Series-side converter control method based on dynamic sampling and detection

Abstract

This application relates to the field of power system signal processing technology, specifically to a control method for a series-side converter based on dynamic sampling and detection. The method includes: acquiring the three-phase voltage signal of the line where the series-side converter is located; performing mode decomposition on each phase voltage signal, obtaining a first feature value for each mode component based on frequency characteristics, and obtaining a first filter component set and a second filter component set; obtaining a first filter based on the first filter component set; performing wavelet decomposition and feature extraction on each mode component in the second filter component set to obtain a second feature value, and reassigning the amplitude of the wavelet coefficients; reconstructing all scales of each reassigned mode component to obtain a second filter; and controlling the series-side converter based on the three-phase voltage signal processed by the two filters. This application aims to filter out noise interference components in the three-phase voltage and improve the control effect of the series-side converter.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of power system signal processing, in particular to a series side converter control method based on dynamic sampling detection. BACKGROUND

[0002] The installation scale of distributed new energy power generation systems in distribution networks is rapidly expanding. However, with a high proportion of photovoltaic access to the distribution network, the randomness and intermittency of the photovoltaic itself can bring many adverse effects on the power quality of the grid-connected distribution network, such as voltage fluctuation of the distribution network caused by frequent start-stop of intermittent distributed power and change of power output, low power factor, and other power quality problems. Among them, intermittent power usually needs to use power electronic converters to output to meet the frequency and voltage requirements of user load.

[0003] By connecting the series side converter and the transformer together in series at the low-voltage side of the transformer, the three-phase voltage signals collected by the sampling circuit are used to control the action of the series side converter, thereby providing compensation voltage in the line, which can realize fine voltage regulation and voltage compensation, and prevent voltage overrun. Among them, accurate sampling of signals is an important part of series side converter control. Accurate three-phase voltage signals can help the series side converter accurately calculate the required voltage compensation amount, so as to quickly and effectively control the voltage of the distribution network within a reasonable range.

[0004] However, the sampling circuits in the prior art have more or less certain drawbacks, such as poor anti-interference ability, improper handling of sampled voltage, etc., which results in low reliability and accuracy of the three-phase voltage signals collected by the existing sampling circuit, thereby affecting the control accuracy of the series side converter in the distribution network. Therefore, a filter circuit is usually used to filter the voltage signals obtained by sampling to improve the sampling accuracy of the three-phase voltage, such as a three-phase voltage and current sampling and analog-digital conversion circuit and method disclosed in CN118944668A. However, this method can only filter out high-frequency interference signals in the power grid, and cannot effectively filter out other noise interference components in the three-phase voltage. SUMMARY

[0005] In view of the above, it is necessary to provide a series side converter control method based on dynamic sampling detection to solve the above problems.

[0006] One embodiment of the application provides a series side converter control method based on dynamic sampling detection, which comprises:

[0007] S1: acquiring three-phase voltage signals of a line where the series side converter is connected;

[0008] S2: modal decomposition is performed on each phase voltage signal, a contrast component of each modal component of each phase voltage signal in other phase voltage signals is obtained based on frequency characteristics, a first characteristic value of each modal component is obtained based on a similarity degree of a frequency spectrum characteristic of each modal component and its contrast component, threshold segmentation is performed on all modal components of all phase voltage signals based on the first characteristic value, a first filter component set and a second filter component set of each phase voltage signal are obtained, and a first FIR filter of each phase voltage signal is obtained based on a frequency spectrum range of each modal component in the first filter component set of each phase voltage signal;

[0009] S3: wavelet decomposition is performed on each modal component in the second filter component set of each phase voltage signal and all contrast components thereof at a preset scale, all wavelet coefficients at each scale are obtained, a second characteristic value is obtained according to a local distribution correlation of wavelet coefficients in the same scale in different phase voltage signals and in combination with a local distribution correlation of wavelet coefficients at different scales in each phase voltage signal, and whether to reassign an amplitude of the wavelet coefficient of each modal component at each scale is judged, reconstruction is performed on all scales of each modal component after reassignment, a second FIR filter of each phase voltage signal is obtained based on a frequency spectrum range of each reconstructed modal component in the second filter component set;

[0010] S4: the series side converter is controlled according to the three-phase voltage signals processed by the two FIR filters.

