Harmonic suppression method and system for power distribution cabinet based on virtual synchronous sampling and dictionary learning
By employing virtual synchronous sampling and dictionary learning methods, the problems of insufficient synchronization accuracy and system stability in existing technologies are solved, achieving efficient multi-order harmonic reconstruction and stable compensation under changes in grid impedance.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- TIANJIN HUAJIE POWER EQUIP MFG CO LTD
- Filing Date
- 2026-05-20
- Publication Date
- 2026-06-16
AI Technical Summary
In existing technologies, physical phase-locked loops suffer from temperature drift and device aging issues, leading to sampling clock jitter and limited synchronization accuracy; online adaptive filtering methods involve large computational loads and are difficult to efficiently extract high-order harmonics; and the lack of online sensing and adaptive adjustment capabilities for changes in grid impedance results in insufficient system stability.
The system employs virtual synchronous sampling and dictionary learning methods, achieves phase alignment through software phase-locked loop, combines offline dictionary with online sparse reconstruction, uses transient gating to avoid parameter jumps, and ensures system stability through impedance estimation reliability and compensation limiting.
Phase alignment is achieved during grid frequency shifts and impedance changes, reducing phase mismatch, improving multi-harmonic reconstruction efficiency, and ensuring system stability and fast response during disturbances.
Smart Images
Figure CN122225451A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of active power filtering technology for distribution cabinets, and in particular to a method and system for harmonic suppression in distribution cabinets based on virtual synchronous sampling and dictionary learning. Background Technology
[0002] As a critical node for power distribution and control at the end of the power grid, the distribution cabinet generates harmonic currents due to the numerous nonlinear devices connected to its load side. This leads to voltage distortion, equipment overheating, and malfunctions of protection devices in the distribution system. Active power filters (APFs) detect harmonic components in the load current and generate a counter-phase compensating current to inject into the grid, which is a primary technical means of suppressing harmonic pollution from distribution cabinets. The harmonic suppression performance of an APF depends on two key aspects: first, the sampling synchronization accuracy of the harmonic detection stage, as synchronization errors between the sampling time and the fundamental phase of the power grid directly introduce calculation deviations in harmonic amplitude and phase; second, the real-time performance and stability of the compensation control strategy, as control parameters need to be dynamically adjusted according to operating conditions.
[0003] However, existing technical solutions have the following shortcomings: First, relying on physical phase-locked loop (PLL) hardware circuits suffers from temperature drift and component aging, leading to cumulative jitter in the sampling clock over time, thus limiting the synchronization accuracy between the sampling time and the fundamental phase of the power grid. Second, online adaptive filtering methods require iterative convergence of each harmonic component separately. As the number of harmonics requiring compensation increases, the computational load increases linearly, making it difficult to achieve high-precision extraction of higher-order harmonics within a specified sampling period under the computational constraints of embedded digital signal processors. Third, existing solutions lack online sensing and adaptive adjustment capabilities for changes in power grid impedance. When power grid impedance fluctuates due to line switching or load changes, fixed controller parameters may lead to system instability. Summary of the Invention
[0004] This invention aims to at least solve one of the technical problems existing in related technologies. To this end, this invention provides a method and system for harmonic suppression in distribution cabinets based on virtual synchronous sampling and dictionary learning. Software phase-locked loops and virtual phase grids can reduce phase mismatch caused by grid frequency offsets; offline dictionary and online sparse reconstruction can obtain multi-order harmonic reconstruction results within a limited number of iterations; transient gating avoids unbounded jumps in dictionary and control parameters during disturbances; impedance estimation reliability and compensation limiting enable the system to maintain grid-connected stability conservatively under weak grid conditions or impedance changes.
