A power grid impedance adaptive power quality compensation system based on perturbation injection

Abstract

The present application relates to the technical field of power quality treatment, and provides a power grid impedance adaptive power quality compensation system based on perturbation injection, which comprises a grid-connected point voltage and current collection terminal, a perturbation injection terminal, a frequency domain separation and background harmonic cancellation terminal, a power grid impedance online identification terminal, a compensation instruction generation terminal and a collaborative compensation execution terminal; the power grid impedance online identification terminal is used for calculating equivalent impedance parameters of the grid-connected point in each frequency band according to the frequency band response data and constructing a power grid impedance spectrum; the compensation instruction generation terminal is used for generating a compensation instruction based on the power grid impedance spectrum and real-time power quality indexes obtained from the real-time sampling data; and the collaborative compensation execution terminal is used for executing parallel compensation and / or series compensation according to the compensation instruction. The present application has the effect of improving comprehensive power quality treatment under complex working conditions.

Description

Technical Field

[0001] This invention relates to the technical field of power quality management, and specifically to a power quality compensation system based on perturbation injection and adaptive power grid impedance. Background Technology

[0002] With the continuous connection of converters, rectifiers, frequency converters, arc loads, and other nonlinear electrical equipment to the power grid, harmonic pollution, reactive power fluctuations, voltage fluctuations, negative sequence imbalances, and the resulting power quality degradation in the power supply system are becoming increasingly prominent. Although existing power quality management devices can compensate for some harmonic or reactive power issues, in scenarios where the impedance state at the grid connection point changes dynamically with load variations and external grid conditions, if compensation control still mainly relies on preset parameters, static models, or single instantaneous power quality indicators, the problems of inaccurate identification of compensation targets and inadequate matching of compensation path allocation to actual impedance characteristics remain significant.

[0003] A search of existing publications reveals that one type of solution focuses on online identification of grid impedance at the grid connection point or the grid-connected inverter side. For example, CN110112776B discloses a grid impedance identification method for grid-connected inverters that considers grid background harmonics, calculating impedance values ​​by injecting high-frequency voltage signals and extracting high-frequency voltage and current components; CN114465289B discloses a converter synchronous stability control method and device based on real-time grid impedance identification, which calculates equivalent impedance and adjusts converter output commands by superimposing harmonic currents in the current control stage and extracting corresponding voltage and current components at the PCC point. While these solutions can improve impedance identification capabilities or grid-connected stability control, their technical focus is mainly on impedance measurement, improving impedance identification accuracy, or adjusting converter control parameters. However, disclosures on how to further couple frequency-band impedance characteristics with multi-dimensional power quality indicators such as real-time harmonic distortion, reactive power fluctuations, voltage fluctuations, and negative sequence imbalances, and generate adaptive collaborative compensation commands for parallel and series compensation, remain insufficient.

[0004] Another type of publicly available solution focuses on comprehensive compensation structures for harmonics, reactive power, or negative sequence problems. For example, CN102832630B discloses an integrated power quality management device that uses a static var generator (SVA) and a static var compensator (SVC) to achieve reactive power, harmonics, and negative sequence compensation. CN102545235B discloses a delta-connected cascaded active filter and reactive power compensator integrated compensation system for dynamic compensation of reactive power and harmonic current in medium-voltage power grids. While these solutions can achieve comprehensive compensation at the device level, their compensation actions are largely based on predetermined compensation structures and control logic. There is a lack of specific disclosures regarding technical approaches that involve actively injecting micro-perturbation signals into the grid connection point during operation, extracting incremental responses through frequency domain differential analysis before and after injection in the presence of background harmonics, further constructing the grid impedance spectrum at the grid connection point, and dynamically allocating priority amounts for parallel and series compensation accordingly.

[0005] Therefore, existing technologies still have at least the following shortcomings: First, some solutions can perform impedance identification, but there is a lack of close linkage between the impedance identification results and power quality compensation decisions, making it difficult to formulate compensation strategies that better fit the current impedance state for complex disturbance conditions. Second, some solutions can perform harmonic, reactive, or negative sequence compensation, but they lack sufficient perception of the dynamic changes in grid connection impedance at different frequency bands, making it difficult to maintain the pertinence and continuity of compensation decisions in scenarios with rapid changes in impedance state, persistent background harmonics, or concurrent compound disturbances. Third, for the coordinated allocation of parallel and series compensation, most solutions still lack an adaptive decision-making mechanism based on the joint analysis of online impedance spectrum and real-time power quality indicators. Based on this, it is still necessary to propose a power quality compensation system that can be oriented towards the real-time operating state of the grid connection point, utilize perturbation injection to obtain frequency response information, extract effective incremental response under background harmonic conditions, construct the grid impedance spectrum online, and further generate adaptive coordinated compensation commands. Summary of the Invention

[0006] The purpose of this invention is to address the aforementioned shortcomings by proposing a power quality compensation system based on perturbation injection that adapts to grid impedance.

[0007] The present invention adopts the following technical solution:

[0008] An adaptive power quality compensation system based on grid impedance based on perturbation injection includes a grid connection point voltage and current acquisition terminal, a perturbation injection terminal, a frequency domain separation and background harmonic cancellation terminal, a grid impedance online identification terminal, a compensation command generation terminal, and a collaborative compensation execution terminal.

[0009] The grid connection point voltage and current acquisition terminal is used to acquire the voltage and current signals of the grid connection point and output real-time sampling data;

[0010] The perturbation injection terminal is connected to the grid connection point voltage and current acquisition terminal and is used to inject multi-frequency small-amplitude perturbation signals into the grid connection point according to a preset frequency, preset amplitude and preset injection period, so that the grid connection point can form voltage and current response data containing perturbation response information.

[0011] The frequency domain separation and background harmonic cancellation terminal is connected to the grid connection point voltage and current acquisition terminal and the micro-perturbation injection terminal, respectively. It is used to perform frequency domain separation processing on the real-time sampling data and the voltage and current response data, and to cancel the background harmonic influence based on the frequency domain difference of at least two sampling results, so as to obtain the frequency band response data corresponding to the micro-perturbation signal.

