A filtering method of a charging module based on a variable capacitance array and related apparatus

Abstract

The application discloses a filtering method of a charging module based on a variable-capacitance array and related devices, which is used for adjusting the equivalent impedance parameters of a filtering loop in real time, improving the dynamic performance of the system under the premise of ensuring the steady-state filtering effect. The method includes: performing frequency domain analysis on an output voltage signal to extract output ripple spectrum characteristics; identifying frequency components with amplitudes greater than a preset threshold at the current time according to the output ripple spectrum characteristics, and determining the frequency component with the maximum amplitude as the dominant noise frequency; based on the dominant noise frequency, calculating a target topology combination that makes the filtering loop composed of a filtering inductor and a variable-capacitance array present a target impedance characteristic at the dominant noise frequency; according to the target topology combination, controlling the on-off state of a controlled switching unit, and controlling a reconfigurable switching network to switch and connect the variable-capacitance array to the corresponding circuit node, and reconfiguring the variable-capacitance array into a filtering network with the target impedance characteristic.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of filtering of charging modules based on variable capacitance arrays, and in particular to a filtering method for a charging module based on a variable capacitance array and related apparatus. BACKGROUND

[0002] The main function of a charging module is to convert grid power into direct current that meets the needs of a load. Due to the use of high-frequency switching power supply technology inside the charging module, the high-speed on-off of power devices will generate voltage ripple containing rich high-order harmonics at the output end. Therefore, it is essential to configure a filter circuit at the output end of the charging module, which aims to filter out high-frequency noise and reduce the output voltage ripple coefficient, so as to ensure that the output power quality meets the battery charging specification and prevent electromagnetic interference from causing damage to the load.

[0003] In existing charging module designs, the output filter circuit usually uses electrolytic capacitors or film capacitors with fixed capacitance values and inductors to form LC or CLC filter networks. The fixed inductor and capacitor parameters are usually calculated and selected according to the rated maximum power of the charging module, the minimum switching frequency, and the maximum ripple voltage peak-to-peak value allowed in the design index.

[0004] However, in actual applications, the charging process usually includes different stages such as constant current CC and constant voltage CV, and the load impedance and ripple characteristics of different stages differ greatly. The existing filter with fixed parameters is designed according to the maximum ripple condition. In the large current stage, a large capacitor is needed to suppress fluctuations, while in the constant voltage and small current stage, the excessive fixed capacitor will reduce the control loop bandwidth of the system, resulting in slower dynamic response speed, and the steady-state filtering effect and dynamic performance cannot be balanced. SUMMARY

[0005] The present application provides a filtering method for a charging module based on a variable capacitance array and related apparatus for real-time adjustment of the equivalent impedance parameters of the filter circuit, which improves the dynamic performance of the system while ensuring the steady-state filtering effect.

[0006] The first aspect of the present application provides a filtering method for a charging module based on a variable capacitance array, the charging module comprising a filter inductor and a variable capacitance array arranged on the periphery of the filter inductor, the variable capacitance array being composed of a plurality of capacitor branches with different impedance characteristics, each capacitor branch having a controlled switch unit connected in series, and the variable capacitance array being connected to the circuit node of the charging module through a reconfigurable switch network; the filtering method comprising:

[0007] acquiring an output voltage signal of the output end of the charging module at a preset high-frequency sampling rate, and performing frequency domain analysis on the output voltage signal to extract the output ripple spectrum characteristics;

[0008] According to the output ripple spectrum, a frequency component with an amplitude greater than a preset threshold at a current time is identified, and a frequency component with the largest amplitude is determined as a dominant noise frequency;

[0009] Based on the dominant noise frequency, a target topology combination is calculated to make the filter loop composed of the filter inductor and the variable capacitance array exhibit a target impedance characteristic at the dominant noise frequency;

[0010] According to the target topology combination, the on-off state of the controlled switch unit is controlled, and the reconfigurable switch network is controlled to switch the variable capacitance array to a corresponding circuit node, and the variable capacitance array is reconfigured into a filter network with the target impedance characteristic to filter out the ripple component corresponding to the dominant noise frequency.

[0011] Optionally, the variable capacitance array includes a low ESR branch using a thin film capacitor, a high ESR branch using a ceramic capacitor in series with a damping resistor, and a notch branch using a capacitor and an inductor in series.

[0012] The target topology combination is calculated based on the dominant noise frequency to make the filter loop composed of the filter inductor and the variable capacitance array exhibit a target impedance characteristic at the dominant noise frequency, including:

[0013] According to the relative relationship between the dominant noise frequency and the resonance frequency of the filter loop, a target impedance characteristic is determined, including a low impedance characteristic, a high impedance characteristic, and a notch characteristic;

[0014] According to the target impedance characteristic, one or more of the low ESR branch, the high ESR branch, and the notch branch are selected for combination to construct a target topology combination.

[0015] Optionally, according to the target impedance characteristic, one or more of the low ESR branch, the high ESR branch, and the notch branch are selected for combination to construct a target topology combination, including:

[0016] A three-dimensional state space including a capacitance branch type, a circuit node position, and a switch combination state is constructed.

[0017] A comprehensive evaluation model is established with a ripple attenuation rate as a benefit function and a transient voltage disturbance amount generated by a switching action as a cost function.

[0018] In the three-dimensional state space, a candidate subset conforming to the target impedance characteristic is screened out, and each combination in the candidate subset is substituted into the comprehensive evaluation model for calculation. The combination that maximizes the weighted difference between the benefit function and the cost function is determined as the target topology combination.

[0019] Optionally, the filtering method further comprises:

[0020] When the target topology combination contains the trap branch, determining an equivalent capacitance value corresponding to the trap branch in the current switch combination state, the equivalent capacitance value being an equivalent capacity presented by the trap branch in the current switch combination state;

[0021] On the basis of the equivalent capacitance value, applying positive and negative alternating discrete capacitance perturbation by changing the combination state of the controlled switch unit, so that the actual resonance frequency of the filter circuit is dithered within a preset fine-tuning range;

[0022] During the application of the discrete capacitance perturbation, the rate of change of the ripple amplitude corresponding to the dominant noise frequency is monitored synchronously, and the gradient direction of the ripple amplitude with respect to the capacitance change is calculated;

[0023] The target topology combination is dynamically updated according to the gradient direction.

[0024] Optionally, the filtering method further comprises:

[0025] The ripple amplitude corresponding to the dominant noise frequency is continuously monitored, and if it is monitored that the ripple amplitude corresponding to the dominant noise frequency presents an upward trend after the variable capacitance array is reconstructed, it is determined that the current target topology combination forms a parallel resonance with the load impedance;

[0026] The target impedance characteristic is updated from the current characteristic to a high impedance characteristic, the target topology combination containing the high ESR branch is reconstructed, and switching is performed.