[0005] This invention provides a method for harmonic suppression in power distribution cabinets based on virtual synchronous sampling and dictionary learning, comprising: S1: Perform phase synchronization sampling on the distribution box system to obtain the phase-aligned current sampling matrix; S2: Calculate the dictionary reconstruction confidence level based on the phase-aligned current sampling matrix and the complete harmonic dictionary. If the dictionary reconstruction confidence level is less than the confidence threshold, proceed to step S3. If the dictionary reconstruction confidence level is greater than or equal to the confidence threshold, proceed to step S5. S3: Calculate the change in spectral energy based on the FFT spectral energy of the phase-aligned current sampling matrix. If the change in spectral energy is greater than the threshold for the change in spectral energy, proceed to step S4. If the change in spectral energy is less than or equal to the threshold for the change in spectral energy, proceed to step S5. S4: Limit the virtual damping and adjust the complete harmonic dictionary according to the spectrum change after limiting; S5: Calculate the compensation current reference value based on the phase-aligned current sampling matrix and the complete harmonic dictionary.
[0006] According to the present invention, a method for harmonic suppression of distribution cabinets based on virtual synchronous sampling and dictionary learning is provided. The phase-aligned current sampling matrix includes: phase-aligned current sampling values, harmonic voltage phasors, complex spectrum values of harmonic frequencies, and control parameters of harmonic frequencies.
[0007] According to the present invention, a method for harmonic suppression of power distribution cabinets based on virtual synchronous sampling and dictionary learning is provided, and step S1 is as follows: S11: Calculate the virtual phase sampling grid points based on the fundamental phase of the phase-locked loop output: in, For the p-th virtual phase sampling grid point, For grid point numbers, This represents the number of phase grid points for each power frequency cycle. S12: Obtain the phase alignment resampling matrix by performing phase alignment resampling based on the virtual phase sampling grid points: in, This is the p-th phase-aligned current sample value. This is a timestamped load current sampling sequence. The fundamental phase is obtained by software digital phase-locked loop. This is the interpolation function.
[0008] According to the present invention, a method for harmonic suppression in power distribution cabinets based on virtual synchronous sampling and dictionary learning is provided. The process of step S2 is as follows: S21: Calculate the optimal solution for the sparse coefficients based on the phase-aligned current sample values: in, This is the optimal solution for sparse coefficients. These are phase-aligned current sample values. For offline training of an incomplete harmonic dictionary, For regularization parameters, To achieve maximum sparsity, The sparsity coefficient is . It is a zero-norm. For a first-order norm, It is a quadratic norm. To obtain the function with the minimum sparsity coefficients, Indicates a precondition; S22: Calculate the confidence level of dictionary reconstruction based on the optimal solution of sparse coefficients: in, To improve the credibility of dictionary reconstruction, To reconstruct the error threshold; S23: If the confidence level of dictionary reconstruction is less than the confidence threshold, proceed to step S3; if the confidence level of dictionary reconstruction is greater than or equal to the confidence threshold, proceed to step S5.
[0009] According to the present invention, a method for harmonic suppression in power distribution cabinets based on virtual synchronous sampling and dictionary learning, the process of step S3 is as follows: S31: Calculate the spectral energy change based on the complex spectral values of the harmonic frequencies in the phase-aligned current sampling matrix: in, This represents the change in spectral energy. The set of harmonic orders to be monitored. This represents the complex spectral value of the h-th harmonic frequency point in the current period. This is the spectral value corresponding to the previous period; S32: If the change in spectral energy is greater than the threshold for the change in spectral energy, proceed to step S4; if the change in spectral energy is less than or equal to the threshold for the change in spectral energy, proceed to step S5.
[0010] According to the present invention, a method for harmonic suppression of power distribution cabinets based on virtual synchronous sampling and dictionary learning is provided. The process of step S4 is as follows: S41: Perform virtual damping limiting based on the control parameters of the harmonic frequency points in the phase-aligned current sampling matrix: in, The control parameters are for the h-th harmonic in the k-th period. For the target parameter, The previous valid parameter, For the maximum change, It is an interpolation function; S42: If the change in spectral energy is greater than the threshold and continues to exist in multiple consecutive steady-state windows after the transient ends, then a candidate dictionary is generated using the data from the most recent steady-state window, and the complete harmonic dictionary is updated accordingly.