[0012] The online grid impedance identification terminal is connected to the frequency domain separation and background harmonic cancellation terminal, and is used to calculate the equivalent impedance parameters of the grid connection point in each frequency band based on the frequency band response data, and to construct the grid impedance spectrum.

[0013] The compensation instruction generation terminal is connected to the grid connection point voltage and current acquisition terminal and the grid impedance online identification terminal, respectively. It is used to generate compensation instructions based on the grid impedance spectrum and the real-time power quality indicators obtained from the real-time sampling data. The real-time power quality indicators include at least one or more of the following: harmonic distortion indicators, voltage fluctuation indicators, negative sequence imbalance indicators, and reactive power fluctuation indicators.

[0014] The collaborative compensation execution terminal is connected to the compensation instruction generation terminal and is used to perform parallel compensation and / or series compensation according to the compensation instruction to adaptively compensate for harmonic current, reactive power fluctuation, voltage distortion and three-phase imbalance at the grid connection point.

[0015] Optionally, the grid connection point voltage and current acquisition terminal includes a voltage sampling module, a current sampling module, and a timing synchronization module;

[0016] The voltage sampling module is used to collect the three-phase voltage signal of the grid connection point, the current sampling module is used to collect the three-phase current signal of the grid connection point, and the timing synchronization module is used to perform time alignment processing on the collected three-phase voltage signal and three-phase current signal to obtain real-time sampling data corresponding to the same sampling time.

[0017] Optionally, the perturbation injection terminal is used to inject small-amplitude sinusoidal perturbation signals into the grid connection point sequentially or simultaneously according to multiple preset frequency points within a preset injection window, and to limit the injection amplitude corresponding to each preset frequency point so that the injection process of the perturbation signal does not change the normal power supply state of the grid connection point.

[0018] Optionally, the frequency domain separation and background harmonic cancellation terminal is used to acquire the first frequency domain sampling result before perturbation injection and the second frequency domain sampling result after perturbation injection, and to perform differential operation on the first frequency domain sampling result and the second frequency domain sampling result to separate the incremental response component caused by the perturbation signal from the mixed spectrum, thereby obtaining the frequency segment response data corresponding to each preset frequency point.

[0019] Optionally, the online grid impedance identification terminal is used to calculate the equivalent impedance parameters corresponding to each preset frequency point based on the voltage increment response and current increment response at each preset frequency point, and arrange the equivalent impedance parameters in frequency order to construct a grid impedance spectrum that reflects the frequency response characteristics of the grid connection point.

[0020] Optionally, the compensation instruction generation terminal is used to determine the priority of parallel compensation and the priority of series compensation based on the impedance characteristics of different frequency bands in the power grid impedance spectrum and the real-time power quality index, and generate corresponding compensation instructions.

[0021] The collaborative compensation execution terminal is used to perform parallel compensation when harmonic current and reactive power fluctuations are dominant, to perform series compensation when voltage distortion and three-phase imbalance are dominant, or to perform collaborative compensation of parallel and series compensation under combined disturbance conditions.

[0022] The beneficial effects achieved by this invention are:

[0023] By setting up terminals for acquiring grid connection point voltage and current, injecting disturbances, separating frequency domains and canceling background harmonics, identifying grid impedance online, generating compensation commands, and executing collaborative compensation, a complete closed-loop processing chain is formed, from acquiring grid connection point status, actively inducing disturbances, separating responses, identifying impedance online, generating compensation decisions, to executing compensation actions. This enables the system to synchronously sense and coordinately adjust grid impedance characteristics and power quality status while operating in grid-connected mode. Compared to compensation control based solely on static parameters or a single power quality index, using both grid impedance spectrum and real-time power quality indicators as the basis for compensation decisions allows compensation commands to reflect not only the current disturbance manifestations but also the actual response characteristics of the grid connection point at different frequency bands, thereby improving the matching degree between compensation actions and the current grid status. Furthermore, by executing parallel and / or series compensation through the collaborative compensation execution terminal, targeted adaptive compensation can be implemented for different types of disturbances such as harmonic currents, reactive power fluctuations, voltage distortion, and three-phase imbalance, which is beneficial for improving the overall power quality management effect under complex operating conditions.

[0024] By setting up voltage sampling modules, current sampling modules, and timing synchronization modules in the grid-connected point voltage and current acquisition terminal, the three-phase voltage and current signals at the grid connection point can form real-time sampled data corresponding to the same sampling time under a unified sampling benchmark. This provides a consistent data foundation for subsequent frequency domain separation processing, incremental response extraction, and equivalent impedance parameter calculation. Since the impedance identification process is sensitive to the amplitude and phase relationships between voltage and current quantities, deviations in sampling timing can easily lead to distorted impedance calculation results. Therefore, setting up a timing synchronization module to perform time alignment processing helps reduce data offset problems caused by asynchronous sampling, thereby improving the accuracy and stability of subsequent online grid impedance identification. Simultaneously, using three-phase voltage and current signals as acquisition objects also provides more complete basic data support for negative sequence imbalance analysis, three-phase imbalance compensation judgment, and composite disturbance identification.

[0025] By injecting small-amplitude sinusoidal perturbation signals sequentially or simultaneously into the grid connection point at multiple preset frequency points within a preset injection window, and by limiting the injection amplitude corresponding to each preset frequency point, the system can actively acquire multi-band frequency response information without significantly altering the normal power supply state of the grid connection point. Compared to methods that rely on natural disturbances or offline test parameters to estimate grid impedance, actively injecting controllable perturbation signals can more effectively stimulate the response characteristics of the grid connection point at different frequency bands during operation, thereby improving the real-time performance and controllability of impedance identification. Furthermore, by setting a preset injection window and amplitude limiting control, excessive additional disturbances to the original power supply state during the perturbation injection process can be avoided, which is beneficial for balancing impedance identification effectiveness and operational safety in engineering applications. Especially in scenarios with continuous load changes or fluctuations in the external power grid state, this method is beneficial for improving the system's ability to acquire effective impedance information.