[0027] Optionally, the circuit node includes a first node located at the output side of the filter inductor, a second node located at the input side of the filter inductor, and a third node connected in parallel to the power switch unit of the charging module;

[0028] The control of the reconfigurable switch network to switch the variable capacitance array to the corresponding circuit node comprises:

[0029] When the dominant noise frequency is a low-frequency ripple and the amplitude is less than a first threshold value, it is determined to be a steady-state output working condition, and the reconfigurable switch network is controlled to be connected to the first node to form an LC low-pass filtering topology with the filter inductor to smooth the output voltage;

[0030] When the dominant noise frequency contains high-frequency interference and the frequency exceeds a second threshold value, it is determined to be a conducted interference working condition, and the reconfigurable switch network is controlled to be connected to the second node to form a CLCπ type filtering topology with the filter inductor and the output capacitance inherent to the charging module to improve high-frequency insertion loss;

[0031] When the rate of change of the drain-source voltage of the power switch unit is detected to exceed a safety threshold, it is determined that a voltage stress condition exists, and the reconfigurable switch network is preferentially controlled to connect to the third node to build a voltage clamping buffer topology to absorb voltage spikes generated by the operation of the power switch unit.

[0032] Optionally, the filtering method further comprises:

[0033] In the process of controlling the on-off state of the controlled switch unit, the phase of the branch current flowing through the controlled switch unit is monitored in real time, and the disconnection action is preferentially triggered at the moment when the phase of the branch current is at zero crossing;

[0034] In the process of controlling the reconfigurable switch network to switch the variable capacitance array to connect to the corresponding circuit node, the end voltage of the variable capacitance array and the end voltage of the target circuit node are monitored before the variable capacitance array is connected to the new circuit node.

[0035] If the voltage difference between the two exceeds a preset safety range, the controlled switch unit is first controlled to perform pre-charging or pre-discharging in a high-frequency PWM modulation manner until the voltage difference falls within the safety range, and then the variable capacitance array is connected to the new circuit node.

[0036] The second aspect of the present application provides a filtering system of a charging module based on a variable capacitance array, the charging module comprising a filter inductor and a variable capacitance array arranged on the periphery of the filter inductor, the variable capacitance array being composed of a plurality of groups of capacitance branches with different impedance characteristics, each of the capacitance branches being connected in series with a controlled switch unit, and the variable capacitance array being connected to a circuit node of the charging module through a reconfigurable switch network; the filtering system comprises:

[0037] A collection unit is configured to collect an output voltage signal of an output end of the charging module at a preset high-frequency sampling rate, and perform frequency domain analysis on the output voltage signal to extract output ripple spectrum characteristics;

[0038] An identification unit is configured to identify a frequency component with an amplitude greater than a preset threshold at a current time according to the output ripple spectrum characteristics, and determine a frequency component with the largest amplitude as a dominant noise frequency;

[0039] A calculation unit is configured to calculate a target topology combination that makes a filtering loop composed of the filter inductor and the variable capacitance array exhibit a target impedance characteristic at the dominant noise frequency based on the dominant noise frequency;

[0040] The control unit is configured to control on-off states of the controlled switch units according to the target topology combination, and control the reconfigurable switch network to switch the variable capacitance array to corresponding circuit nodes, and reconfigure the variable capacitance array into a filter network with the target impedance characteristic to filter out a ripple component corresponding to the dominant noise frequency.

[0041] The third aspect of the present application provides a filter device of a charging module based on a variable capacitance array, the device comprising:

[0042] a processor, a memory, an input-output unit, and a bus;

[0043] The processor is connected to the memory, the input-output unit, and the bus;

[0044] The memory stores a program, and the processor invokes the program to execute the filter method of the charging module based on the variable capacitance array according to the first aspect and any optional embodiment of the first aspect.

[0045] The fourth aspect of the present application provides a computer-readable storage medium, which stores a program, and the program executes the filter method of the charging module based on the variable capacitance array according to the first aspect and any optional embodiment of the first aspect when executed on a computer.

[0046] As can be seen from the above technical solutions, the present application has the following advantages:

[0047] By collecting voltage signals in real time and performing frequency domain analysis, the dominant noise frequency can be quantitatively identified. The target impedance characteristic determined according to the dominant noise frequency and the target topology combination further calculated are variables dynamically generated according to the actual noise demand at the current time. Finally, the on-off states of the controlled switch units are controlled according to the target topology combination, and the reconfigurable switch network is controlled to switch the variable capacitance array to corresponding circuit nodes, and reconfigure the variable capacitance array into a filter network with the target impedance characteristic. That is, by changing the series-parallel structure inside the variable capacitance array and its access position in the circuit, the equivalent impedance parameters of the filter circuit are adjusted in real time. This mechanism enables the filter network to present the optimal impedance for a specific noise frequency, ensuring effective filtering of the dominant noise frequency while meeting the steady-state output indicators, and minimizing the impact of the filter network on the system control loop bandwidth, thereby maximizing the dynamic response speed of the system. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the present application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0049] Figure 1 An embodiment flowchart of a filtering method of a charging module based on a variable-capacitance array provided by the present application;

[0050] Figure 2 An embodiment flowchart of calculating a target topology combination in the filtering method of the charging module based on the variable-capacitance array provided by the present application;

[0051] Figure 3 An embodiment flowchart of determining a target topology combination based on a comprehensive evaluation model in the filtering method of the charging module based on the variable-capacitance array provided by the present application;

[0052] Figure 4 An embodiment flowchart of switching the variable-capacitance array to be connected to different circuit nodes in the filtering method of the charging module based on the variable-capacitance array provided by the present application;

[0053] Figure 5 An embodiment flowchart of a soft switching control strategy in the filtering method of the charging module based on the variable-capacitance array provided by the present application

[0054] Figure 6 An embodiment structural diagram of a filtering system of a charging module based on a variable-capacitance array provided by the present application;

[0055] Figure 7 An embodiment structural diagram of a filtering device of a charging module based on a variable-capacitance array provided by the present application. DETAILED DESCRIPTION

[0056] The present application provides a filtering method of a charging module based on a variable-capacitance array and related devices, which are used for adjusting the equivalent impedance parameters of a filtering loop in real time, and improve the dynamic performance of the system under the premise of ensuring the steady-state filtering effect.

