of the circuit.

[0018] Further, if Figure 2 As shown, the present application provides a method for constructing a target three-dimensional model, including:

[0019] The predetermined circuit characteristics include transmission line length and matching component values; constructing a target initial model of the target RF circuit according to the transmission line length and the matching component values; reading an initialization layout strategy, and performing layout processing on the target initial model based on the initialization layout strategy to obtain the target three-dimensional model.

[0020] Preferably, when constructing the three-dimensional model of the target radio frequency circuit, the system terminal obtains a predetermined circuit feature for feature collection of the target radio frequency circuit. The predetermined circuit feature includes the transmission line length and the matching element value. Among them, the transmission line length refers to the geometric length of the transmission line in the radio frequency circuit. It directly affects the transmission efficiency of the signal and the performance of the circuit. The matching element value refers to the electrical parameter value of the element used to achieve impedance matching in the radio frequency circuit, such as inductance, capacitance, etc. After obtaining the predetermined circuit feature, the system terminal performs feature collection on the target radio frequency circuit according to these predetermined circuit features to obtain the target circuit feature information. Subsequently, the system terminal uses the circuit design software to construct the target initial model of the target radio frequency circuit based on the transmission line length and matching element value in the target circuit feature information. In the target initial model, the transmission line and matching element are placed and connected according to the correct connection mode and layout. Ensure that the various components and connection relationships in the target initial model are accurate to reflect the actual working state and performance of the target radio frequency circuit. Afterwards, read the initialization layout strategy. This strategy includes the initialization output impedance strategy, the initialization boundary condition strategy, the initialization differential equation strategy, and the initialization field component strategy. Based on this initial layout strategy, the system terminal performs layout processing on the target initial model. This process involves setting the output impedance, boundary conditions, electric field and magnetic field components, etc., to ensure that their positions and relationships in three-dimensional space meet the design requirements. Then, the system terminal uses the target initial model after layout processing as the target three-dimensional model. This target three-dimensional model reflects the physical structure and electrical connection relationship of the circuit, providing a basis for subsequent circuit analysis and optimization.

[0021] Furthermore, the present application provides an initialization deployment strategy, including:

[0022] The initialization layout strategy includes an initialization output impedance strategy, an initialization boundary condition strategy, an initialization differential equation strategy and an initialization field component strategy. The initialization boundary condition strategy includes a perfect matching layer. The initialization field component strategy includes an initialization electric field component, an initialization magnetic field component and a predetermined excitation source. The predetermined excitation source refers to a sine wave or a Gaussian pulse.

[0023] Optionally, the initialization layout strategy is of great significance in constructing the target three-dimensional model, which ensures the accuracy and effectiveness of the three-dimensional modeling process. The initialization layout strategy includes the initialization output impedance strategy, the initialization boundary condition strategy, the initialization differential equation strategy and the initialization field component strategy. The system terminal adjusts the circuit element parameters related to the output impedance in the target initial model according to the output impedance value defined in the initialization output impedance strategy to ensure that the output impedance of the circuit matches the expected load impedance. According to the initialization boundary condition strategy, a perfect matching layer (PML) is set in the peripheral area of ​​the target initial model to define the boundary behavior of the three-dimensional modeling area to simulate the infinite space environment and reduce the reflection of electromagnetic waves at the boundary. Among them, the perfect matching layer is a commonly used boundary condition strategy, which can effectively absorb the outward propagating electromagnetic waves and prevent them from reflecting at the boundary, thereby simulating the effect of infinite space. According to the initialization differential equation strategy, the Maxwell equations are discretized using the central difference method to obtain the differential equations about the electric field and magnetic field components, thereby simulating the behavior of the electromagnetic field. Among them, the Maxwell equations describe the electric field, the magnetic field and the relationship between them how they change over time. Then set the initial conditions of the differential equations, such as the initial values ​​of the electric field and magnetic field components. The selection of these initial values ​​is based on the specific three-dimensional modeling scenario and physical conditions. According to the settings in the initialization field component strategy, the initial values ​​are assigned to the electric field and magnetic field components in the target initial model. When initializing the electric field component, the system terminal uses a sine wave or a Gaussian pulse as a predetermined excitation source to set the initial value for the electric field component. Similarly, when initializing the magnetic field component, an initial value is also assigned to it according to the settings in the initialization field component strategy. Under the guidance of the above strategy, the system terminal lays out the target initial model. This includes adjusting the position, connection relationship, parameter setting, etc. of the circuit components to ensure that the target initial model meets the requirements of the initialization layout strategy and generates the target three-dimensional model.