[0011] Preferably, the contrast component of each modal component of each phase voltage signal is selected from modal components in voltage signals of other phases, and the selected modal component has the minimum absolute value of a central frequency difference from each modal component.

[0012] Preferably, the first characteristic value of each modal component is obtained by:

[0013] a waveform similarity and a frequency spectrum similarity between each modal component and each contrast component thereof are calculated, a mean value of the waveform similarity of each modal component and all contrast components thereof is recorded as a first mean value;

[0014] a mean value of the frequency spectrum similarity of each modal component and all contrast components thereof is recorded as a second mean value;

[0015] a result of forward fusion of the first mean value and the second mean value corresponding to each modal component is taken as the first characteristic value of each modal component.

[0016] Preferably, the first filter component set and the second filter component set of each phase voltage signal are obtained by:

[0017] inputting the first eigenvalue of all modal components of all phase voltage signals into the maximum between-cluster variance algorithm to output a first threshold value;

[0018] a set of all modal components with the first eigenvalue greater than the first threshold value in each phase voltage signal is recorded as a first filter component set of each phase voltage signal, and a set of remaining modal components in each phase voltage signal is recorded as a second filter component set of each phase voltage signal.

[0019] Preferably, the first FIR filter of each phase voltage signal is specifically obtained by:

[0020] for each modal component in the first filter component set of each phase voltage signal, the maximum frequency and the minimum frequency of the frequency with the amplitude greater than 0 in each frequency spectrum are respectively taken as the upper limit and the lower limit of the frequency distribution range;

[0021] the union of the frequency distribution ranges of all modal components is taken as the passband frequency range of the FIR filter, and the first FIR filter of each phase voltage signal is obtained by combining the Parks-McClellan algorithm.

[0022] Preferably, the specific process of obtaining the second eigenvalue is:

[0023] the first characteristic factor is obtained by comparing the local distribution similarity of each wavelet coefficient of each modal component and its control component in the same scale;

[0024] the second characteristic factor is obtained by comparing the local distribution similarity of each wavelet coefficient of each modal component in different scales;

[0025] the product of the amplitude of each wavelet coefficient, the first characteristic factor and the second characteristic factor is taken as the second eigenvalue.

[0026] Preferably, the first characteristic factor is specifically obtained by:

[0027] for each wavelet coefficient in each scale, a sequence composed of the amplitudes of a preset number of wavelet coefficients closest to the sampling time of each wavelet coefficient is recorded as the first sequence of each wavelet coefficient;

[0028] the mean value of the correlation between each wavelet coefficient of each modal component and the first sequence of each wavelet coefficient of all control components of the modal component is taken as the first characteristic factor.

[0029] Preferably, the second characteristic factor is specifically the mean value of the correlation degree between each wavelet coefficient in each scale and the first sequence of each wavelet coefficient in all other scales.

[0030] Preferably, the process of judging whether to reassign the amplitude of the wavelet coefficient in each scale of each modal component is:

[0031] The second threshold value is obtained by threshold segmentation of the second eigenvalue of all wavelet coefficients in each scale of each modal component, and the amplitude of all wavelet coefficients in each scale whose second eigenvalue is less than the second threshold value is assigned as 0.

[0032] Preferably, the passband frequency range of the second FIR filter is the union of the frequency spectrum distribution range of the reconstructed modal component of each modal component in the second filter component set.

[0033] The present application has at least the following beneficial effects:

[0034] 1. In the present application, the first eigenvalue of each modal component is obtained based on the similarity degree of the spectral characteristics of each modal component and its control component, and the first FIR filter is constructed based on the first eigenvalue, which can avoid the useful information in the signal component in a certain frequency range in the three-phase voltage signal without noise being excessively filtered out.