[0011] According to the present invention, a method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning is provided. The process of step S5 is as follows: S51: Calculate and update the equivalent impedance based on harmonic voltage phasors and current phasors: in, Let h be the updated equivalent impedance of the h-th harmonic. For the first Second harmonic voltage phasor These are harmonic current phasors of the same frequency; S52: Generate a reference value for the reverse-phase compensation current based on the harmonic reconstruction current, dictionary reconstruction reliability, impedance estimation reliability, and phase-locked loop quality indicator. in, To compensate for the current reference value, To improve the credibility of dictionary reconstruction, To improve the reliability of impedance estimation, For phase-locked loop quality weighting, For harmonic compensation gain, For the h-th reconstructed harmonic current, This is the rated compensation limit. It is a saturation function.
[0012] According to the present invention, a method for harmonic suppression in distribution cabinets based on virtual synchronous sampling and dictionary learning is provided. When | |Greater than the observable threshold and Update when quality exceeds the threshold Otherwise, retain the previous effective value or enter the conservative damping parameter.
[0013] This invention also provides a harmonic suppression system for power distribution cabinets based on virtual synchronous sampling and dictionary learning, comprising: Data acquisition module: used to perform phase synchronization sampling of the distribution box system to obtain a phase-aligned current sampling matrix; Dictionary credibility judgment module: used to calculate the dictionary reconstruction credibility based on the phase-aligned current sampling matrix and the complete harmonic dictionary. If the dictionary reconstruction credibility is less than the credibility threshold, the spectrum energy judgment module is executed. If the dictionary reconstruction credibility is greater than or equal to the credibility threshold, the current prediction module is executed. Spectrum energy judgment module: used to calculate the spectrum energy change based on the FFT spectrum energy of the phase-aligned current sampling matrix. If the spectrum energy change is greater than the spectrum energy change threshold, the damping limiting module is executed. If the spectrum energy change is less than or equal to the spectrum energy change threshold, the current prediction module is executed. Damping Limiting Module: Used to limit virtual damping and adjust the complete harmonic dictionary according to the spectrum changes after limiting; Current prediction module: used to calculate the compensation current reference value based on the phase-aligned current sampling matrix and the complete harmonic dictionary.
[0014] The above-described one or more technical solutions in the embodiments of the present invention have at least one of the following technical effects: This invention introduces multiple criteria to form an observable, verifiable, and protected closed-loop compensation link encompassing synchronous sampling, dictionary sparse reconstruction, transient response, and impedance adaptation. Software phase-locked loops and virtual phase grids reduce phase mismatch caused by grid frequency shifts; offline dictionary and online sparse reconstruction obtain multi-harmonic reconstruction results within a limited number of iterations; transient gating prevents unbounded jumps in the dictionary and control parameters during disturbances; and impedance estimation reliability and compensation limiting enable the system to maintain grid-connected stability conservatively under weak grid conditions or impedance variations.
[0015] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0017] Figure 1 This is a flowchart illustrating the harmonic suppression method for power distribution cabinets based on virtual synchronous sampling and dictionary learning provided by the present invention.
[0018] Figure 2 This is a schematic diagram of the structure of the power distribution cabinet harmonic suppression system based on virtual synchronous sampling and dictionary learning provided by the present invention.
[0019] Figure label: 101. Data acquisition module; 102. Dictionary reliability judgment module; 103. Spectrum energy judgment module; 104. Damping limiting module; 105. Current prediction module. Detailed Implementation
[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The following embodiments are used to illustrate this invention but cannot be used to limit the scope of this invention.
[0021] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0022] The following is combined with Figures 1 to 2 This invention is described.