[0024] The target three-dimensional model is meshed to obtain a target meshing result, wherein the target meshing result includes a plurality of meshes with point identifiers.

[0025] In one embodiment, after constructing the target three-dimensional model, the system terminal imports the target three-dimensional model into the mesh processing software, and then sets the relevant parameters of meshing, such as mesh size, mesh type (such as hexahedron, tetrahedron, etc.), mesh quality, etc., according to the characteristics of the target three-dimensional model and the simulation requirements. These parameters will directly affect the accuracy and computational efficiency of the meshing results. Subsequently, the mesh is automatically generated using the mesh processing software. In this process, the mesh processing software divides the model into multiple meshes according to the preset parameters. Each mesh has a unique point identifier, so that the system terminal can accurately track and locate each part of the target three-dimensional model.

[0026] The key point screening strategy is read, and the plurality of grids with point identifiers are analyzed and screened based on the key point screening strategy to obtain a target key grid set.

[0027] In one embodiment, the system terminal first reads the key point screening strategy, which includes a first strategy and a second strategy. The first strategy is used to screen any neighborhood grid set of any matching element in multiple grids with point identifiers. The second strategy is used to screen and obtain a sub-key grid set. The system terminal can form a target key grid set by merging the screening results of the two strategies. This target key grid set contains all grids that are considered key for subsequent analysis and adjustment.

[0028] Further, the present application provides a method for obtaining a target key grid set, including:

[0029] The key point screening strategy includes a first strategy and a second strategy; extracting any matching element in the target RF circuit, and screening any neighborhood grid set of the any matching element in the multiple grids with point identifiers based on the first strategy.

[0030] Preferably, the key point screening strategy includes a first strategy and a second strategy. The first strategy is used to screen a set of grids adjacent to the selected matching element from a plurality of grids with point identifiers. The second strategy is used to screen a set of grids greater than a predetermined distance limit from a plurality of grids with point identifiers excluding the grids screened by the first strategy. For the first strategy, the system terminal first randomly selects an arbitrary matching element from the target RF circuit, and then randomly selects a grid from a plurality of grids with point identifiers as the first grid, and its point is used as the first point. Subsequently, the spatial distance from the first point to the arbitrary matching element is calculated using the Euclidean distance to obtain the first spatial distance. If the first spatial distance is within the predetermined distance threshold range, then the first grid is used as the neighborhood grid of the arbitrary matching element. On the contrary, if the spatial distance exceeds the predetermined distance threshold, the grid does not belong to the neighborhood grid set. The predetermined distance threshold is determined based on historical experience. In summary, the essence of this first strategy is to regard the area around any matching element in the RF circuit as an important monitoring point. By dynamically monitoring the grids around these elements, the performance and status of these areas in circuit operation can be more effectively focused on and analyzed. This approach helps to accurately capture and analyze the interactions between components and their impact on the overall circuit performance.

[0031] Based on the multiple grids with point identifications excluding the arbitrary neighborhood grid set, a secondary screening grid set is formed; based on the second strategy, the secondary screening grid set is screened to obtain a secondary key grid set; the arbitrary neighborhood grid set and the secondary key grid set constitute the target key grid set.

[0032] Optionally, after obtaining any neighborhood grid set, the system terminal removes the grids contained in any neighborhood grid set from multiple grids with point identifiers to form a secondary screening grid set. Subsequently, the secondary screening grid set is further screened using the second strategy. This screening is similar to the above and is also based on a predetermined distance limit. Through the screening of the second strategy, the system terminal obtains a secondary critical grid set. Afterwards, the obtained arbitrary neighborhood grid set is combined with the secondary critical grid set to form a target critical grid set.

[0033] Further, the present application provides a method for obtaining a sub-critical grid set, including:

[0034] A first grid in the secondary screening grid set is randomly extracted; and a spatial distance from the first grid to each grid in the arbitrary neighborhood grid set is calculated in sequence to obtain a plurality of spatial distance values.