[0035] 2. In the present application, the second eigenvalue is obtained according to the local distribution correlation of the wavelet coefficients in the same scale in different phase voltage signals, combined with the local distribution correlation of the wavelet coefficients in different scales in each phase voltage signal, and the second FIR filter is constructed based on the second eigenvalue. Compared with directly using the filter in the preset frequency range to filter the collected three-phase voltage signal, the correlation between the wavelet coefficients can be fully utilized to improve the filtering effect of the filter on the noise in the three-phase voltage signal and the retention degree of the useful information in the signal.

[0036] 3. The present application realizes the filtering processing of the three-phase voltage signal by using the constructed FIR filter, and controls the series side converter based on the filtered three-phase voltage. Compared with the existing filtering circuit for filtering the sampled voltage signal, the noise component in the three-phase voltage signal can be effectively filtered out and the useful information in the signal can be retained, thereby improving the control accuracy of the series side converter in the power distribution network, realizing the fine voltage regulation and voltage compensation of the power distribution network, and preventing voltage out-of-limit. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 The flowchart of the series side converter control method based on dynamic sampling detection provided by the present application is provided.

[0038] Figure 2 The flowchart of the second eigenvalue provided by the present application is provided. DETAILED DESCRIPTION

[0039] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0041] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0043] This application proposes a series-side converter control method based on dynamic sampling and detection, which is applied to the field of power system signal processing technology. (See attached document.) Figure 1 The method includes the following steps:

[0044] S1: Obtain the three-phase voltage signal of the line where the series converter is located.

[0045] This application uses a pulse width modulation (PWM) signal sampling and detection circuit to acquire the three-phase voltage signals of the line where the series-side converter is located. The sampling rate of the three-phase voltage signals is dynamically adjusted based on changes in the input signal of the PWM signal sampling and detection circuit to ensure the sampling accuracy and effectiveness of the three-phase voltage signals, thus achieving dynamic sampling and detection of the three-phase voltage signals of the line where the series-side converter is located. An amplifier circuit is used to amplify the acquired three-phase voltage signals to improve the signal-to-noise ratio. An analog-to-digital converter circuit is used to process the amplified three-phase voltage signals, converting the analog voltage signals into digital signals for subsequent filtering. The three-phase voltage signals include phase A, phase B, and phase C voltage signals. The PWM signal sampling and detection circuit, amplifier circuit, and analog-to-digital converter circuit are all well-known technologies, and their specific processes will not be described in detail.

[0046] S2: Perform mode decomposition on each phase voltage signal, pre-define the reference component of each mode component of each phase voltage signal in other phase voltage signals, and obtain the first feature value of each mode component based on the similarity of the spectral characteristics of each mode component and its reference component; perform threshold segmentation on all mode components of all phase voltage signals based on the first feature value to obtain the first filter component set and the second filter component set of each phase voltage signal; obtain the first FIR filter of each phase voltage signal based on the spectral range of each mode component in the first filter component set of each phase voltage signal.

[0047] In the actual control process of series-side converters in power distribution networks, the three-phase voltage signals collected usually contain noise interference due to factors such as equipment installation location, external electromagnetic interference, and mutual inductance between parallel lines. The presence of noise will affect the calculation results of voltage compensation of the series-side converter. Therefore, it is necessary to filter the collected three-phase voltage signals.

[0048] Since useful information in the three-phase voltage signal in the distribution network is usually concentrated in certain specific frequency bands, while noise, due to its diversity and complexity, is usually distributed over a wide frequency range, directly using a single filter with a pre-set frequency range to filter the three-phase voltage signal collected by the series-side converter can easily lead to excessive filtering of useful information in the signal components within a noise-free frequency band of the three-phase voltage signal. Therefore, the following processing is performed.

[0049] Specifically, the three-phase AC power system used in the distribution network is designed to achieve symmetrical operation, that is, the amplitude of the three-phase voltage is equal and the phase angle differs by 120 degrees. This symmetry can ensure the stable operation of the power system and the effective transmission of electrical energy. Under normal circumstances, the three-phase voltage signals in the distribution network should have similar signal distribution characteristics, that is, the waveform and frequency changes of each phase voltage should be basically consistent. When the signal component of a certain phase voltage signal in a certain frequency band has a large signal distribution difference from the signal components of other phase voltage signals in similar frequency bands, then this phase voltage signal is more likely to be subject to noise interference in similar frequency bands. When using filters for signal filtering, this frequency should be set outside the pass frequency band.