[0057] To facilitate the understanding of the basic structure and implementation idea of the present application, a schematic structural description is provided as follows: In the present application, the charging module includes a filter inductor arranged at the output side of the power conversion stage, and a variable capacitance array arranged at the periphery of the filter inductor. The variable capacitance array can be composed of multiple groups of capacitance branches, each of which can have different equivalent impedance characteristics in structure, such as a capacitance branch with low equivalent series resistance, a capacitance branch with damping characteristics, or a capacitance-inductance composite branch for forming a notch effect, etc. To achieve flexible scheduling of each capacitance branch, a controlled switch unit such as a MOSFET or a relay can be arranged in series with each capacitance branch for controlling the on-off state of the branch. In order to enable the variable capacitance array to participate in the filter circuit in multiple ways, the present application is also provided with a reconfigurable switch network, which enables each capacitance branch to be selectively connected to different circuit nodes of the charging module according to the control requirements. Through the above structure, the present application can dynamically adjust the topology combination and connection position of the variable capacitance array during operation, thereby forming a filter network for specifically filtering out the ripple of a specific frequency band or suppressing the switching peak. It should be noted that the above structural description is only used to help understand the basic concept of the present application, and does not limit the technical solutions of the present application. The specific branch form, branch number, switch unit type and specific implementation method of the reconfigurable switch network of the variable capacitance array can be adjusted and extended according to actual requirements, as long as the spirit and essence of the present application are not deviated, they should be considered to fall within the protection scope of the present application.

[0058] Please refer to Figure 1 , Figure 1 An embodiment of the filtering method of the charging module based on the variable capacitance array provided in the present application is provided, which comprises:

[0059] 101, collecting the output voltage signal of the output end of the charging module at a preset high-frequency sampling rate, and performing frequency domain analysis on the output voltage signal to extract the output ripple spectrum characteristics;

[0060] During normal operation of the charging module, the power conversion unit is affected by load mutation, power grid interference or change of the working state of the switching device, and the output end usually superimposes ripple and noise components of different sources. In order to accurately judge the frequency distribution of these components, the output voltage signal needs to be collected based on a high-frequency sampling rate. Thereafter, frequency domain analysis is performed on the collected output voltage signal sequence, which can be realized by discrete frequency spectrum decomposition of the sampling data, so as to identify the energy strength and distribution of different frequency components. Through the frequency domain analysis process, the main frequency components contained in the output ripple at a certain moment can be obtained, such as the switching main frequency, the resonance frequency and its multiple components, i.e. the output ripple spectrum characteristics are obtained.

[0061] The spectrum feature will dynamically change when the charging module experiences different working conditions. For example, the high-frequency peak component can be enhanced when the device temperature rise causes the switching speed to decrease; the low-frequency ripple can be temporarily increased when the load experiences a step change. The spectrum analysis result provides a basis for subsequent filter topology switching.

[0062] 102. Identify the frequency component with an amplitude greater than a preset threshold at the current time based on the output ripple spectrum feature, and determine the frequency component with the largest amplitude as the dominant noise frequency;

[0063] After completing the frequency domain analysis of the output voltage, the amplitude distribution curve of each frequency point of the output ripple in a period of time can be obtained. Due to the complex actual working environment of the charging module, there are usually low-frequency voltage fluctuations, switching fundamental waves generated by power switching devices, and high-frequency resonance peaks caused by device parasitic parameters in the spectrum. Different types of ripples have different effects on output quality and device stress, so it is necessary to identify the frequency component that causes main interference to the system.

[0064] In this embodiment, first, the amplitudes of each frequency point in the spectrum are compared one by one based on a preset threshold of the amplitude. The preset threshold is generally set based on the output ripple characteristics allowed by the charging module, electromagnetic compatibility requirements, and the safe working range of the switching device, and is not limited here. For example, in an electric vehicle charging module, if the output end ripple amplitude exceeds a certain proportion (such as 1% of the rated output voltage), it may affect the stability of constant voltage or constant current control, so this proportion can be used as a reference standard to set the amplitude threshold. By filtering out frequency components below the threshold, frequency components with less impact on the output can be effectively filtered out, so that subsequent processing can focus on key issues.

[0065] After screening all frequency components with amplitudes exceeding the threshold, the frequency component with the largest amplitude is determined as the dominant noise frequency at the current time. The dominant noise frequency usually best reflects the main noise source affecting the output quality or device stress, i.e., it reflects the noise component that most affects the output quality. For example, during the large current climbing stage, low-frequency ripple can be dominant; while in a high-temperature or parasitic parameter sensitive environment, high-frequency peaks can become the dominant noise frequency.

[0066] 103. Based on the dominant noise frequency, calculate the target topology combination that makes the filter circuit composed of a filter inductor and a variable capacitor array exhibit a target impedance characteristic at the dominant noise frequency;

[0067] After identifying the dominant noise frequency that has the greatest impact on the output quality at the current moment, the target topology combination is calculated, which makes the filter circuit composed of the filter inductance and the variable capacitance array present a target impedance characteristic at the dominant noise frequency. The target impedance characteristic refers to the electrical characteristic that the filter circuit should present at a certain specific frequency, such as low impedance absorption, impedance notch, or presenting a specific attenuation slope, etc. In this embodiment, the target impedance characteristic generally presents a significant impedance concave at the dominant noise frequency, so that the noise energy is preferentially absorbed by the filter branch to reduce the ripple at the output end. The target topology combination refers to the specific hardware connection scheme of the on-off state of each capacitance branch in the variable capacitance array and the access node of the variable capacitance array in the charging module circuit, in order to physically realize the above-mentioned target impedance characteristic.

[0068] In actual operation scenarios, the calculation of the target topology combination can be understood as on-demand deployment of capacitance branches with different capacity levels, ESR and ESL differences. For example, when the charging module works in a steady state and the dominant noise frequency is low, the filter circuit needs a large equivalent capacitance to weaken the large amplitude voltage fluctuation, and the target topology combination usually contains more parallel branches to build a low-frequency high-capacity filter channel; on the contrary, when the dominant noise frequency increases or the system is in a dynamic adjustment process, if the capacitive impedance is maintained too large, it will limit the system control loop bandwidth and affect the dynamic response speed. At this time, it is more necessary to control the resonance point position or reduce the influence of capacitive load on the control bandwidth, so the target topology combination may reduce the parallel branches or select a more appropriate access node, so as to release the dynamic control ability of the system while ensuring that the noise at a specific frequency is filtered out.

[0069] Through the above calculation, the hardware connection scheme that can make the filter circuit present the optimal impedance characteristic at the dominant noise frequency can be obtained.

[0070] 104、According to the target topology combination, the on-off state of the controlled switch unit is controlled, and the reconfigurable switch network is controlled to switch and connect the variable capacitance array to the corresponding circuit node, so as to reconfigure the variable capacitance array into a filter network with a target impedance characteristic, so as to filter out the ripple component corresponding to the dominant noise frequency.

[0071] After calculating the theoretically optimal hardware connection scheme, the hardware circuit is reconfigured, and the target topology combination is converted into the actual circuit connection state. This process involves two-dimensional cooperative control: reconfiguration of the internal parameters of the variable capacitance array and reconfiguration of the external access position of the variable capacitance array, which will be described below:

[0072] For the reconstruction of the internal parameters of the variable capacitance array, specifically, the controlled switch units in series in each capacitance branch are sent drive signals according to the instructions in the target topology combination, and in the actual running scene, the on-off state of the physical circuit changes. For example, when the filtering capability needs to be enhanced, multiple groups of branches can be turned on to increase the equivalent capacitance; when the dynamic response needs to be improved or the resonance characteristics need to be changed, part of the branches can be turned off.