[0035] Optionally, the system terminal randomly extracts a grid from the obtained secondary screening grid set as the first grid, and uses the corresponding point as the first point. Then, the spatial distance between the first grid and each grid in any neighborhood grid set is calculated using the same method as above, thereby obtaining multiple spatial distance values. These distance values ​​will be used for subsequent analysis or screening.

[0036] Filter the minimum spatial distance value among the multiple spatial distance values, and determine whether the minimum spatial distance value is greater than a predetermined distance limit; according to the second strategy, if the minimum spatial distance value is greater than the predetermined distance limit, add the first grid to the secondary key grid set.

[0037] Optionally, after obtaining multiple spatial distance values, the system terminal selects the minimum spatial distance value from these spatial distance values, that is, the distance from the first grid to the nearest grid in the neighborhood grid set. Subsequently, it is determined whether this minimum spatial distance value exceeds the predetermined distance limit. According to the provisions of the second strategy, if the minimum spatial distance value is greater than the predetermined distance limit, it means that this first grid maintains a certain distance from all known grids in the neighborhood grid set, so it itself has the potential to become a key grid. Based on this judgment, the system terminal adds this first grid to the secondary key grid set.

[0038] Furthermore, the present application provides for adjusting the dynamic monitoring results of each grid in the arbitrary neighborhood grid set, including:

[0039] When the arbitrary matching element is a nonlinear element, a predetermined nonlinear impedance adjustment coefficient is obtained; and the dynamic monitoring results of each grid in the arbitrary neighborhood grid set are adjusted based on the predetermined nonlinear impedance adjustment coefficient.

[0040] Optionally, when any matching element obtained is a nonlinear element (transistor, diode, etc.), the impedance of these nonlinear elements will change with the change of signal strength. This nonlinear characteristic increases the complexity of circuit matching. In order to cope with this complexity, the system terminal pre-sets a nonlinear impedance adjustment coefficient for the nonlinear element. This coefficient reflects the characteristic that the impedance of the nonlinear element changes with the signal strength, and is derived based on experimental data and theoretical derivation. Subsequently, in the process of dynamically monitoring any neighborhood grid set, that is, the grid set adjacent to the nonlinear element, the system terminal adjusts the monitoring result according to this predetermined nonlinear impedance adjustment coefficient. When the signal strength or impedance in the neighborhood grid is monitored to change, the nonlinear impedance adjustment coefficient is used to correct these changes to more accurately reflect the impact of the nonlinear element on the circuit performance. In this way, the performance of the circuit can be more accurately evaluated and analyzed when processing a radio frequency circuit containing nonlinear elements. The introduction of the nonlinear impedance adjustment coefficient enables the system terminal to better capture the impact of nonlinear elements on circuit performance, thereby improving the accuracy of circuit design and analysis.

[0041] Dynamically monitor the first key grid in the target key grid set to obtain a first grid feature time sequence, and extract the first grid feature at the first step time in the first grid feature time sequence.

[0042] In one embodiment, the system terminal first dynamically monitors the first key grid in the target key grid set. Dynamic monitoring refers to real-time monitoring of the state changes of the grid to obtain detailed information about its performance. When the first key grid is dynamically monitored, the system terminal will obtain a time series data representing the change of its state over time, namely the first grid feature time series. This time series data contains various characteristic values ​​of the first key grid at different time points. Subsequently, the system terminal extracts the grid characteristic value at the first step time from this first grid feature time series as the first grid feature. This first step time refers to the time point when the grid characteristic value is recorded for the first time during the dynamic monitoring process. This time point is set according to the accuracy and needs of the monitoring.

[0043] The first grid feature is analyzed in combination with an impedance fitness evaluation function to obtain a first impedance fitness at the first step time.

[0044] In one embodiment, after the system terminal obtains the first grid features of the first key grid in the target key grid set at the first step time, these features are analyzed in combination with the impedance fitness evaluation function. This impedance fitness evaluation function is defined according to the needs and performance requirements of the circuit design, and is used to evaluate how the grid features affect the impedance matching degree in the circuit. By applying the impedance fitness evaluation function, the first grid features can be converted into a quantitative indicator, namely the first impedance fitness. This indicator reflects the performance of the grid in circuit impedance matching at the first step time.