[0026] By having the frequency domain separation and background harmonic cancellation terminals separately acquire the first frequency domain sampling result before perturbation injection and the second frequency domain sampling result after perturbation injection, and performing differential operations on the two, the incremental response component caused by the perturbation signal is separated from the mixed spectrum. This allows the system to more effectively distinguish the target perturbation response from the original background harmonic components and the original operating disturbance components. Since harmonic background, voltage fluctuations, and other non-target spectral components usually exist at the grid connection point during actual operation, directly performing impedance identification on the single-sample spectrum result can easily lead to the background components being mistakenly included in the impedance calculation result. However, by performing differential processing on the frequency domain results before and after perturbation injection, stable background components that do not have a direct correspondence with the current perturbation can be weakened, thereby improving the relevance and purity of the extracted frequency band response data. This is beneficial to improving the effectiveness of the input data acquired by the subsequent online grid impedance identification terminal and reducing the interference of background harmonics on the impedance spectrum construction process.

[0027] By enabling the online grid impedance identification terminal to calculate equivalent impedance parameters based on the voltage and current increment responses at each preset frequency point, and arranging these parameters in frequency order, a grid impedance spectrum reflecting the frequency response characteristics of the grid connection point is constructed. This allows the system to characterize the impedance features of the grid connection point across different frequency ranges in a segmented manner. Compared to using only a single frequency impedance value or an overall impedance approximation, constructing a grid impedance spectrum provides a more detailed reflection of impedance differences, coupling characteristics, and response trends across different frequency bands. This provides a more targeted basis for the subsequent compensation command generation terminal to identify disturbance types, determine compensation paths, and allocate compensation priorities. Especially in scenarios with complex harmonic disturbance frequency band distributions and significant frequency variations in impedance characteristics, the constructed grid impedance spectrum helps improve the consistency between compensation decisions and the actual response characteristics of the grid connection point.

[0028] By having the compensation command generation terminal determine the priority values ​​for parallel and series compensation based on the impedance characteristics of different frequency bands in the grid impedance spectrum and real-time power quality indicators, and generating corresponding compensation commands, the system enables the collaborative compensation execution terminal to perform parallel, series, or collaborative compensation according to different disturbance dominance states. This allows the system to implement more directional compensation control for different dominant disturbance types. Compared to schemes that use a fixed single compensation path or simply switch compensation methods based on experience, introducing the priority values ​​for parallel and series compensation into the compensation decision-making process allows the selection of compensation paths and the allocation of compensation intensity to simultaneously consider the current disturbance performance and current impedance characteristics, thereby improving the adaptability of compensation actions to actual operating conditions. Furthermore, performing parallel compensation when harmonic currents and reactive power fluctuations are dominant, performing series compensation when voltage distortion and three-phase imbalance are dominant, and performing collaborative compensation under complex disturbance conditions helps avoid the problem of insufficient coverage of a single compensation method under complex operating conditions, thereby improving the overall compensation effect and operational flexibility of the system in scenarios with complex power quality problems.

[0029] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the overall structure of the present invention;

[0031] Figure 2 This is a schematic diagram of the method flow of an adaptive power quality compensation method for grid impedance based on perturbation injection in this invention.

[0032] Figure 3This is a statistical chart showing the trend of credibility change under the combined effect of "background contamination suppression" and "effective response enhancement" in this invention.

[0033] Figure 4 This is a statistical graph of the impedance spectrum stability index Sz in this invention. Detailed Implementation

[0034] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated in advance. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0035] Example 1: This example provides a power quality compensation system based on perturbation injection and adaptive power grid impedance. Combined with... Figure 1 As shown, a grid impedance adaptive power quality compensation system based on perturbation injection includes a grid connection point voltage and current acquisition terminal, a perturbation injection terminal, a frequency domain separation and background harmonic cancellation terminal, a grid impedance online identification terminal, a compensation command generation terminal, and a collaborative compensation execution terminal.

[0036] The grid connection point voltage and current acquisition terminal is used to acquire the voltage and current signals of the grid connection point and output real-time sampling data;

[0037] The perturbation injection terminal is connected to the grid connection point voltage and current acquisition terminal and is used to inject multi-frequency small-amplitude perturbation signals into the grid connection point according to a preset frequency, preset amplitude and preset injection period, so that the grid connection point can form voltage and current response data containing perturbation response information.

[0038] The frequency domain separation and background harmonic cancellation terminal is connected to the grid connection point voltage and current acquisition terminal and the micro-perturbation injection terminal, respectively. It is used to perform frequency domain separation processing on the real-time sampling data and the voltage and current response data, and to cancel the background harmonic influence based on the frequency domain difference of at least two sampling results, so as to obtain the frequency band response data corresponding to the micro-perturbation signal.

[0039] The online grid impedance identification terminal is connected to the frequency domain separation and background harmonic cancellation terminal, and is used to calculate the equivalent impedance parameters of the grid connection point in each frequency band based on the frequency band response data, and to construct the grid impedance spectrum.

[0040] The compensation instruction generation terminal is connected to the grid connection point voltage and current acquisition terminal and the grid impedance online identification terminal, respectively. It is used to generate compensation instructions based on the grid impedance spectrum and the real-time power quality indicators obtained from the real-time sampling data. The real-time power quality indicators include at least one or more of the following: harmonic distortion indicators, voltage fluctuation indicators, negative sequence imbalance indicators, and reactive power fluctuation indicators.

[0041] The collaborative compensation execution terminal is connected to the compensation instruction generation terminal and is used to perform parallel compensation and / or series compensation according to the compensation instruction to adaptively compensate for harmonic current, reactive power fluctuation, voltage distortion and three-phase imbalance at the grid connection point.

[0042] Optionally, the grid connection point voltage and current acquisition terminal includes a voltage sampling module, a current sampling module, and a timing synchronization module;

[0043] The voltage sampling module is used to collect the three-phase voltage signal of the grid connection point, the current sampling module is used to collect the three-phase current signal of the grid connection point, and the timing synchronization module is used to perform time alignment processing on the collected three-phase voltage signal and three-phase current signal to obtain real-time sampling data corresponding to the same sampling time.

[0044] Optionally, the perturbation injection terminal is used to inject small-amplitude sinusoidal perturbation signals into the grid connection point sequentially or simultaneously according to multiple preset frequency points within a preset injection window, and to limit the injection amplitude corresponding to each preset frequency point so that the injection process of the perturbation signal does not change the normal power supply state of the grid connection point.