[0073] For the reconstruction of the external access position of the variable capacitance array, specifically, the reconfigurable switch network is operated to change the connection node between the variable capacitance array and the main circuit of the charging module, so that it is in a topology position suitable for the current noise characteristics in the circuit. In the prior art, the filter capacitor is usually permanently welded at the output port. In the present embodiment, the reconfigurable switch network can physically switch the variable capacitance array from one circuit node (such as the back stage of the inductor) to another circuit node (such as the front stage of the inductor or near the power device), substantially changing the topology order and structure form of the filter, so as to provide the most targeted suppression effect for different properties of the dominant noise frequency.

[0074] Through the coordinated control of the above two dimensions, the variable capacitance array can be reconstructed into a filter network with low impedance or matching target impedance at the dominant noise frequency, thereby effectively shunting or attenuating the ripple current of the frequency band.

[0075] In the present embodiment, by real-time acquisition of voltage signals and frequency domain analysis, the current dominant noise frequency can be quantitatively identified, and the target impedance characteristics determined according to the dominant noise frequency and the further calculated target topology combination are variables dynamically generated according to the actual noise demand at the current time. Finally, the on-off state of the controlled switch unit is controlled according to the target topology combination, and the reconfigurable switch network is controlled to switch and connect the variable capacitance array to the corresponding circuit node, so as to reconstruct the variable capacitance array into a filter network with target impedance characteristics. That is, by changing the series-parallel structure of the variable capacitance array and its access position in the circuit, the equivalent impedance parameters of the filter loop are adjusted in real time. This mechanism enables the filter network to present the optimal impedance for a specific noise frequency, ensuring effective filtering of the dominant noise frequency while meeting the steady-state output indicators, and minimizing the impact of the filter network on the system control loop bandwidth, thereby maximizing the dynamic response speed of the system.

[0076] In some specific embodiments, in order to achieve precise suppression of different frequency bands and different properties of noise, the variable capacitance array includes a low ESR branch using a thin film capacitor, a high ESR branch using a ceramic capacitor in series with a damping resistor, and a notch branch using a capacitor and an inductor in series; for the corresponding hardware structure, please refer to Figure 2 , Figure 2One embodiment of calculating the target topology combination in the filtering method of the variable-capacitance array-based charging module provided in the present application comprises:

[0077] 201. Determine the target impedance characteristic according to the relative relationship between the dominant noise frequency and the resonant frequency of the filter circuit, the target impedance characteristic including a low impedance characteristic, a high impedance characteristic and a notch characteristic;

[0078] After identifying the dominant noise frequency, read the current inherent parameters of the filter circuit, i.e. the inductance of the inductor and the current base capacitance value, and calculate the inherent resonant frequency point of the filter circuit according to the same. Compare the real-time identified dominant noise frequency with the inherent resonant frequency, and according to the physical meaning of the comparison result, decide the target impedance characteristic required at present. The target impedance characteristics in different scenarios are described as follows:

[0079] Scenario one: low impedance characteristic, used for steady-state smoothing. When the identified dominant noise frequency is in the low frequency band (for example, several tens of hertz to several hundred hertz of power grid ripple), and the frequency is far lower than the inherent resonant frequency of the filter circuit, it is determined that the current is in the normal steady-state filtering working condition. In this scenario, the filter network should maintain a low capacitive impedance in a wide frequency range, so that the low-frequency ripple current can be fully discharged, thereby realizing the suppression of output voltage fluctuation. At this time, the determined target impedance characteristic is the low impedance characteristic.

[0080] Scenario two: high impedance characteristic, used for suppressing resonance. When the identified dominant noise frequency is close to or equal to the inherent resonant frequency of the filter circuit, it is determined that there is a high risk of parallel resonance at present. In the resonant state, the quality factor Q value of the LC circuit will increase, and a small excitation may cause oscillation, threatening the safety of the device. Therefore, the quality factor needs to be reduced by increasing the equivalent damping to make the voltage oscillation decay quickly. At this time, the target impedance characteristic is determined as the high impedance characteristic, i.e. the filter network introduces appropriate resistive components on the basis of the capacitive component to effectively weaken the potential resonance.

[0081] Scenario three: notch characteristic, used for specific frequency point suppression. When the dominant noise frequency is the high-order harmonic of the switching frequency or a specific high-frequency peak, and the frequency is much higher than the inherent resonant frequency of the filter circuit, it is determined that there is high-frequency interference concentrated in a narrow band area at present. The traditional low-pass structure is difficult to achieve sufficient attenuation in this frequency band. Therefore, the target impedance characteristic is determined as the notch characteristic, i.e. the filter network forms an impedance zero point at this frequency, so that the noise current flows into the notch branch preferentially, realizing effective suppression of the specific interference frequency point.

[0082] 202. Select one or more from the low-ESR branch, the high-ESR branch and the notch branch according to the target impedance characteristic to combine to build the target topology combination.

[0083] According to the target impedance characteristic type determined in step 201, a target topology combination under the current working condition is constructed by selecting from three heterogeneous branches. The specific scheme is as follows:

[0084] Low impedance characteristic selects low ESR branch: The low ESR branch contains a capacitor unit with small equivalent series resistance and large capacity, which can form a low impedance path in the low frequency band and is beneficial to suppress the output ripple amplitude. Therefore, when the target is low impedance characteristic, the low ESR branch is selected to be turned on to improve the equivalent capacity of the filter network and smooth the output voltage.

[0085] High impedance characteristic introduces high ESR branch: The high ESR branch has a resistance element with a proper resistance value in series, which can provide a necessary energy dissipation path when close to the resonance frequency, thereby reducing the loop quality factor. When the target is high impedance characteristic, the high ESR branch is added to the target topology combination to suppress resonance or overshoot.

[0086] Notch characteristic selects notch branch: The notch branch is composed of an inductor and a capacitor in series, and its inherent series resonance point can be designed to match the dominant noise frequency. When the branch is connected, a low impedance path is formed at a specific frequency, which can reduce high frequency spikes. Therefore, when the target is notch characteristic, the corresponding notch unit with the resonance point is selected according to the dominant noise frequency to connect the circuit, realizing directional filtering of the specified frequency noise.

[0087] Further, in step 202 of constructing the target topology combination, not only the filtering effect but also the safety of the switching action itself must be considered. If the pressure difference of the capacitor branch is forcibly switched to pursue the ultimate filtering effect, a serious inrush current or voltage spike may be caused, which may even impact the system. Please refer to Figure 3 , Figure 3 An embodiment of the method for determining the target topology combination based on the comprehensive evaluation model in the filter method of the charging module based on the variable capacitance array provided in the present application comprises the following steps:

[0088] 301. Construct a three-dimensional state space containing the type of capacitor branch, the position of circuit node, and the state of switch combination;

[0089] There may be dozens or even hundreds of possible connection states in the actual hardware of the variable capacitance array. In order to simplify the calculation process, a complete hardware state mapping library, i.e. a three-dimensional state space, can be established. The first dimension describes the physical properties of the capacitor branch, i.e. whether the branch belongs to low ESR film capacitor, high ESR damping capacitor or LC notch unit; the second dimension describes the physical position of the branch connected through the reconfigurable switch network; and the third dimension describes the binary on-off state of each controlled switch unit.