[0045] Furthermore, the present application provides an impedance fitness evaluation function, including:

[0046] The first grid feature includes a first reflection coefficient, a first transmission coefficient, a first standing wave ratio, a first return loss, a first insertion loss and a first bandwidth, and the first standing wave ratio, the first return loss, the first insertion loss and the first bandwidth constitute a first correction coefficient group, and the expression of the impedance fitness evaluation function is as follows:

[0047] ;

[0048] ;in, Refers to the first key grid The first impedance fitness, and The first key grid The first reflection coefficient and the first transmission coefficient, and are the first reflection coefficients and the first transmission coefficient The weight of , Refers to the first key grid The first correction factor, refers to the first correction coefficient group parameters, refers to the The feedback adjustment coefficient of the parameter, and .

[0049] Preferably, when evaluating the performance of the target key grid, the system terminal combines the first reflection coefficient, the first transmission coefficient, the first standing wave ratio, the first return loss, the first insertion loss and the first bandwidth. These characteristics jointly reflect the performance of the grid in the RF circuit, and the first standing wave ratio, the first return loss, the first insertion loss and the first bandwidth are combined into a first correction coefficient group. In order to obtain a comprehensive impedance fitness evaluation, the system terminal constructs an impedance fitness evaluation function and defines the parameters involved. The impedance fitness evaluation function is as follows:

[0050] ;

[0051] ;in, Refers to the first key grid The first impedance fitness is a quantitative indicator used to evaluate the performance of the mesh in terms of impedance matching. This indicator is calculated based on the reflection coefficient and transmission coefficient. and The first key grid The first reflection coefficient and the first transmission coefficient are the performance evaluation indicators of the RF circuit. They reflect the reflection and transmission of the signal at the grid and are obtained based on dynamic monitoring. and They refer to the weights of the first reflection coefficient and the first transmission coefficient, respectively, and are used to adjust the relative importance of the reflection coefficient and the transmission coefficient in the impedance fitness calculation. The sum of these two weights is equal to 1, which ensures the fairness and rationality of the evaluation. Refers to the first key grid The first correction coefficient is calculated based on the first correction coefficient group. Refers to the first correction coefficient group parameters. It refers to The feedback adjustment coefficients of the parameters are used to correct the reflection coefficient and the transmission coefficient, so as to more comprehensively consider the actual performance of the grid in the circuit. K is equal to 4, which means that the first correction coefficient group contains four parameters. These four parameters represent the performance of the grid in different aspects. For example, the standing wave ratio reflects the reflection of the signal, the return loss and insertion loss reflect the loss of the signal, and the bandwidth reflects the working frequency range of the grid.

[0052] A first impedance fitness time series of the first key grid is constructed according to a first corresponding relationship between the first step time and the first impedance fitness.

[0053] In one embodiment, in order to more comprehensively understand the impedance fitness change of the first key grid during the entire monitoring process, the system terminal combines the first impedance fitness at the first step time with the impedance fitness at other time points to construct a time series data, namely, the first impedance fitness time series of the first key grid. This time series data records the impedance fitness change of the first key grid from the beginning to the end of monitoring, helping the system terminal to more intuitively understand the dynamic performance of the grid performance.

[0054] When the first impedance adaptability timing does not reach the impedance adaptability index constraint, the initialization output impedance of the target RF power source is adjusted.

[0055] In one embodiment, after the first impedance fitness time series of the first key grid is constructed, the system terminal uses a pre-built fitness index prediction model to evaluate the first predicted fitness index of this time series data. Subsequently, it is determined whether this first predicted fitness index meets the pre-set impedance fitness index constraint. The impedance fitness index constraint is determined based on actual needs and engineering specifications to ensure that the target RF power supply can meet certain performance indicators and stability requirements during operation. If the first impedance fitness time series fails to meet this constraint, it means that there may be a problem with the current working state of the target RF power supply and it needs to be adjusted to optimize its performance. In this case, the system terminal adjusts the initialization output impedance of the target RF power supply. The initialization output impedance is an initial value set when the RF power supply is turned on or restarted, which has an important impact on the subsequent working state of the power supply. By adjusting the initialization output impedance, you can try to improve the matching degree between the RF power supply and the target key grid, thereby improving the impedance fitness and enabling the power supply to better adapt to the impedance changes of the grid.