[0045] Optionally, the frequency domain separation and background harmonic cancellation terminal is used to acquire the first frequency domain sampling result before perturbation injection and the second frequency domain sampling result after perturbation injection, and to perform differential operation on the first frequency domain sampling result and the second frequency domain sampling result to separate the incremental response component caused by the perturbation signal from the mixed spectrum, thereby obtaining the frequency segment response data corresponding to each preset frequency point.

[0046] Optionally, the online grid impedance identification terminal is used to calculate the equivalent impedance parameters corresponding to each preset frequency point based on the voltage increment response and current increment response at each preset frequency point, and arrange the equivalent impedance parameters in frequency order to construct a grid impedance spectrum that reflects the frequency response characteristics of the grid connection point.

[0047] Optionally, the compensation instruction generation terminal is used to determine the priority of parallel compensation and the priority of series compensation based on the impedance characteristics of different frequency bands in the power grid impedance spectrum and the real-time power quality index, and generate corresponding compensation instructions.

[0048] The collaborative compensation execution terminal is used to perform parallel compensation when harmonic current and reactive power fluctuations are dominant, to perform series compensation when voltage distortion and three-phase imbalance are dominant, or to perform collaborative compensation of parallel and series compensation under combined disturbance conditions.

[0049] In this embodiment, the working process of the system includes the following steps:

[0050] First, the grid connection point voltage and current acquisition terminal continuously acquires the three-phase voltage and current signals at the grid connection point. The timing synchronization module then performs time alignment processing on the acquired three-phase voltage and current signals to form real-time sampling data corresponding to the same sampling moment. The reason for performing time alignment processing first is that subsequent frequency domain separation, incremental response extraction, and equivalent impedance parameter calculation require ensuring that the voltage and current quantities correspond to the same sampling reference. This avoids phase distortion due to sampling timing offset, which would affect the identification results of the grid connection point impedance characteristics.

[0051] Subsequently, the perturbation injection terminal determines the current operating status of the grid-connected point based on the real-time sampling data. When the preset injection conditions are met, it injects multi-frequency small-amplitude perturbation signals into the grid-connected point within a preset injection window according to multiple preset frequency points, preset amplitudes, and preset injection periods. The preset injection window and amplitude limiting control are set here because the perturbation injection terminal needs to obtain identifiable frequency response information without significantly altering the normal power supply status of the grid-connected point. By controlling the injection amplitude within a preset range and combining it with injection period control, the perturbation signal can be detectable while avoiding significant additional disturbances to the original power supply quality and load operating status.

[0052] Before and after perturbation injection, the frequency domain separation and background harmonic cancellation terminal acquires the first frequency domain sampling result before perturbation injection and the second frequency domain sampling result after perturbation injection, respectively. Frequency domain differential processing is then performed on the first and second frequency domain sampling results to separate the incremental response component caused by the perturbation signal from the mixed spectrum, thereby obtaining the frequency band response data corresponding to each preset frequency point. The reason for using this processing method is that grid connection points may inherently contain background harmonics, voltage fluctuations, negative sequence components, and other operational disturbances. If impedance calculations are performed directly based on a single sampling result, background components may be mistakenly identified as perturbation response components. By differentially processing the frequency domain sampling results before and after perturbation injection, stable background components unrelated to perturbation injection can be canceled, thereby improving the specificity of the extracted frequency band response data.

[0053] Furthermore, the online grid impedance identification terminal receives the frequency band response data and extracts the voltage increment response and current increment response at each preset frequency point. It calculates the equivalent impedance parameters based on the voltage increment response and current increment response at the corresponding frequency point, and then arranges the equivalent impedance parameters in frequency order to form a grid impedance spectrum reflecting the frequency response characteristics of the grid connection point. The reason for using frequency band calculation and constructing the impedance spectrum in frequency order is that the impedance characteristics of the grid connection point are usually different in different frequency bands. If judgment is based solely on a single frequency point or a single overall indicator, it is difficult to accurately reflect the differences in grid response under different disturbance types. Organizing the impedance parameters of each frequency band in frequency order provides a more detailed basis for subsequent compensation object identification and compensation method allocation.

[0054] After obtaining the grid impedance spectrum, the compensation command generation terminal jointly analyzes the grid impedance spectrum with the real-time power quality indicators obtained from the real-time sampling data to determine the main disturbance types at the current grid connection point and the corresponding compensation requirements. Specifically, when harmonic distortion and reactive power fluctuation indicators are dominant, the compensation command generation terminal determines the compensation priority corresponding to parallel compensation; when voltage distortion and negative sequence imbalance indicators are dominant, the compensation command generation terminal determines the compensation priority corresponding to series compensation; when multiple disturbances exist simultaneously at the grid connection point and exhibit a composite disturbance condition, the compensation command generation terminal determines the priority for parallel compensation and the priority for series compensation respectively, and generates corresponding compensation commands accordingly. The grid impedance spectrum and the real-time power quality indicators are used together as the basis for compensation decisions because while the real-time power quality indicators alone can reflect the current disturbance characteristics, they are insufficient to reflect the compensability and coupling characteristics of the grid connection point at different frequency bands; combining the two allows for a more appropriate allocation of compensation resources to the current impedance state of the grid connection point.

[0055] Subsequently, the collaborative compensation execution terminal receives the compensation instruction and performs parallel compensation and / or series compensation according to the instruction. Specifically, when harmonic current and reactive power fluctuations dominate, the collaborative compensation execution terminal performs parallel compensation to compensate for current-side disturbances at the grid connection point; when voltage distortion and three-phase imbalance dominate, the collaborative compensation execution terminal performs series compensation to compensate for voltage-side disturbances at the grid connection point; under combined disturbance conditions, the collaborative compensation execution terminal simultaneously performs coordinated compensation of parallel and series compensation to suppress both current-side and voltage-side disturbances. By adopting the above-mentioned collaborative compensation method, the problem of insufficient coverage of compensation objects under combined disturbance conditions by a single compensation method can be avoided.