[0023] Example like Figure 1 As shown, Figure 1 The flowchart of the harmonic suppression method for power distribution cabinets based on virtual synchronous sampling and dictionary learning provided by this invention includes: S1: Perform phase synchronization sampling on the distribution box system to obtain the phase-aligned current sampling matrix; S2: Calculate the dictionary reconstruction confidence level based on the phase-aligned current sampling matrix and the complete harmonic dictionary. If the dictionary reconstruction confidence level is less than the confidence threshold, proceed to step S3. If the dictionary reconstruction confidence level is greater than or equal to the confidence threshold, proceed to step S5. S3: Calculate the change in spectral energy based on the FFT spectral energy of the phase-aligned current sampling matrix. If the change in spectral energy is greater than the threshold for the change in spectral energy, proceed to step S4. If the change in spectral energy is less than or equal to the threshold for the change in spectral energy, proceed to step S5. S4: Limit the virtual damping and adjust the complete harmonic dictionary according to the spectrum change after limiting; S5: Calculate the compensation current reference value based on the phase-aligned current sampling matrix and the complete harmonic dictionary.
[0024] Specifically, the phase-aligned current sampling matrix includes: phase-aligned current sampling values, harmonic voltage phasors, complex spectrum values of harmonic frequencies, and control parameters of harmonic frequencies.
[0025] Specifically, step S1 is as follows: S11: Calculate the virtual phase sampling grid points based on the fundamental phase of the phase-locked loop output: in, For the p-th virtual phase sampling grid point, For grid point numbers, This represents the number of phase grid points for each power frequency cycle. S12: Obtain the phase alignment resampling matrix by performing phase alignment resampling based on the virtual phase sampling grid points: in, This is the p-th phase-aligned current sample value. This is a timestamped load current sampling sequence. The fundamental phase is obtained by software digital phase-locked loop. This is the interpolation function.
[0026] Specifically, the process of step S2 is as follows: S21: Calculate the optimal solution for the sparse coefficients based on the phase-aligned current sample values: in, This is the optimal solution for sparse coefficients. These are phase-aligned current sample values. For offline training of an incomplete harmonic dictionary, For regularization parameters, To achieve maximum sparsity, The sparsity coefficient is . It is a zero-norm. For a first-order norm, It is a quadratic norm. To obtain the function with the minimum sparsity coefficients, Indicates a precondition; S22: Calculate the confidence level of dictionary reconstruction based on the optimal solution of sparse coefficients: in, To improve the credibility of dictionary reconstruction, To reconstruct the error threshold; S23: If the confidence level of dictionary reconstruction is less than the confidence threshold, proceed to step S3; if the confidence level of dictionary reconstruction is greater than or equal to the confidence threshold, proceed to step S5.
[0027] Specifically, the process of step S3 is as follows: S31: Calculate the spectral energy change based on the complex spectral values of the harmonic frequencies in the phase-aligned current sampling matrix: in, This represents the change in spectral energy. The set of harmonic orders to be monitored. This represents the complex spectral value of the h-th harmonic frequency point in the current period. This is the spectral value corresponding to the previous period; S32: If the change in spectral energy is greater than the threshold for the change in spectral energy, proceed to step S4; if the change in spectral energy is less than or equal to the threshold for the change in spectral energy, proceed to step S5.
[0028] Specifically, the process for step S4 is as follows: S41: Perform virtual damping limiting based on the control parameters of the harmonic frequency points in the phase-aligned current sampling matrix: in, The control parameters are for the h-th harmonic in the k-th period. For the target parameter, The previous valid parameter, For the maximum change, It is an interpolation function; If within several consecutive steady-state windows after the amplitude is limited Still approaching If the boundary value is not met, the spectrum change is determined to be a non-transient disturbance, triggering the dictionary update process and executing S42; otherwise, the current dictionary is maintained and S5 is returned.
[0029] S42: If the change in spectral energy is greater than the threshold and multiple steady-state windows persist after the transient ends, the background task can generate a candidate dictionary using the most recent steady-state window. The candidate dictionary must be verified by reconstruction error, compensation current limit, and simulation playback or low-gain trial operation before switching is allowed near the current zero crossing or during low-compensation load periods. If the protection module detects an anomaly after switching, it can revert to the previous dictionary version and return to step S2. The most recent steady-state window refers to a window where the change in spectral energy is continuously lower than the threshold for a duration exceeding a preset time.