[0056] Furthermore, the present application provides for adjusting the initialization output impedance, including:

[0057] Perform multi-domain feature collection on the first impedance fitness time series to obtain first multi-domain feature parameters; use the first multi-domain feature parameters as input data of a fitness index prediction model to obtain an output result, wherein the output result includes a first predicted fitness index of the first impedance fitness time series; when the first predicted fitness index does not meet the impedance fitness index constraint, adjust the initialized output impedance.

[0058] Preferably, when optimizing the impedance matching between the RF power supply and the key grid, in order to more accurately predict and evaluate the change of impedance fitness, the system terminal collects multi-domain features of the first impedance fitness time series. Multi-domain features include time domain features, frequency domain features, etc., which can fully reflect the characteristics of the impedance fitness time series in different dimensions. By collecting the multi-domain features of the first impedance fitness time series, the system terminal can obtain the first multi-domain feature parameters to help understand the dynamic changes of impedance fitness. Subsequently, these first multi-domain feature parameters are used as input data of the fitness index prediction model. The fitness index prediction model is a neural network model constructed based on historical impedance matching records, which can predict the future impedance fitness index based on the input multi-domain feature parameters. After obtaining the first predicted fitness index, the system terminal compares the first predicted fitness index with the pre-set impedance fitness index constraint. If the first predicted fitness index does not meet the impedance fitness index constraint, it means that the current RF power output impedance setting may not meet the performance requirements. In this case, the system terminal adjusts the initialization output impedance to change the output characteristics of the RF power supply, thereby improving the matching degree between it and the key grid and improving the impedance fitness.

[0059] Furthermore, the present application provides a method for constructing a fitness index prediction model, including:

[0060] The fitness index prediction model refers to an intelligent model obtained by supervised training of data in historical impedance matching records based on the principle of neural network.

[0061] Optionally, when training a fitness index prediction model based on the principle of a neural network, the system terminal first determines the number of network layers, the number of units in each layer, the activation function and other parameters according to the complexity of the problem and the characteristics of the data. Then import a deep learning library (such as TensorFlow or PyTorch), and use the functions and classes in the library to define and construct a fitness index prediction model. Subsequently, the historical impedance matching records are divided into a training set and a test set. The historical impedance matching records include historical multi-domain feature parameters and corresponding historical predicted fitness indexes. The weights and biases of the fitness index prediction model are initialized according to random numbers. After that, the fitness index prediction model is trained using the training set. During the training process, the Adam optimizer, the mean square error loss function and the back propagation algorithm are used to guide the parameter update of the model to minimize the loss function. When the training reaches the maximum number of iterations, the test set is used to evaluate the current fitness index prediction model, and the error, accuracy and other indicators between the predicted fitness index and the historical predicted fitness index are calculated. If the calculated indicators meet the corresponding preset values, the current fitness index prediction model is output. Otherwise, the model is tuned, including adjusting the model's hyperparameters such as learning rate, batch size, number of layers, and number of units, and retraining.

[0062] In summary, the embodiments of the present application have at least the following technical effects:

[0063] The embodiment of the present application constructs a three-dimensional model of the target RF circuit and performs gridding processing on it so as to accurately analyze and control the impedance characteristics in the circuit. Subsequently, based on a specific key point screening strategy, a target key grid set closely related to impedance matching is identified from a large number of grids. These key grids are then dynamically monitored, feature timings are extracted, and impedance fitness is calculated using an impedance fitness evaluation function. Afterwards, based on the real-time feedback of the impedance fitness timing, it is determined whether the output impedance of the RF power supply needs to be adjusted to achieve adaptive impedance matching. In addition, the initialization processing of the model, the adjustment of the impedance of the nonlinear element, and the fitness index prediction model constructed using the principle of neural network are used to predict the future impedance fitness, providing a more scientific and accurate basis for adjustment decisions. These technical effects jointly solve the technical problem of low transmission efficiency caused by the mismatch between the output impedance of the RF power supply and the impedance of the target RF circuit, and realize the effect of improving the accuracy and efficiency of impedance matching through the impedance adaptive matching method, thereby ensuring the transmission efficiency of the RF power supply.