[0056] Furthermore, after a compensation cycle ends, the grid connection point voltage and current acquisition terminal continues to acquire the compensated grid connection point voltage and current signals. The compensation command generation terminal updates the real-time power quality index based on the post-compensation sampling results, and the grid impedance online identification terminal updates the grid impedance spectrum based on the new frequency band response data. When the updated grid impedance spectrum or real-time power quality index changes relative to the previous compensation cycle, the compensation command generation terminal recalculates the parallel compensation priority and series compensation priority, and outputs the updated compensation command to the collaborative compensation execution terminal to form a continuously operating online identification and adaptive compensation closed loop. The reason for setting up this continuous update process is that the grid connection point impedance characteristics and power quality status will dynamically change with load changes, external grid status changes, and the compensation process itself. If the compensation command remains unchanged for a long time, compensation lag or compensation deviation is likely to occur.

[0057] As a preferred implementation, when injecting multi-frequency small-amplitude perturbation signals, the perturbation injection terminal can inject the perturbation signals corresponding to each preset frequency point in batches, from low frequency to high frequency, or inject the corresponding perturbation signals sequentially in groups according to preset frequency points. Using batch injection or group injection helps reduce the superimposed disturbances to the original operating state of the grid connection point caused by simultaneous injection of multiple frequency components. It also facilitates the frequency domain separation and background harmonic cancellation terminal in extracting the corresponding responses of each frequency point, thereby improving the distinguishability of the frequency band response data.

[0058] As a preferred implementation, after acquiring the first frequency domain sampling result and the second frequency domain sampling result, the frequency domain separation and background harmonic cancellation terminal can continuously compare the differential results within adjacent sampling periods and retain the incremental response components that meet the preset consistency conditions in multiple consecutive sampling periods as valid response components. The purpose of setting up this continuous comparison process is to reduce the impact of occasional transient disturbances, random noise, or short-term background fluctuations on the incremental response extraction results, making the frequency band response data entering the online grid impedance identification terminal more stable.

[0059] As a preferred implementation, after constructing the power grid impedance spectrum, the online power grid impedance identification terminal can further smooth or segment the spectrum based on the impedance variation amplitude between adjacent frequency points to improve the discriminability of impedance characteristics in different frequency bands. The reason for this processing is that, during actual sampling, incremental response data at different frequency points may experience local jumps due to measurement fluctuations. Directly using these local jump results for compensation decisions may lead to unstable allocation of compensation priorities. Further processing of the power grid impedance spectrum improves the continuity of the compensation decision-making process.

[0060] As a preferred implementation, after determining the priority amounts for parallel and series compensation, the compensation instruction generation terminal can also perform a transitional correction on the current compensation instruction based on the execution result of the previous compensation cycle, so that the change in compensation amount between adjacent compensation cycles remains within a preset range. The reason for setting this transitional correction is that when the grid connection point disturbance state changes rapidly, if the compensation amount abruptly changes between adjacent cycles, it may introduce new voltage and current fluctuations; by performing a transitional correction on the compensation instruction, the stability of the collaborative compensation execution terminal during mode switching can be improved.

[0061] As a preferred implementation, when performing coordinated compensation of parallel and series compensation, the coordinated compensation execution terminal can first perform pre-compensation for harmonic current and reactive power fluctuations based on the priority of parallel compensation, and then perform further compensation for voltage distortion and three-phase imbalance based on the priority of series compensation, or perform series compensation first and then parallel compensation. Adopting a phased approach to coordinated compensation allows for adaptation of the compensation sequence according to the different dominant disturbance sides, thereby improving the targeted nature of compensation for complex disturbance conditions.

[0062] In summary, this embodiment achieves online acquisition, online identification, and adaptive compensation of grid connection point impedance characteristics and power quality status through the coordinated operation of the grid connection point voltage and current acquisition terminal, the perturbation injection terminal, the frequency domain separation and background harmonic cancellation terminal, the grid impedance online identification terminal, the compensation command generation terminal, and the collaborative compensation execution terminal. Furthermore, by optimizing the perturbation injection process, the frequency domain differential process, the impedance spectrum construction process, and the compensation switching process, the stability of the impedance identification results and the continuity of the compensation execution process are improved, making the system more suitable for grid application scenarios with dynamically changing impedance status and complex perturbation types.

[0063] An adaptive power quality compensation method based on perturbation injection for grid impedance is proposed, which is used to perform adaptive power quality compensation based on perturbation injection for grid impedance, combined with... Figure 2 As shown, the adaptive power quality compensation method based on perturbation injection for grid impedance includes:

[0064] S1 collects the voltage and current signals at the grid connection point and outputs real-time sampling data;

[0065] S2, inject multi-frequency small-amplitude perturbation signals into the grid connection point according to a preset frequency, preset amplitude and preset injection period, so that the grid connection point can form voltage and current response data containing perturbation response information;

[0066] S3, perform frequency domain separation processing on the real-time sampling data and the voltage and current response data, and cancel the background harmonic influence based on the frequency domain difference of at least two sampling results to obtain the frequency band response data corresponding to the perturbation signal;

[0067] S4. Calculate the equivalent impedance parameters of the grid connection point in each frequency band based on the frequency band response data, and construct the grid impedance spectrum.

[0068] S5. Generate a compensation command based on the grid impedance spectrum and the real-time power quality index obtained from the real-time sampling data. The real-time power quality index includes at least one or more of the following: harmonic distortion index, voltage fluctuation index, negative sequence imbalance index, and reactive power fluctuation index.

[0069] S6, execute parallel compensation and / or series compensation according to the compensation instruction to adaptively compensate for harmonic current, reactive power fluctuation, voltage distortion and three-phase imbalance at the grid connection point.

[0070] Example 2: Based on Example 1, combined with Figure 3 and Figure 4 As shown, this embodiment further elaborates on the collaborative processing between the frequency domain separation and background harmonic cancellation terminal, the grid impedance online identification terminal, and the compensation command generation terminal. The purpose of this embodiment is to achieve more stable separation of the incremental response before and after disturbance injection, more reliable online identification of the equivalent impedance parameters of the grid connection point in each frequency band, and, based on this, generate compensation commands more suitable for the current impedance and disturbance states, under conditions of continuously changing background harmonics, load fluctuations, and compensation states at the grid connection point. To this end, this embodiment does not directly use single frequency domain results or a single linear weighting method to determine the compensation direction. Instead, it first constructs a disturbance response reliability index, then an impedance spectrum stability index, followed by a composite disturbance dominance bias index, and finally generates parallel compensation priority quantities and series compensation priority quantities, so that the calculation results are ultimately implemented in the action control of the collaborative compensation execution terminal.