[0030] Specifically, the process of step S5 is as follows: S51: Calculate and update the equivalent impedance based on harmonic voltage phasors and current phasors: in, Let h be the updated equivalent impedance of the h-th harmonic. For the first Second harmonic voltage phasor For harmonic current phasors of the same frequency; when | |Greater than the observable threshold and Update only when the quality exceeds the threshold Otherwise, retain the previous effective value or enter the conservative damping parameter.
[0031] S52: Generate a reference value for the reverse-phase compensation current based on the harmonic reconstruction current, dictionary reconstruction reliability, impedance estimation reliability, and phase-locked loop quality indicator. in, To compensate for the current reference value, To improve the credibility of dictionary reconstruction, To improve the reliability of impedance estimation, For phase-locked loop quality weighting, For harmonic compensation gain, For the h-th reconstructed harmonic current, This is the rated compensation limit. It is a saturation function. To find the minimum value function, For the previous effective impedance, for Maximum permissible variation.
[0032] Specifically, on the same distribution cabinet test platform, this method embodiment was compared with a comparative harmonic suppression scheme in three comparative experiments: steady-state test, transient test, and grid impedance change test.
[0033] The test platform's distribution cabinet has an incoming voltage of 380V / 50Hz three-phase four-wire and is equipped with a 15kW rated nonlinear programmable load. It can simulate the harmonic current characteristics of frequency converters and switching power supplies, and can simulate transient operating conditions by switching the load. The inverter has a rated compensation capacity of 20kVA, a switching frequency of 20kHz, and uses a 32-bit floating-point digital signal processor.
[0034] Steady-state test: With a rated nonlinear load connected, the total harmonic distortion (THD) of the load current was approximately 28.3%, dominated by the 3rd, 5th, and 7th harmonics. Both the comparative scheme and this method were run. After 5 minutes of stable operation, voltage and current data at the PCC point were continuously collected for 50 power frequency cycles, and the mean and standard deviation of the THD were calculated. After compensation, the THD of the PCC point current was 4.9% with the comparative scheme, and 4.7% with this method; the steady-state compensation accuracy of the two schemes was at the same level. Regarding real-time computation time, on a 32-bit floating-point digital signal processor, the average time for single harmonic extraction with the comparative scheme was 83.4 μs, while the average time for single extraction with dictionary projection and sparse solution using this method was 30.1 μs, a reduction of 63.9%. This difference stems from the fact that this method changes harmonic extraction from iterative iteration to a single sparse projection solution; the time difference becomes more significant as the number of harmonics to be extracted increases.
[0035] Transient testing: Under stable operating conditions, the load power is suddenly increased to 200% of the rated power, maintained for 200ms, and then switched back. This test is repeated 10 times, and the average value is taken. The comparative solution shows a transient overshoot amplitude of 58% of the rated compensation current and a recovery time of 4.2 power frequency cycles. In fast response mode, this method limits the transient overshoot amplitude to 31% using a virtual damping term, and the recovery time is 2.1 power frequency cycles. This method reduces the overshoot amplitude by an average of 46.6% and shortens the recovery time by an average of 50%. The improvement is due to the virtual damping term limiting the update step size of the controller gain in the early stages of the transient response, avoiding the compensation current overshoot caused by instantaneous parameter jumps in the comparative solution.
[0036] Grid impedance variation test: A 0.5mH inductor was connected in series on the inverter output side, and the system response was observed. The test was repeated 5 times. The comparative scheme exhibited continuous oscillation after the series inductor, with an amplitude of approximately 15% of the rated output current, which persisted for more than 20 power frequency cycles without convergence, requiring manual controller reset. This method completed the estimation of new grid impedance parameters within 3 power frequency cycles. Based on the identification results, this method adjusted the output impedance, and the compensation current stabilized within 5 power frequency cycles. No continuous oscillations were observed in any of the 5 tests. The results show that the online impedance identification and adaptive controller parameter adjustment mechanism can effectively cope with sudden changes in grid impedance.
[0037] like Figure 2 As shown below, a harmonic suppression system for a distribution cabinet based on virtual synchronous sampling and dictionary learning provided by the present invention will be described. The harmonic suppression system for a distribution cabinet based on virtual synchronous sampling and dictionary learning described below can be referred to in correspondence with the harmonic suppression method for a distribution cabinet based on virtual synchronous sampling and dictionary learning described above.