[0050] Based on the above analysis, the A-phase voltage signal, B-phase voltage signal, and C-phase voltage signal are used as inputs to the Variational Mode Decomposition (VMD) algorithm, respectively. The algorithm outputs K modal components of each phase voltage signal and the center frequency of each modal component, which are then used to analyze the signal distribution differences between signal components in similar frequency bands of different phase voltage signals. In this embodiment, the number of modes K in the VMD algorithm is set to 10. Implementers can set this number according to their actual situation; this application does not impose any restrictions on it. It should be noted that the VMD algorithm is a well-known technology, and the specific process will not be described in detail here.

[0051] This embodiment takes the i-th mode component A(i) in the A-phase voltage signal of the three-phase voltage signal as an example, and uses Fourier transform to extract the spectrum of the mode component A(i). The extraction of the signal spectrum is a well-known technique, and the specific process will not be described in detail.

[0052] From the modal components of the B-phase and C-phase voltage signals, the modal component with the smallest absolute value of the difference between its center frequency and that of modal component A(i) is selected as the reference component of modal component A(i). The waveform similarity and spectral similarity between modal component A(i) and each reference component are calculated. The mean of the waveform similarity between modal component A(i) and all its reference components is denoted as the first mean. The mean of the spectral similarity between modal component A(i) and all its reference components is denoted as the second mean. The result of positively fusing the first mean and the second mean corresponding to modal component A(i) is used as the first feature value of modal component A(i), which is used to characterize the similarity of signal distribution between signal components in the B-phase and C-phase voltage signals that are in similar frequency ranges to modal component A(i). In this embodiment, the waveform similarity is obtained through the Pearson correlation coefficient; the spectral similarity is obtained through the calculation method of cosine similarity; the positive fusion of multiple variables is performed by multiplication; it should be noted that the Pearson correlation coefficient and cosine similarity are both existing known technologies, and this application will not elaborate on them.

[0053] The first eigenvalues ​​of all modal components of all phase voltage signals are used as inputs to the Otsu's inter-class variance algorithm, and the first threshold is output. Taking the A-phase voltage signal as an example, the set of all modal components in the A-phase voltage signal whose first eigenvalues ​​are greater than the first threshold is denoted as the first filter component set D1(A) of the A-phase voltage signal. This set is used to characterize the set of signal components in the frequency band of the A-phase voltage signal that are not affected by noise, so as to facilitate the subsequent design of the first filter of the A-phase voltage signal. The set of the remaining modal components in the A-phase voltage signal is denoted as the second filter component set D2(A) of the A-phase voltage signal. This set is used to characterize the set of signal components in the frequency band of the A-phase voltage signal that may be affected by noise, so as to facilitate the subsequent design of the second filter of the A-phase voltage signal. The Otsu's inter-class variance algorithm is a well-known technique, and the specific process will not be described in detail.

[0054] Furthermore, the spectrum of each modal component in the first filter component set D1(A) of the phase A voltage signal is extracted. The maximum and minimum frequencies with amplitudes greater than 0 in each spectrum are used as the upper and lower limits of the frequency distribution range, respectively. The union of the frequency distribution ranges of all modal components is denoted as the first union, which is used to characterize the frequency range of the signal components in the phase A voltage signal that are not affected by noise. The first union is used as the passband frequency range of the FIR filter. The Parks-McClellan algorithm is used to design the optimal FIR filter, resulting in the first FIR filter Q1(A) of the phase A voltage signal. In this embodiment, the filter order of the FIR filter is 60. The extraction of the signal spectrum and the Parks-McClellan algorithm are well-known techniques, and the specific process will not be described in detail in this application.