[0071] Specifically, suppose the perturbation injection terminal injects a first perturbation into the grid connection point within a preset injection window. The perturbation angular frequency corresponding to each preset frequency point is: Before and after the perturbation injection, the voltage frequency domain response at that frequency point is obtained by the frequency domain separation and background harmonic cancellation terminals, respectively. , and current frequency domain response , .in, Indicates the frequency point before perturbation injection. The extracted complex voltage frequency domain component, This indicates that after perturbation injection, at the frequency point The extracted complex voltage frequency domain component, Indicates the frequency point before perturbation injection. The extracted complex current frequency domain components, This indicates that after perturbation injection, at the frequency point The complex current frequency domain component extracted at that frequency. Then the voltage increment response at that frequency. With current increment response They are represented as follows:

[0072] ;

[0073] ;

[0074] The reason for using the above incremental expression method is that the original frequency domain results before and after the perturbation injection simultaneously include the inherent background harmonic components of the grid connection point, the disturbance components caused by the original load, and the response components introduced by this perturbation injection. If we directly use... and Calculating impedance can easily lead to the miscalculation of existing background components into the current identification result. However, by extracting it incrementally, stable background terms that are not directly related to the current perturbation can be weakened during the difference process, thus making the subsequent impedance identification closer to the real response triggered by the current perturbation.

[0075] Furthermore, considering that short-term fluctuations, measurement noise, or occasional spectral drift may still exist at some frequency points in actual operating conditions, the frequency domain separation and background harmonic cancellation terminal further constructs a second... The reliability index of the perturbation response corresponding to each frequency point The perturbation response reliability index Represented as:

[0076] ;

[0077] in, Indicates the first Background harmonic residual factors at each frequency point Indicates the first Discrete offset factor between adjacent sampling periods of each frequency point Indicates the first Incremental response significance factor at each frequency point , , Let represent the background residual suppression coefficient, the discrete offset suppression coefficient, and the significance amplification coefficient, respectively. , , All are preset constants greater than zero. The reason for adopting the nonlinear form of the above-mentioned exponential decay term combined with the hyperbolic tangent amplification is that this embodiment needs to quickly reduce the credibility of the corresponding frequency point when the background residue is large and the periodic fluctuation is large, while improving the retention of the corresponding frequency point when the incremental response is indeed clearly distinguishable; if a simple linear addition method is used, it is difficult to form a sufficiently obvious distinction between high noise frequency points and high credibility frequency points.

[0078] Specifically, the background harmonic residual factor It can be represented as:

[0079] ;

[0080] in, Indicates the first The residual amplitude of background harmonics extracted within a preset frequency band near each frequency point. This represents the current increment response reduction factor. This represents a small positive number to prevent the denominator from being zero. The background harmonic residual factor... Used to reflect the The proportion of background components that are not sufficiently weakened by the differential process at each frequency point relative to the effective incremental response, when An increase in frequency indicates that the incremental response at that frequency point is more likely to be contaminated by residual background components, therefore it is necessary to use the aforementioned methods. This reduces the reliability of that frequency point. The reason for setting this parameter is that the magnitude of the incremental response alone cannot distinguish whether the increment has been distorted by background fluctuations, thus introducing [this parameter]. Subsequently, the frequency domain separation and background harmonic cancellation terminal can avoid directly sending the heavily polluted frequency points into the online impedance identification process.

[0081] The discrete offset factor of adjacent sampling periods It can be represented as:

[0082] ;

[0083] in, Indicates the first The first sampling period corresponds to the first Voltage increment response at each frequency point Indicates the first The first sampling period corresponds to the first Incremental current response at each frequency point This indicates the number of sampling periods used for consecutive comparisons. This represents the current discrete offset weighting coefficient. The discrete offset factor for adjacent sampling periods. Used to characterize the degree of incremental response jump at the same frequency point within adjacent sampling periods, when A large value indicates that the perturbation response at that frequency is not stable enough and may be significantly affected by occasional fluctuations, sampling glitches, or temporary disturbances. Therefore, the degree to which this frequency participates in impedance calculation should be reduced. The reason for setting this parameter is that although the incremental response at a single moment may seem effective, if it exhibits a large jump in adjacent periods, directly using this result to construct the impedance spectrum will cause subsequent compensation commands to oscillate frequently with occasional fluctuations.

[0084] The incremental response significance factor It can be represented as:

[0085] ;

[0086] in, This represents the current significance conversion factor. Indicates the first The noise baseline amplitude corresponding to each frequency point. The incremental response significance factor. Used to measure the prominence of the effective incremental response at the current frequency relative to the local noise baseline, when An increase in value indicates a clearer perturbation response at that frequency, making it more suitable as a basis for impedance identification. The reason for this is that when When in the low-value range, this item can quickly differentiate between low-significance and medium-significance frequencies, while when... Once a certain range is exceeded, the term gradually approaches saturation, which can avoid unnecessary over-amplification of the overall calculation due to excessively high significance frequencies.

[0087] After obtaining the perturbation response reliability index corresponding to each frequency point Subsequently, the online grid impedance identification terminal does not directly use... Instead of using the final impedance result, it constructs equivalent impedance parameters under credible constraints. Its expression is:

[0088] ;

[0089] in, Indicates the first The historical equivalent impedance parameters retained for each frequency point in the previous update cycle. The meaning of the above expression is that when the reliability of the current frequency point... At higher levels, More emphasis is placed on using impedance results directly calculated from the current incremental response; when At lower levels, To mitigate the impact of abnormal or unstable frequencies on the real-time impedance spectrum, more historical impedance results from the previous period are retained. This historical fusion term is included because the impedance spectrum is a crucial basis for subsequent compensation decisions. Relying solely on current measurement results could lead to local distortions in the impedance spectrum during sudden disturbances or fluctuations in sampling conditions. By fusing historical information through reliability constraints, the continuity and engineering usability of the impedance spectrum can be improved.