[0038] Data acquisition module 101: used to perform phase synchronization sampling on the distribution box system to obtain a phase-aligned current sampling matrix; Dictionary confidence judgment module 102: is used to calculate the dictionary reconstruction confidence based on the phase-aligned current sampling matrix and the complete harmonic dictionary. If the dictionary reconstruction confidence is less than the confidence threshold, the spectrum energy judgment module is executed. If the dictionary reconstruction confidence is greater than or equal to the confidence threshold, the current prediction module is executed. Spectrum energy judgment module 103: used to calculate the spectrum energy change based on the FFT spectrum energy of the phase-aligned current sampling matrix. If the spectrum energy change is greater than the spectrum energy change threshold, the damping limiting module is executed. If the spectrum energy change is less than or equal to the spectrum energy change threshold, the current prediction module is executed. Damping limiting module 104: used to limit the virtual damping and adjust the complete harmonic dictionary according to the spectrum change after limiting; Current prediction module 105: used to calculate the compensation current reference value based on the phase-aligned current sampling matrix and the complete harmonic dictionary.
[0039] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.
[0040] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device to execute the methods described in the various embodiments or some parts of the embodiments.
[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention 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; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
[0042] It should be noted that the embodiments of this disclosure can be implemented using hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a programmable memory or a data carrier such as an optical or electronic signal carrier.
[0043] Furthermore, although the operation of the methods of this disclosure is described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Rather, the steps depicted in the flowcharts may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps. It should also be noted that the features and functions of two or more devices according to this disclosure may be embodied in one device. Conversely, the features and functions of one device described above may be further divided and embodied by multiple devices.
[0044] While this disclosure has been described with reference to several specific embodiments, it should be understood that this disclosure is not limited to the specific embodiments disclosed. This disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.
Claims
1. A method for harmonic suppression in power distribution cabinets based on virtual synchronous sampling and dictionary learning, characterized in that, include: S1: Perform phase synchronization sampling on the distribution box system to obtain the phase-aligned current sampling matrix; S2: Calculate the dictionary reconstruction confidence level based on the phase-aligned current sampling matrix and the complete harmonic dictionary. If the dictionary reconstruction confidence level is less than the confidence threshold, proceed to step S3. If the dictionary reconstruction confidence level is greater than or equal to the confidence threshold, proceed to step S5. S3: Calculate the change in spectral energy based on the FFT spectral energy of the phase-aligned current sampling matrix. If the change in spectral energy is greater than the threshold for the change in spectral energy, proceed to step S4. If the change in spectral energy is less than or equal to the threshold for the change in spectral energy, proceed to step S5. S4: Limit the virtual damping and adjust the complete harmonic dictionary according to the spectrum change after limiting; S5: Calculate the compensation current reference value based on the phase-aligned current sampling matrix and the complete harmonic dictionary.
2. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 1, characterized in that, The phase-aligned current sampling matrix includes: phase-aligned current sampling values, harmonic voltage phasors, complex spectrum values of harmonic frequencies, and control parameters of harmonic frequencies.
3. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 1, characterized in that, Step S1 is as follows: S11: Calculate the virtual phase sampling grid points based on the fundamental phase of the phase-locked loop output: in, For the p-th virtual phase sampling grid point, For grid point numbers, This represents the number of phase grid points for each power frequency cycle. S12: Obtain the phase alignment resampling matrix by performing phase alignment resampling based on the virtual phase sampling grid points: in, This is the p-th phase-aligned current sample value. This is a timestamped load current sampling sequence. The fundamental phase is obtained by software digital phase-locked loop. This is the interpolation function.
4. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 1, characterized in that, The process for step S2 is as follows: S21: Calculate the optimal solution for the sparse coefficients based on the phase-aligned current sample values: in, This is the optimal solution for sparse coefficients. These are phase-aligned current sample values. For offline training of an incomplete harmonic dictionary, For regularization parameters, To achieve maximum sparsity, The sparsity coefficient is . It is a zero-norm. For a first-order norm, It is a quadratic norm. To obtain the function with the minimum sparsity coefficients, Indicates a precondition; S22: Calculate the confidence level of dictionary reconstruction based on the optimal solution of sparse coefficients: in, To improve the credibility of dictionary reconstruction, To reconstruct the error threshold; S23: If the confidence level of dictionary reconstruction is less than the confidence threshold, proceed to step S3; if the confidence level of dictionary reconstruction is greater than or equal to the confidence threshold, proceed to step S5.
5. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 1, characterized in that, The process for step S3 is as follows: S31: Calculate the spectral energy change based on the complex spectral values of the harmonic frequencies in the phase-aligned current sampling matrix: in, This represents the change in spectral energy. The set of harmonic orders to be monitored. This represents the complex spectral value of the h-th harmonic frequency point in the current period. This is the spectral value corresponding to the previous period; S32: If the change in spectral energy is greater than the threshold for the change in spectral energy, proceed to step S4; if the change in spectral energy is less than or equal to the threshold for the change in spectral energy, proceed to step S5.
6. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 1, characterized in that, The process for step S4 is as follows: S41: Perform virtual damping limiting based on the control parameters of the harmonic frequency points in the phase-aligned current sampling matrix: in, The control parameters are for the h-th harmonic in the k-th period. For the target parameter, The previous valid parameter, For the maximum change, It is an interpolation function; S42: If the change in spectral energy is greater than the threshold and continues to exist in multiple consecutive steady-state windows after the transient ends, then a candidate dictionary is generated using the data from the most recent steady-state window, and the complete harmonic dictionary is updated accordingly.
7. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 1, characterized in that, The process for step S5 is as follows: S51: Calculate and update the equivalent impedance based on harmonic voltage phasors and current phasors: in, Let h be the updated equivalent impedance of the h-th harmonic. For the first Second harmonic voltage phasor These are harmonic current phasors of the same frequency; S52: Generate a reference value for the reverse-phase compensation current based on the harmonic reconstruction current, dictionary reconstruction reliability, impedance estimation reliability, and phase-locked loop quality indicator. in, To compensate for the current reference value, To improve the credibility of dictionary reconstruction, To improve the reliability of impedance estimation, For phase-locked loop quality weighting, For harmonic compensation gain, For the h-th reconstructed harmonic current, This is the rated compensation limit. It is a saturation function.
8. The method for harmonic suppression in a distribution cabinet based on virtual synchronous sampling and dictionary learning according to claim 7, characterized in that, When | |Greater than the observable threshold and Update when quality exceeds the threshold ; Otherwise, maintain the previous effective value or enter the conservative damping parameter.
9. A harmonic suppression system for a distribution cabinet based on virtual synchronous sampling and dictionary learning, used to execute the harmonic suppression method for a distribution cabinet based on virtual synchronous sampling and dictionary learning as described in any one of claims 1 to 8, characterized in that, include: Data acquisition module: used to perform phase synchronization sampling of the distribution box system to obtain a phase-aligned current sampling matrix; Dictionary credibility judgment module: used to calculate the dictionary reconstruction credibility based on the phase-aligned current sampling matrix and the complete harmonic dictionary. If the dictionary reconstruction credibility is less than the credibility threshold, the spectrum energy judgment module is executed. If the dictionary reconstruction credibility is greater than or equal to the credibility threshold, the current prediction module is executed. Spectrum energy judgment module: used to calculate the spectrum energy change based on the FFT spectrum energy of the phase-aligned current sampling matrix. If the spectrum energy change is greater than the spectrum energy change threshold, the damping limiting module is executed. If the spectrum energy change is less than or equal to the spectrum energy change threshold, the current prediction module is executed. Damping Limiting Module: Used to limit virtual damping and adjust the complete harmonic dictionary according to the spectrum changes after limiting; Current prediction module: used to calculate the compensation current reference value based on the phase-aligned current sampling matrix and the complete harmonic dictionary.