[0055] S3: Perform wavelet decomposition at a preset scale on each modal component and all its reference components in the second filter component set of each phase voltage signal to obtain all wavelet coefficients at each scale; based on the local distribution correlation of wavelet coefficients at the same scale in different phase voltage signals, and combined with the local distribution correlation of wavelet coefficients at different scales in each phase voltage signal, obtain the second eigenvalue, and determine whether to reassign the amplitude of the wavelet coefficients at each scale of each modal component; reconstruct all scales after reassigning each modal component, and obtain the second FIR filter for each phase voltage signal based on the spectral range of each reconstructed modal component in the second filter component set.

[0056] When a signal passes through a filter, the filter can only accept signals within a preset frequency range. However, traditional filtering methods typically set the upper and lower limits of the frequency range based on empirical values ​​and process the entire signal. This method does not take into account that noise components in different phase voltage signals may exist in different frequency bands. Therefore, to improve the filter's noise removal effect on three-phase voltage signals and the degree of retention of useful information in the signal, the following processing is performed.

[0057] Specifically, in a power distribution network, the frequency ranges of three-phase voltage signals are similar. Therefore, the wavelet coefficients obtained after wavelet transform of the modal components of these signals within similar frequency bands usually exhibit similar distribution characteristics. Noise, however, is different. Due to its randomness and complexity, the wavelet coefficients corresponding to noise often do not exhibit such similar distribution characteristics. Furthermore, after wavelet decomposition, the amplitude of the wavelet coefficients of the signal is usually greater than that of the noise wavelet coefficients. The different propagation characteristics of the wavelet transform coefficients of the signal and noise at different scales indicate that the wavelet transform of the signal has a strong correlation across scales, while the wavelet transform of the noise does not show a significant correlation across scales.

[0058] Based on the above analysis, taking the A-phase voltage signal as an example, and the j-th modal component a(j) in the second filter component set D2(A) as an example, the corresponding components of modal component a(j) in the B-phase and C-phase voltage signals are denoted as modal component b(j) and modal component c(j), respectively. Wavelet decomposition is performed on modal components a(j), b(j), and c(j) to obtain the wavelet coefficients of all scales of modal components a(j), b(j), and c(j). In this embodiment, the Daubechies 8 wavelet basis function is used for wavelet decomposition; the number of wavelet decomposition layers M=4, which can be adjusted by the implementer according to the actual situation, and this application does not impose any restrictions on this; it should be noted that wavelet decomposition is a well-known technique, and the specific process will not be described in detail.

[0059] Taking the nth wavelet coefficient a(j,m,n) in the mth scale a(j,m) of modal component a(j) as an example, the amplitudes of the h wavelet coefficients in the mth scale a(j,m) whose sampling time is closest to that of wavelet coefficient a(j,m,n) are arranged in ascending order according to their sampling time. This sequence is denoted as the first sequence L(j,m,n) of wavelet coefficients a(j,m,n), which characterizes the local wavelet coefficient distribution of wavelet coefficients a(j,m,n). In this embodiment, h is a preset number, which is 10. Implementers can adjust it according to the actual situation.

[0060] The mean of the Pearson correlation coefficient between the first sequence L(j,m,n) and the first sequence of the first wavelet coefficients at the m-th scale in the modal components b(j) and c(j) is denoted as the first characteristic factor of the wavelet coefficient a(j,m,n), which is used to characterize the degree of similarity of the local distribution of the wavelet coefficient a(j,m,n) with the n-th wavelet coefficients at the same scale in the modal components of other phase voltage signals.

[0061] The mean of the Pearson correlation coefficient between the first sequence L(j,m,n) and the first sequence of the nth wavelet coefficients in each scale of the modal component a(j) is denoted as the second characteristic factor of the wavelet coefficients a(j,m,n), which is used to characterize the local wavelet coefficient distribution characteristics of the wavelet coefficients a(j,m,n) and the degree of correlation between them and the other scales of the modal component a(j).

[0062] The product of the amplitude, first eigenfactor, and second eigenfactor of wavelet coefficient a(j,m,n) is denoted as the second eigenvalue of wavelet coefficient a(j,m,n). This eigenvalue characterizes the probability that wavelet coefficient a(j,m,n) represents useful information in the A-phase voltage signal at the m-th scale of the modal component a(j). The flowchart for obtaining the second eigenvalue is shown below. Figure 2 As shown.