[0090] Furthermore, to avoid the impedance results at different frequencies existing only as isolated points, the online power grid impedance identification terminal further constructs the local bending degree of the impedance spectrum. To assess the smoothness of impedance changes between adjacent frequency points, the local bending degree of the impedance spectrum. Represented as:

[0091] ;

[0092] in, Indicates the first Credible constraint equivalent impedance parameters at each frequency point Indicates the first The reliable constraint equivalent impedance parameters at each frequency point. The local bending degree of the impedance spectrum. Used to describe the impedance spectrum in the 1st The degree of local bending at each frequency point, when A large value indicates an abnormal jump in impedance change between that frequency and adjacent frequencies, which may originate from local noise, instantaneous errors, or highly sensitive resonance characteristics in actual operating conditions. The reason for setting this parameter is that although the impedance spectrum at the grid connection point may be non-flat with frequency changes, if the local bend is too large, directly allocating compensation priority based on this can easily lead to the compensation decision being overly sensitive to a single local jump point.

[0093] Based on the local bending degree of the impedance spectrum The online power grid impedance identification terminal further outputs smoothed impedance parameters. Its expression is:

[0094] ;

[0095] in, This represents the bending smoothing adjustment coefficient. Indicates the bending sensitivity index, and The meaning of the above formula is that when When smaller, Mainly maintain the impedance result at the current frequency; when When it is large, By gradually converging towards the average value of adjacent frequency points, the disruptive effect of abnormal local jumps on the overall impedance spectrum morphology is reduced. A power-law approach is adopted. The reason for using a non-linear enhancement method is that this embodiment tries to preserve realistic details when there are slight bends, while rapidly increasing the smoothing force when there are obvious jumps. Therefore, using a non-linear enhancement method is more conducive to balancing realism and stability.

[0096] After obtaining the smoothed impedance parameters for all frequency points Subsequently, the online grid impedance identification terminal constructs the grid impedance spectrum at the grid connection point according to frequency order, and further calculates the impedance spectrum stability index. The impedance spectrum stability index Represented as:

[0097] ;

[0098] in, This indicates the total number of frequency points involved in constructing the impedance spectrum. This represents the historical offset penalty coefficient. This represents the local bending penalty coefficient. The impedance spectrum stability index... Used to characterize the overall stability of the current periodic impedance spectrum relative to the historical impedance spectrum, when A higher impedance level indicates that the current impedance spectrum is generally continuous and well-connected with historical states, making it suitable for direct participation in compensation decisions; when... A lower value indicates significant fluctuations in the current impedance spectrum, necessitating enhanced transition corrections during compensation command generation. This stability index is set because directly basing compensation decisions on an unstable impedance spectrum would lead to substantial fluctuations in the compensation amount between adjacent periods. Therefore, the overall reliability of the impedance spectrum must be clearly considered during the decision-making phase.

[0099] During the compensation command generation phase, the compensation command generation terminal simultaneously receives the smoothed impedance spectrum and real-time power quality indicators. Let... Indicates harmonic distortion index, This indicates a voltage fluctuation index. Indicates a negative-order imbalance index. To represent the reactive power fluctuation index, this embodiment further constructs a composite disturbance-dominant bias index. The composite disturbance dominant bias index Represented as:

[0100] ;

[0101] in, , , These represent the bias amplification factor, reactive power conversion factor, and negative order conversion factor, respectively. This represents the set of frequency bands associated with parallel compensation. This represents the set of frequency bands associated with series compensation. The dominant bias index of the composite disturbance... Used to reflect whether the current disturbance is more inclined to be mitigated on the current side or the voltage side; when When the value approaches 1, it indicates that harmonic distortion and reactive power fluctuations are relatively more dominant, and the impedance characteristics of the corresponding parallel compensation frequency band are more prominent; when When the value approaches 0 or is low, it indicates that voltage fluctuations and negative sequence imbalances are relatively more significant, and the impedance characteristics of the series compensation frequency band are more noteworthy. The reason for simultaneously coupling real-time power quality indicators and impedance spectrum frequency division characteristics into this indicator is that... , , , The judgment can only reflect the appearance of the disturbance, but cannot reflect the actual control sensitivity of these disturbances under the current impedance state; after adding the impedance amplitude ratio of the associated frequency band, the compensation bias judgment can be more in line with the current grid connection point state.

[0102] Based on the aforementioned composite disturbance dominant bias index This embodiment further generates parallel compensation priority quantities. and series compensation priority Among them, the priority of parallel compensation Represented as:

[0103] ;

[0104] Series compensation priority Represented as:

[0105] ;

[0106] in, , , , , , These represent priority adjustment coefficients. The two priority factors mentioned above are used to measure the priority of parallel and series compensation in the allocation of compensation resources within the current cycle. The impedance spectrum stability index... The reason for multiplying all values ​​in the aforementioned priority formula is that even if a certain type of disturbance index is currently high, if the current impedance spectrum is still unstable, it is not advisable to immediately and significantly increase the corresponding compensation amount; otherwise, the compensation decision may oscillate due to impedance identification jitter. A logarithmic function is used. The reason is that when the disturbance index is in the low to medium range, the priority quantity will increase significantly as the index grows, while when the disturbance index is already in the high range, the growth trend gradually slows down, which can prevent the compensation priority quantity from amplifying without limit with a single abnormal index.

[0107] To ensure better continuity of the compensation action, the compensation command generation terminal further constructs a compensation switching vibration suppression coefficient. Its expression is:

[0108] ;

[0109] in, This indicates the priority amount of parallel compensation in the previous cycle. This indicates the priority amount of series compensation in the previous cycle. , , These represent the parallel variation penalty coefficient, the series variation penalty coefficient, and the stability penalty coefficient, respectively. The compensation switching vibration suppression coefficient... Used to suppress abrupt changes in priority quantities between adjacent compensation cycles, when the current cycle changes too rapidly compared to the previous cycle. This reduces the instantaneous adjustment range of the compensation command in the current cycle. The reason for setting this parameter is that even if the disturbance state in the current cycle does change, if the compensation amount changes too abruptly between adjacent cycles, it may introduce new voltage and current fluctuations. Therefore, it is necessary to constrain the update speed by switching the damping coefficient.