[0090] 302、establish a comprehensive evaluation model with ripple attenuation rate as the benefit function and the transient voltage disturbance caused by switching action as the cost function;

[0091] In this embodiment, a double-target evaluation model containing benefits and costs is established, in which the benefit function (ripple attenuation rate) is used to quantify the filtering ability of a certain topology combination to the dominant noise frequency. The larger the value, the smaller the residual ripple at the output end under the same noise excitation, and the better the filtering effect; the cost function (transient voltage disturbance) is used to quantify the electrical impact that may be caused when switching from the current state to the target state. Specifically, according to the difference between the current capacitor voltage and the target node voltage, the voltage drop or peak amplitude that may be caused at the switching moment is estimated. The larger the difference, the higher the cost.

[0092] 303、In the three-dimensional state space, filter out the candidate subset that meets the target impedance characteristics, and calculate each combination in the candidate subset by substituting it into the comprehensive evaluation model. The combination that maximizes the weighted difference between the benefit function and the cost function is determined as the target topology combination.

[0093] According to the three-dimensional state space and the comprehensive evaluation model, the final optimization calculation is performed. In order to improve the calculation efficiency, first, a rough screening will be carried out, that is, according to the target impedance characteristics determined in the previous step, those combinations that do not meet the characteristics are quickly eliminated in the three-dimensional state space, thereby locking a smaller candidate subset. Then, the candidate subset is calculated by substituting each candidate combination in the candidate subset into the comprehensive evaluation model to calculate the expected benefit (how much the ripple can be reduced) and the expected cost (how big the switching impact is) respectively. The difference between the two is calculated by a weighting algorithm (i.e. comprehensive score = benefit × weight A - cost × weight B). The weight coefficients A and B can be dynamically adjusted according to the current working condition, for example, increasing the weight B of the cost function when the system is fragile.

[0094] Finally, compare the comprehensive scores of all candidate combinations, and select the one with the highest score as the final target topology combination. For example, suppose there are two candidate combinations A and B that can filter out noise. Combination A can filter out 99% of the ripple, but it needs to switch a large-capacitance capacitor with a high voltage difference, resulting in a high cost function; combination B can filter out 95% of the ripple, but its voltage is close to the current node, resulting in a smooth switching and a very low cost function. In this case, the comprehensive evaluation model may determine that the weighted difference of combination B is larger, thereby abandoning the aggressive combination A and selecting the more robust combination B.

[0095] In this embodiment, not only can the theoretical best hardware combination be found by traversing the optimization, but also the electrical impact risk caused by forcibly switching a high-voltage difference to pursue filtering indicators can be predicted and avoided through the cost function, thereby significantly improving the safety and robustness of the filtering network reconstruction process while ensuring the output power quality.

[0096] In some specific embodiments, an automatic calibration mechanism based on closed-loop feedback is introduced for technical characteristics that are extremely sensitive to element parameter accuracy of the notch branch. This is used to solve the technical problem of the inductance or capacitance drifting due to device aging and environmental temperature changes, thereby causing the notch frequency to deviate from the dominant noise frequency. The steps specifically include: when the target topology combination contains a notch branch, determining the equivalent capacitance value of the notch branch corresponding to the current switch combination state, which is the equivalent capacity presented by the notch branch in the current switch combination state; on the basis of the equivalent capacitance value, by changing the combination state of the controlled switch unit, a positive and negative alternating discrete capacitance disturbance is applied to make the actual resonance frequency of the filter circuit jitter within the preset fine-tuning range; during the application of the discrete capacitance disturbance, the rate of change of the ripple amplitude corresponding to the dominant noise frequency is monitored synchronously, and the gradient direction of the ripple amplitude relative to the capacitance change is calculated; the target topology combination is dynamically updated according to the gradient direction.

[0097] In this embodiment, after confirming that the target topology combination currently accesses the LC series notch branch, first read the on-off state vector of the current controlled switch unit, and analyze the current equivalent capacitance value of the notch branch at the current time in theory in combination with the pre-stored capacitance array parameter library. In order to find the best resonance point in physics, the auxiliary switch unit with the smallest capacitance step in the variable capacitance array needs to be controlled to perform fast on-off action, alternately superimposing a positive small capacitance value and a negative small capacitance value on the current reference equivalent capacitance value. From the physical phenomenon, this action will force the actual resonance frequency of the LC filter circuit to jitter slightly on both sides of the theoretical center frequency. For example, if the theoretical resonance point is 200 kHz, through the disturbance, the actual resonance point will scan back and forth within the fine-tuning range of 198 kHz to 202 kHz. The voltage ripple amplitude of the output end at the dominant noise frequency is captured during the frequency jitter. By comparing the ripple amplitude difference between the two times of applying positive and negative disturbances in real time, the gradient direction is calculated, which reflects the logical relationship between the change of capacitance value and the filtering effect. If increasing the capacitance causes the ripple amplitude to decrease, it means that the actual best resonance point is located to the right of the current capacitance value, i.e. a larger capacitance is needed; if increasing the capacitance causes the ripple amplitude to increase, it means that the actual best resonance point is located to the left of the current capacitance value, i.e. a smaller capacitance is needed; if the ripple amplitude increases regardless of increasing or decreasing the capacitance, it means that the current capacitance value is already at the best point. Based on the calculated gradient direction, the target topology combination can be updated, and the switch state of the variable capacitance array is switched to the direction of reducing the ripple, i.e. the reference capacitance value is corrected. This process is performed in a cyclic iteration manner until the gradient tends to zero or oscillates around zero. Through this process, the actual resonance frequency of the filter circuit can be locked at the physical extreme point with the minimum noise amplitude.

[0098] In some specific embodiments, in order to prevent system instability caused by improper parameter matching during dynamic reconstruction, a resonance suppression strategy based on negative feedback mechanism can also be introduced. This is used to solve the technical problem of abnormal amplification of output ripple when the filter network parameters and the load impedance accidentally satisfy the parallel resonance condition. The specific steps include: continuously monitoring the change of the ripple amplitude corresponding to the dominant noise frequency, if it is monitored that the ripple amplitude corresponding to the dominant noise frequency presents an upward trend after the reconstruction of the variable capacitance array, it is determined that the current target topology combination forms a parallel resonance with the load impedance; the target impedance characteristic is updated from the current characteristic to a high impedance characteristic, the target topology combination containing a high ESR branch is reconstructed and switching is performed.