[0063] Taking the m-th scale a(j,m) as an example, the second eigenvalue of all wavelet coefficients in the m-th scale a(j,m) is used as the input of the maximum inter-class variance algorithm. The second threshold is output, and the amplitude of all wavelet coefficients in the m-th scale a(j,m) whose second eigenvalue is less than the second threshold is assigned to 0. This yields the m-th scale after reassignment of the modal component a(j), so as to retain all wavelet coefficients corresponding to the useful information in the A-phase voltage signal in the m-th scale of the modal component a(j).

[0064] By using inverse wavelet transform to reconstruct all scales of the modal component a(j) after reassignment, the reconstructed modal component of the j-th modal component in the second filter component set D2(A) is obtained, so as to retain the signal component corresponding to the useful information in the A-phase voltage signal in the modal component a(j).

[0065] The spectrum of each reconstructed mode component in the second filter component set D2(A) is extracted. The maximum and minimum frequencies with amplitudes greater than 0 in each spectrum are used as the upper and lower limits of the frequency distribution range, respectively. The union of the frequency distribution ranges of all mode components is denoted as the second union, which is used to characterize the frequency range of the signal components in the second filter component set D2(A) that are not affected by noise. The second union is used as the passband frequency range of the FIR filter. The Parks-McClellan algorithm is used to design the optimal FIR filter, resulting in the second FIR filter Q2(A) for the A-phase voltage signal. In this embodiment, the filter order of the FIR filter is 60. The extraction of the signal spectrum and the Parks-McClellan algorithm are well-known techniques, and the specific process will not be described in detail in this application.

[0066] S4: Control the series-side converter based on the three-phase voltage signal processed by two FIR filters.

[0067] Each phase voltage signal in the three-phase voltage signal is filtered using a first FIR filter and a second FIR filter, respectively. The filtered three-phase voltage signal is then used as the input to the series-side converter. The corresponding two-phase voltage values ​​are generated using dq transformation, and the corresponding active current i is obtained by comparing these values ​​with voltage reference values. q and reactive current i d At this point, the PI control algorithm is used to obtain the corresponding command signal, and the corresponding pulse drive signal is generated through dq / αβ transformation to control the operation of the series-side converter. The voltage reference value is obtained as follows: the three-phase voltage signal input to the series-side converter is used to obtain the grid voltage angle estimate θ through a phase-locked loop, and then the three-phase voltage signal is compared with the grid voltage component u in the orthogonal stationary coordinate system obtained by Clarke transformation. α and u β The grid-side voltage compensation value is calculated and multiplied by the transformer turns ratio N. T Obtain the voltage reference value of the series converter.

[0068] In the series-side converter, this embodiment mainly uses the SPWM pulse width modulation method for control, which mainly uses a power frequency sine wave and a high-frequency triangular wave for control. The 50Hz power frequency sine wave is used as the modulation wave, and the 6.25kHz triangular wave is used as the carrier wave. Every 16 sampling points constitute a carrier wave, thereby obtaining a square wave array. This square wave is used as the signal to drive the converter. When the intersection of the carrier wave and the modulation wave is a rising edge, the lower bridge arm of the converter is turned on and turned off. When the intersection is a falling edge, the lower bridge arm is turned on and the upper bridge arm is turned off.