[0064] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0066] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An impedance adaptive matching method for a radio frequency power supply, characterized in that: include: Constructing a target three-dimensional model according to target circuit characteristic information of a target radio frequency circuit collected based on predetermined circuit characteristics, wherein the target radio frequency circuit refers to a circuit including a target radio frequency power supply in the circuit; Performing meshing processing on the target three-dimensional model to obtain a target meshing result, wherein the target meshing result includes a plurality of meshes with point identifiers; Reading a key point screening strategy, and analyzing and screening the plurality of grids with point identifiers based on the key point screening strategy to obtain a target key grid set; Dynamically monitoring a first key grid in the target key grid set to obtain a first grid feature time series, and extracting a first grid feature at a first step time in the first grid feature time series; Analyzing the first grid feature in combination with an impedance fitness evaluation function to obtain a first impedance fitness at the first step time; Constructing a first impedance fitness time series of the first key grid according to a first corresponding relationship between the first step time and the first impedance fitness; When the first impedance fitness time sequence does not reach the impedance fitness index constraint, adjusting the initialization output impedance of the target RF power source; The key point screening strategy includes a first strategy and a second strategy; Extracting any matching element in the target radio frequency circuit, and screening any neighborhood grid set of the any matching element in the plurality of grids with point identifiers based on the first strategy; Based on the plurality of grids with point identifiers excluding the arbitrary neighborhood grid set, forming a secondary screening grid set; Screening the secondary screening grid set based on the second strategy to obtain a secondary key grid set; The arbitrary neighborhood grid set and the secondary key grid set constitute the target key grid set; Randomly extracting the first grid from the secondary screening grid set; sequentially calculating the spatial distance from the first grid to each grid in the arbitrary neighborhood grid set to obtain a plurality of spatial distance values; Screening a minimum spatial distance value among the multiple spatial distance values, and determining whether the minimum spatial distance value is greater than a predetermined distance limit; According to the second strategy, if the minimum spatial distance value is greater than the predetermined distance limit, the first sub-grid is added to the sub-critical grid set.

2. The impedance adaptive matching method of a radio frequency power supply according to claim 1, characterized in that: include: The predetermined circuit characteristics include transmission line length and matching component values; Constructing a target initial model of the target radio frequency circuit according to the transmission line length and the matching component value; An initialization layout strategy is read, and layout processing is performed on the target initial model based on the initialization layout strategy to obtain the target three-dimensional model.

3. The impedance adaptive matching method of a radio frequency power supply according to claim 2, characterized in that: The initialization layout strategy includes an initialization output impedance strategy, an initialization boundary condition strategy, an initialization differential equation strategy and an initialization field component strategy. The initialization boundary condition strategy includes a perfect matching layer. The initialization field component strategy includes an initialization electric field component, an initialization magnetic field component and a predetermined excitation source. The predetermined excitation source refers to a sine wave or a Gaussian pulse.

4. The impedance adaptive matching method of a radio frequency power supply according to claim 1, characterized in that: The first grid feature includes a first reflection coefficient, a first transmission coefficient, a first standing wave ratio, a first return loss, a first insertion loss and a first bandwidth, and the first standing wave ratio, the first return loss, the first insertion loss and the first bandwidth constitute a first correction coefficient group, and the expression of the impedance fitness evaluation function is as follows: ； ； in, Refers to the first key grid The first impedance fitness, and The first key grid The first reflection coefficient and the first transmission coefficient, and are the first reflection coefficients and the first transmission coefficient The weight of , Refers to the first key grid The first correction factor, refers to the first correction coefficient group parameters, refers to the The feedback adjustment coefficient of the parameter, and .

5. The impedance adaptive matching method of a radio frequency power supply according to claim 1, characterized in that: include: Collecting multi-domain features of the first impedance fitness time series to obtain first multi-domain feature parameters; Using the first multi-domain characteristic parameter as input data of a fitness index prediction model to obtain an output result, wherein the output result includes a first predicted fitness index of the first impedance fitness time series; When the first predicted fitness index does not reach the impedance fitness index constraint, the initialized output impedance is adjusted.

6. The impedance adaptive matching method of a radio frequency power supply according to claim 5, characterized in that: The fitness index prediction model refers to an intelligent model obtained by supervised training of data in historical impedance matching records based on the principle of neural network.

7. The impedance adaptive matching method of a radio frequency power supply according to claim 1, characterized in that: Also includes: When the arbitrary matching element is a nonlinear element, obtaining a predetermined nonlinear impedance adjustment coefficient; The dynamic monitoring results of each grid in the arbitrary neighborhood grid set are adjusted based on the predetermined nonlinear impedance adjustment coefficient.

Images