[0099] In this embodiment, after the hardware reconstruction of the filter circuit is completed, the voltage ripple amplitude at the dominant noise frequency can also be continuously tracked. Under normal circumstances, the ripple amplitude after reconstruction should present a downward trend. However, due to the complex complex impedance characteristics of the load end of the charging module, and the fact that this impedance dynamically changes with the working condition, there is a small probability but high risk working condition: the newly connected capacitance value exactly forms a High-Q LC parallel resonance circuit with the line inductance and the load impedance at the current frequency point. Under this working condition, the physical manifestation is: although the filter capacitor is connected, the ripple amplitude at the dominant noise frequency does not decrease but increases, and even appears divergent oscillation. Based on this, when it is monitored that the ripple amplitude corresponding to the dominant noise frequency presents an upward trend after the reconstruction of the variable capacitance array, it is determined that the current system is in a parallel resonance state. At this time, the current filtering strategy needs to be suspended, the target impedance characteristic is changed from the current characteristic to a high impedance characteristic, and the target topology combination containing a high ESR branch is reconstructed for switching. The core feature of this high ESR branch is that a damping resistance of a certain resistance value is connected in series. When the controlled switch unit controls the closing of this branch, the damping resistance is connected into the oscillation circuit. According to the circuit principle, the introduction of this series resistance significantly increases the equivalent series resistance ESR of the filter circuit, thereby greatly reducing the Q value of the resonance circuit, forcing the resonance voltage to decay rapidly, thereby protecting the power devices of the charging module.

[0100] In some specific embodiments, the circuit nodes include a first node located at the output side of the filter inductor, a second node located at the input side of the filter inductor, and a third node connected in parallel to the power switch unit of the charging module; based on this hardware structure, the reconfigurable switch network can be physically switched between the above three nodes according to the current working condition, thereby switching the variable capacitance array to different circuit nodes to achieve different suppression effects. Please refer to Figure 4 , Figure 4 One embodiment of the method for filtering the charging module based on the variable capacitance array provided in this application, which switches the variable capacitance array to different circuit nodes, includes:

[0101] 401. When the dominant noise frequency is a low-frequency ripple and the amplitude is less than the first threshold, it is determined to be a steady-state output condition. The reconfigurable switching network is connected to the first node to form an LC low-pass filter topology with the filter inductor to smooth the output voltage.

[0102] Steady-state output operation refers to the situation where, during the normal constant current or constant voltage charging phase of the charging module, the output ripple is mainly manifested as the fundamental frequency or low-order harmonics of the switching frequency, and the amplitude is within the normal allowable range (less than the first threshold). The main task at this time is to maintain the stability of the output voltage and ensure efficient energy transfer. Therefore, the reconfigurable switching network is connected to the first node. In this topology, the variable capacitor array and the preceding filter inductor together constitute a standard second-order LC low-pass filter topology. The variable capacitor array smooths the pulsating DC current output by the inductor. Compared to complex high-order filters, this topology has the lowest conduction loss and highest efficiency in steady state.

[0103] 402. When the dominant noise frequency contains high-frequency interference and the frequency exceeds the second threshold, it is determined to be a conducted interference condition. The reconfigurable switching network is connected to the second node, which together with the filter inductor and the inherent output capacitor of the charging module forms a CLCπ-type filter topology to improve the high-frequency insertion loss.

[0104] When high-frequency common-mode interference is detected from the power grid, or when high-frequency hard switching of power devices causes extremely high-frequency electromagnetic interference (EMI) noise at the output, exceeding the second threshold (e.g., several megahertz), traditional second-order LC filters often suffer from insufficient insertion loss due to high-frequency parasitic parameters, leading to noise penetration and interference with the load. This is classified as conducted interference. To improve the blocking capability against high-frequency noise, a reconfigurable switching network is connected to the second node. Here, the variable capacitor array acts as the first-stage filter capacitor, the filter inductor as the intermediate impedance, and the inherent output capacitor of the charging module acts as the second-stage filter capacitor. These three are physically connected in series to form a third-order CLC π-type filter topology. According to filter theory, a third-order CLC π-type filter has a steeper stopband attenuation slope than a second-order LC filter. This reconfiguration allows the system to instantly achieve extremely strong high-frequency noise suppression capabilities without adding additional components, effectively blocking EMI interference within the power stage and preventing its conduction to the output.

[0105] 403. When the drain-source voltage change rate of the power switching unit exceeds the safety threshold, it is determined to be a voltage stress condition. Priority is given to controlling the reconfigurable switching network to connect to the third node to build a voltage clamping buffer topology to absorb the voltage spikes generated by the operation of the power switching unit.

[0106] In extreme situations such as the start of the charging module, load dump or short circuit protection, the power switch unit will withstand a great drain-source voltage rate at the moment of turning off. Due to the existence of parasitic inductance in the circuit, the sharp current change will induce a very high peak voltage across the switch tube. If not suppressed, it is easy to break through the power device. Therefore, when the drain-source voltage rate dv / dt of the power switch unit is detected to exceed the safety threshold, it is determined that the voltage stress condition is met. At this time, the main task is to protect the device safety, so the reconfigurable switch network will switch the variable capacitance array, specifically the specific high-voltage small-capacity branch in the array, to the third node. At this time, the capacitor is directly connected in parallel across the power switch tube, forming a voltage clamping buffer topology. By using the characteristic that the capacitor voltage cannot change abruptly, this voltage clamping buffer topology can absorb the energy released by the parasitic inductance at the moment of turning off, forcibly reduce the dv / dt slope, clamp the voltage peak within a safe range, and ensure the safety of the device.

[0107] In some specific embodiments, in order to avoid the electrical stress problem caused by switching, a soft switching control strategy containing zero-crossing turn-off and equipotential closing is introduced. Please refer to Figure 5 , Figure 5 An embodiment of the soft switching control strategy in the variable capacitance array-based filtering method of the charging module provided in the present application includes:

[0108] 501. In the process of controlling the on-off state of the controlled switch unit, the phase of the branch current flowing through the controlled switch unit is monitored in real time, and the disconnection action is preferentially triggered at the moment when the phase of the branch current is at the zero-crossing point;

[0109] In the process of controlling the on-off state of the controlled switch unit, if the capacitor branch with large current is forcibly cut off, due to the existence of stray inductance in the circuit, the instantaneous change of current will induce a very high reverse electromotive force across the switch contact, causing the switch device to withstand overvoltage breakdown risk or generate arc interference. Therefore, when it is determined that a certain group of capacitor branches needs to be disconnected, the disconnection action is not immediately performed, but the phase of the branch current flowing through the controlled switch unit is continuously monitored. Only when the current waveform naturally crosses the zero point at the moment will a turn-off signal be sent. At this time, the energy stored in the circuit is zero, and the disconnection action will not produce arc or voltage peak, thus eliminating the switching noise from the source and prolonging the service life of the power device.