[0069] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0070] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method of control of a series side converter based on dynamic sampling detection, characterized in that, The method comprises the following steps: S1: acquiring three-phase voltage signals of a line to which a series side converter is connected; S2: performing modal decomposition on each phase voltage signal, selecting, from modal components in voltage signals of other phases, a modal component with the smallest absolute value of difference from a center frequency of each modal component as a reference component of each modal component in the voltage signals of the other phases, obtaining a first characteristic value of each modal component based on a similarity degree of spectral characteristics of each modal component and the reference component thereof, performing threshold segmentation on all modal components of all phase voltage signals based on the first characteristic value, obtaining a first filter component set and a second filter component set of each phase voltage signal, and obtaining a first FIR filter of each phase voltage signal based on a spectral range of each modal component in the first filter component set of each phase voltage signal; S3: performing wavelet decomposition of each modal component in the second filter component set of each phase voltage signal and all reference components thereof at a preset scale to obtain all wavelet coefficients at each scale, obtaining a second characteristic value based on local distribution correlation of wavelet coefficients in the same scale in different phase voltage signals and combining local distribution correlation of wavelet coefficients at different scales in each phase voltage signal, judging whether to reassign an amplitude of the wavelet coefficients at each scale of each modal component, and reconstructing all scales of each modal component after reassignment, obtaining a second FIR filter of each phase voltage signal based on a spectral range of each reconstructed modal component in the second filter component set; S4: controlling the series side converter based on the three-phase voltage signals processed by the two FIR filters; The first characteristic value of each modal component is obtained in the following manner: a waveform similarity and a spectral similarity between each modal component and each reference component thereof are calculated, an average of the waveform similarities of each modal component and all reference components thereof is denoted as a first average value, an average of the spectral similarities of each modal component and all reference components thereof is denoted as a second average value, and a result of forward fusion of the first average value and the second average value corresponding to each modal component is taken as the first characteristic value of each modal component. The first filter component set and the second filter component set of each phase voltage signal are obtained in the following manner: first characteristic values of all modal components of all phase voltage signals are taken as inputs of a maximum inter-class variance algorithm, and a first threshold value is outputted, 2. The series side converter control method based on dynamic sampling detection of claim 1, wherein, a set of all modal components with the first characteristic values greater than the first threshold value in each phase voltage signal is denoted as the first filter component set of each phase voltage signal, and a set of remaining modal components in each phase voltage signal is denoted as the second filter component set of each phase voltage signal. The first FIR filter of each phase voltage signal is obtained in the following manner: for each modal component in the first filter component set of each phase voltage signal, a maximum frequency and a minimum frequency of frequencies with amplitudes greater than 0 in each spectrum are taken as upper and lower limits of a frequency distribution range, respectively.

3. The method of claim 1, wherein the method further comprises: ​ ​ The union of the frequency distribution ranges of all modal components is taken as the passband frequency range of the FIR filter, and the first FIR filter of each phase voltage signal is obtained by combining the Parks-McClellan algorithm.

4. The method of claim 1, wherein the method further comprises: The specific process of obtaining the second characteristic value is: The first characteristic factor is obtained by comparing the local distribution similarity of each wavelet coefficient of each modal component and its control component in the same scale; The second characteristic factor is obtained by comparing the local distribution similarity of each wavelet coefficient of each modal component in different scales; The product of the amplitude of each wavelet coefficient, the first characteristic factor, and the second characteristic factor is taken as the second characteristic value.

5. The series side converter control method based on dynamic sampling detection of claim 4, wherein, The first characteristic factor is obtained by: For each wavelet coefficient in each scale, a sequence composed of the amplitudes of a preset number of wavelet coefficients closest in sampling time to each wavelet coefficient is taken as the first sequence of each wavelet coefficient. The average of the correlation between each wavelet coefficient in each scale and the first sequence of each wavelet coefficient of all control components of each modal component is taken as the first characteristic factor.

6. The series side converter control method based on dynamic sampling detection of claim 5, wherein, The second characteristic factor is the average of the correlation between each wavelet coefficient in each scale and the first sequence of each wavelet coefficient in all other scales.

7. The dynamic sampling detection based series side converter control method of claim 1, wherein, The process of determining whether to reassign the amplitude of the wavelet coefficient in each scale of each modal component is: The second characteristic value of all wavelet coefficients in each scale of each modal component is threshold segmented to obtain a second threshold value. The amplitudes of all wavelet coefficients in each scale whose second characteristic values are less than the second threshold value are assigned a value of 0.

8. The series side converter control method based on dynamic sampling detection of claim 3, wherein, The passband frequency range of the second FIR filter is the union of the frequency spectrum distribution ranges of the reconstructed modal components of each modal component in the second filter component set.

Images