[0110] 502. In the process of controlling the reconfigurable switch network to switch the variable capacitance array to the corresponding circuit node, the end voltage of the variable capacitance array and the end voltage of the target circuit node are monitored before the variable capacitance array is connected to the new circuit node.

[0111] 503、If the voltage difference between the two exceeds the preset safety range, first control the controlled switch unit to pre-charge or pre-discharge in high-frequency PWM modulation mode until the voltage difference falls within the safety range, and then connect the variable capacitance array to the new circuit node.

[0112] When performing the connection or node switching action of the reconfigurable switch network, there is a risk of facing a pressure difference impact. For example, if the first node of the variable capacitance array stores a charge of 400V, and it is directly switched to a second node with a voltage of only 200V, a huge voltage difference will instantly generate a surge current similar to a short circuit effect, which will cause the contact to burn out or cause the self-healing of the internal metallized film of the capacitor to fail. Therefore, before physically connecting, the current terminal voltage of the variable capacitance array and the real-time voltage of the target circuit node to be connected are collected, the difference between the two is calculated, and the risk of direct closure is evaluated. When it is detected that the voltage difference is too large, i.e., exceeds the preset safety range, the reconfigurable switch network cannot be directly closed, but the controlled switch unit inside the variable capacitance array with extremely fast response speed is used as an adjustment actuator to drive these controlled switch units into a high-frequency PWM operating mode. The controlled switch unit quickly turns on and off microscopically, and macroscopically behaves as a variable equivalent resistance. If the capacitor voltage is lower than the target node voltage, PWM control is equivalent to limiting the charging current and pre-charging the capacitor; if the capacitor voltage is higher than the target node voltage, PWM control is equivalent to pre-discharging the capacitor through energy consumption or feedback. With the pre-charge / discharge process, the terminal voltage of the variable capacitance array gradually approaches the terminal voltage of the target node, and when the difference between the two falls within the safety range, the controlled switch unit can be controlled to be fully on, and the reconfigurable switch network can be controlled to complete the final physical closure, realizing the connection of the variable capacitance array to the new circuit node.

[0113] In this embodiment, the zero-crossing turn-off eliminates the reverse electromotive force breakdown and switch arc interference that may occur when the inductive loop is cut off from the source, and the pre-charge / discharge technology of high-frequency PWM modulation is used to realize the smooth connection of the variable capacitance array and the target circuit node at the same potential, effectively suppressing the closing surge current caused by the pressure difference. This ensures that the hardware components are not damaged by the impact, prolongs the service life of the system, and also ensures the stability of the output voltage of the charging module when the filter topology is frequently reconstructed, avoiding secondary voltage disturbance caused by the adjustment process itself.

[0114] The filter system of the charging module based on the variable capacitance array provided in the present application will be described in detail below. Please refer to Figure 6 , Figure 6For another embodiment of the filtering system of the charging module based on the variable-capacitance array provided in the present application, the charging module comprises a filtering inductor and a variable-capacitance array arranged around the filtering inductor, the variable-capacitance array is composed of a plurality of groups of capacitance branches with different impedance characteristics, each capacitance branch is connected in series with a controlled switch unit, and the variable-capacitance array is connected to a circuit node of the charging module through a reconfigurable switch network; the filtering system comprises:

[0115] The acquisition unit 601 is configured to acquire an output voltage signal of an output end of the charging module at a preset high-frequency sampling rate, and perform frequency domain analysis on the output voltage signal to extract an output ripple spectrum feature;

[0116] The identification unit 602 is configured to identify a frequency component with an amplitude greater than a preset threshold at a current time according to the output ripple spectrum feature, and determine a frequency component with the largest amplitude as a dominant noise frequency;

[0117] The calculation unit 603 is configured to calculate a target topology combination for the filtering circuit composed of the filtering inductor and the variable-capacitance array to exhibit a target impedance characteristic at the dominant noise frequency based on the dominant noise frequency;

[0118] The control unit 604 is configured to control the on-off state of the controlled switch unit according to the target topology combination, and control the reconfigurable switch network to switch and connect the variable-capacitance array to a corresponding circuit node, so as to reconfigure the variable-capacitance array into a filtering network with the target impedance characteristic, and filter out a ripple component corresponding to the dominant noise frequency.

[0119] In the system of the embodiment, the functions of each unit correspond to the steps in the method embodiment described above, and will not be described here again. Figures 1 to 5

[0120] The present application also provides a filtering device of a charging module based on a variable-capacitance array, please refer to Figure 7 , Figure 7 For an embodiment of the filtering device of the charging module based on the variable-capacitance array provided in the present application, the device comprises:

[0121] The processor 701, the memory 702, the input and output unit 703, and the bus 704;

[0122] The processor 701 is connected with the memory 702, the input and output unit 703, and the bus 704;

[0123] The memory 702 stores a program, and the processor 701 invokes the program to perform any of the filtering methods of the charging module based on the variable-capacitance array as described above.

[0124] ​The application also relates to a computer readable storage medium, which stores a program, and when the program is run on a computer, the computer executes the filtering method of any one of the above charge modules based on a variable capacitance array.

[0125] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0126] In the several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only schematic, and the division of units is only a logical function division, and there can be another division manner in actual implementation, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0127] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0128] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0129] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the various embodiments of the method of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, read-only memory), a random access memory (RAM, random access memory), a magnetic disk or an optical disk, and various media that can store program codes.

Claims

1. A filtering method for a charge module based on a variable capacitance array, characterized by, The charging module comprises a filter inductor and a variable capacitance array arranged around the filter inductor, the variable capacitance array is composed of a plurality of groups of capacitor branches with different impedance characteristics, each of the capacitor branches is connected in series with a controlled switch unit, and the variable capacitance array is connected to circuit nodes of the charging module through a reconfigurable switch network, the circuit nodes comprise a first node at an output side of the filter inductor, a second node at an input side of the filter inductor, and a third node connected in parallel to both ends of a power switch unit of the charging module; The filtering method comprises: acquiring an output voltage signal of an output end of the charging module at a preset high-frequency sampling rate, and performing frequency domain analysis on the output voltage signal to extract an output ripple spectrum feature; identifying a frequency component with an amplitude greater than a preset threshold at a current time according to the output ripple spectrum feature, and determining a frequency component with the largest amplitude as a dominant noise frequency; based on the dominant noise frequency, calculating a target topology combination for the filter circuit composed of the filter inductor and the variable capacitance array to exhibit a target impedance characteristic at the dominant noise frequency; controlling the on-off state of the controlled switch unit according to the target topology combination, and controlling the reconfigurable switch network to switch the variable capacitance array to be connected to the corresponding circuit node, and reconfiguring the variable capacitance array into a filter network with the target impedance characteristic to filter out a ripple component corresponding to the dominant noise frequency; the control of the reconfigurable switch network to switch the variable capacitance array to be connected to the corresponding circuit node comprises: when the dominant noise frequency is a low-frequency ripple and the amplitude is less than a first threshold, it is determined to be a steady-state output working condition, and the reconfigurable switch network is controlled to be connected to the first node to form an LC low-pass filter topology with the filter inductor to smooth the output voltage; when the dominant noise frequency contains high-frequency interference and the frequency exceeds a second threshold, it is determined to be a conducted interference working condition, and the reconfigurable switch network is controlled to be connected to the second node to form a CLCπ type filter topology with the filter inductor and the inherent output capacitor of the charging module to improve high-frequency insertion loss; when the rate of change of drain-source voltage of the power switch unit is detected to exceed a safety threshold, it is determined to be a voltage stress working condition, and the reconfigurable switch network is preferentially controlled to be connected to the third node to build a voltage clamping buffer topology to absorb voltage spikes generated by the action of the power switch unit.

2. The filtering method of claim 1, wherein, The variable capacitance array contains a low ESR branch using a thin film capacitor, a high ESR branch using a ceramic capacitor in series with a damping resistor, and a notch branch using a capacitor and an inductor in series; the calculation of the target topology combination for the filter circuit composed of the filter inductor and the variable capacitance array to exhibit a target impedance characteristic at the dominant noise frequency based on the dominant noise frequency comprises: determining a target impedance characteristic according to the relative relationship between the dominant noise frequency and the resonance frequency of the filter circuit, the target impedance characteristic comprising a low impedance characteristic, a high impedance characteristic, and a notch characteristic; According to the target impedance characteristic, one or more of the low-ESR branch, the high-ESR branch and the notch branch are selected to be combined to construct a target topology combination.

3. The filtering method of claim 2, wherein, According to the target impedance characteristic, one or more of the low-ESR branch, the high-ESR branch and the notch branch are selected to be combined to construct a target topology combination. A three-dimensional state space is constructed, including a capacitance branch type, a circuit node position and a switch combination state. A comprehensive evaluation model is established, taking a ripple attenuation rate as a benefit function and a transient voltage disturbance generated by a switching action as a cost function. In the three-dimensional state space, a candidate subset meeting the target impedance characteristic is screened out, and each combination in the candidate subset is substituted into the comprehensive evaluation model for calculation, and a combination that maximizes a weighted difference between the benefit function and the cost function is determined as the target topology combination.

4. The filtering method of claim 2, wherein, The filtering method further includes: When the target topology combination includes the notch branch, an equivalent capacitance value corresponding to the notch branch in the current switch combination state is determined, and the equivalent capacitance value is an equivalent capacity presented by the notch branch in the current switch combination state. Based on the equivalent capacitance value, a positive and negative alternating discrete capacitance disturbance is applied by changing the combination state of the controlled switch unit, so that the actual resonance frequency of the filter circuit is dithered within a preset fine-tuning range. During the application of the discrete capacitance disturbance, the rate of change of the ripple amplitude corresponding to the dominant noise frequency is monitored synchronously, and a gradient direction of the ripple amplitude with respect to the capacitance change is calculated. The target topology combination is dynamically updated according to the gradient direction.

5. The filtering method of claim 2, wherein, The filtering method further includes: The ripple amplitude corresponding to the dominant noise frequency is continuously monitored, and if it is monitored that the ripple amplitude corresponding to the dominant noise frequency presents an upward trend after the variable capacitance array is reconstructed, it is determined that the current target topology combination forms a parallel resonance with the load impedance. The target impedance characteristic is updated from the current characteristic to a high impedance characteristic, the target topology combination including the high-ESR branch is reconstructed, and switching is performed.

6. The filtering method according to any one of claims 1 to 5, characterized in that, The filtering method further includes: During the control of the on-off state of the controlled switch unit, the branch current phase flowing through the controlled switch unit is monitored in real time, and the disconnection action is preferentially triggered at the moment when the branch current phase is at the zero-crossing point. During the process of controlling the reconfigurable switch network to switch the variable capacitance array to be connected to the corresponding circuit node, the end voltage of the variable capacitance array and the end voltage of the target circuit node are monitored before the variable capacitance array is connected to the new circuit node. If the voltage difference between the two exceeds a preset safety range, the variable capacitance array is first connected to the new circuit node after the voltage difference falls within the safety range by controlling the controlled switch unit to perform pre-charging or pre-discharging in a high-frequency PWM modulation manner.

7. A filter system for a charge module based on a variable capacitance array, characterized by, The charging module comprises a filter inductor and a variable capacitance array arranged around the filter inductor, the variable capacitance array is composed of a plurality of groups of capacitor branches with different impedance characteristics, each capacitor branch is connected in series with a controlled switch unit, and the variable capacitance array is connected to circuit nodes of the charging module through a reconfigurable switch network, the circuit nodes comprise a first node located at an output side of the filter inductor, a second node located at an input side of the filter inductor, and a third node connected in parallel to both ends of a power switch unit of the charging module; The filter system comprises: The acquisition unit is configured to acquire an output voltage signal of an output end of the charging module at a preset high-frequency sampling rate, and perform frequency domain analysis on the output voltage signal to extract an output ripple spectrum feature; The identification unit is configured to identify a frequency component with an amplitude greater than a preset threshold at a current time according to the output ripple spectrum feature, and determine a frequency component with the largest amplitude as a dominant noise frequency; The calculation unit is configured to calculate a target topology combination of the filter circuit composed of the filter inductor and the variable capacitance array to make the filter circuit have a target impedance characteristic at the dominant noise frequency; The control unit is configured to control on-off states of the controlled switch units according to the target topology combination, and control the reconfigurable switch network to switch and connect the variable capacitance array to corresponding circuit nodes, and reconfigure the variable capacitance array into a filter network with the target impedance characteristic to filter out a ripple component corresponding to the dominant noise frequency; The control unit is specifically configured to: When the dominant noise frequency is a low-frequency ripple and the amplitude is less than a first threshold, determine that it is a steady-state output working condition, control the reconfigurable switch network to be connected to the first node to form an LC low-pass filter topology with the filter inductor to smooth the output voltage; When the dominant noise frequency contains high-frequency interference and the frequency exceeds a second threshold, determine that it is a conducted interference working condition, control the reconfigurable switch network to be connected to the second node to form a CLCπ type filter topology with the filter inductor and an inherent output capacitor of the charging module to improve high-frequency insertion loss; When a change rate of a drain-source voltage of the power switch unit exceeds a safety threshold, determine that it is a voltage stress working condition, preferentially control the reconfigurable switch network to be connected to the third node to build a voltage clamping buffer topology to absorb voltage spikes generated by the action of the power switch unit.

8. A filter device for a charge module based on a variable capacitance array, characterized by The filter device comprises: a processor, a memory, an input / output unit, and a bus; the processor is connected with the memory, the input / output unit, and the bus; the memory stores a program, and the processor invokes the program to execute the method in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program, and the program performs the method in any one of claims 1 to 6 when executed on a computer.

Images