[0110] Finally, the compensation instruction generation terminal outputs the corrected parallel compensation instruction quantity. and series compensation command quantity Specifically, it can be expressed as:

[0111] ;

[0112] ;

[0113] in, Used to drive the parallel compensation section in the collaborative compensation execution terminal. Used to drive the series compensation section in the collaborative compensation execution terminal. When Greater than the preset parallel trigger threshold and When the threshold is lower than the preset series trigger threshold, the collaborative compensation execution terminal prioritizes parallel compensation; when Greater than the preset series trigger threshold and When the threshold is lower than the preset parallel trigger threshold, the collaborative compensation execution terminal prioritizes performing series compensation; when and Simultaneously, when the values ​​exceed their respective thresholds, the collaborative compensation execution terminal performs collaborative compensation combining parallel and series compensation. Thus, this embodiment forms a complete calculation-judgment-action chain consisting of incremental response extraction, confidence screening, impedance spectrum smoothing, disturbance bias discrimination, priority quantity generation, and compensation switching suppression.

[0114] The physical or logical meanings of the main parameters in this embodiment are further explained below: Used to identify the target of perturbation injection and response extraction. The preset angular frequencies are the basic indexes for constructing frequency band response data and impedance spectra; and These represent the voltage increment response and current increment response caused by perturbation injection at the same frequency point, respectively, and are the direct basis for distinguishing the current injection response from the original background components. Used to characterize the proportion of background harmonic residuals relative to the effective incremental response, reflecting the degree to which the response at this frequency point is contaminated by background noise; Used to characterize the consistency of the response at the same frequency point within adjacent sampling periods, reflecting whether there is a short-term drift or jump at the current frequency point; Used to characterize the prominence of the effective incremental response relative to the noise baseline, reflecting whether the frequency point has sufficiently clear identification value; Used to comprehensively reflect the first The reliability of the incremental response at each frequency point in impedance identification is an important gating quantity before entering the online impedance identification stage; It is used to measure the degree of bending or abrupt jump in the impedance spectrum at local frequency points, and reflects the local continuity of the impedance spectrum; It is used to measure the overall stability of the impedance spectrum in the current period relative to the historical impedance spectrum, and is an overall reference quantity for whether the compensation decision is suitable for rapid updating; It is used to reflect the dominant bias between current-side compensation demand and voltage-side compensation demand under current operating conditions; and These are used to indicate the priority of parallel compensation and series compensation in the current cycle, respectively. This is used to limit the rate of change of the compensation priority in adjacent periods, thereby suppressing compensation switching oscillations; and This is the actual control quantity that is ultimately sent to the collaborative compensation execution terminal.

[0115] Furthermore, this embodiment adopts the aforementioned model system instead of directly issuing compensation commands based solely on harmonic distortion indices, voltage fluctuation indices, or a single impedance value because grid-connected power quality issues are often not determined by a single factor, but are jointly influenced by background harmonic states, frequency band impedance distribution, disturbance dominance type, and the compensation state of the previous cycle. Without constraints on the reliability of the micro-disturbance response, contaminated frequency points are easily included in the identification process; without consideration of the local bending degree and overall stability of the impedance spectrum, the impedance spectrum is prone to instability due to measurement jumps; without quantification of the dominant bias of composite disturbances, the allocation between parallel and series compensation easily relies solely on empirical settings; without constraints on the compensation switching damping coefficient, compensation overshoot is likely to occur under rapid disturbance switching conditions. Therefore, this embodiment, through the aforementioned multi-layer nonlinear chain processing, establishes a more stable and interpretable collaborative mechanism between the frequency domain separation and background harmonic cancellation terminal, the grid impedance online identification terminal, and the compensation command generation terminal.

[0116] In summary, based on the system structure and workflow described in Example 1, this embodiment further provides a specific implementation method consisting of reliable extraction of incremental perturbation response, stable identification of impedance spectrum, determination of dominant bias of composite perturbation, and smooth generation of compensation priority quantity. This enables the system to stably complete online identification of grid connection point impedance and adaptive power quality compensation decision-making even in application scenarios where background harmonics exist, frequency band impedance dynamically changes, and composite perturbations coexist. This improves the practical engineering applicability of the entire compensation system.

[0117] Furthermore, in order to ensure that each computational model in this embodiment has a clear data source, a clear physical orientation, and reproducible processing logic, the definition, acquisition method, physical meaning, and control function of the main parameters involved in Embodiment 2 will be further explained.

[0118] in, Indicates the first Each preset frequency point corresponds to an angular frequency that is one-to-one with the frequency configuration of the multi-frequency small-amplitude perturbation signal executed by the perturbation injection terminal. This setting is used to separate the response characteristics of the grid connection point at different frequency bands, enabling the frequency domain separation and background harmonic cancellation terminal to extract the voltage and current responses around each preset frequency point. The reason for setting this parameter is that the grid connection point impedance does not exhibit uniform characteristics across all frequencies; different frequency bands often correspond to different types of harmonic coupling, reactive power fluctuation sensitivity, and voltage disturbance propagation characteristics. Without this parameter, the response characteristics can be significantly altered. This frequency index makes it impossible to establish an impedance spectrum oriented towards frequency distribution.

[0119] in, and They represent the first The voltage frequency domain response at a preset frequency point before and after perturbation injection. and They represent the first The current frequency domain response at preset frequency points before and after perturbation injection. All of these parameters are complex frequency domain quantities; their amplitude is used to characterize the response intensity, and their phase is used to characterize the response phase relationship. These parameters are obtained by synchronous sampling via a grid-connected voltage and current acquisition terminal, followed by frequency domain separation and background harmonic cancellation to perform frequency domain transformation and extract corresponding frequency points from the sampled sequence. The reason for setting these four parameters is that the corresponding frequency domain response quantities before and after perturbation injection form the basis for incremental extraction. Without these two sets of frequency domain quantities, it would be impossible to distinguish between the background state and the current perturbation excitation state.

[0120] in, Indicates the first The voltage increment response caused by this perturbation injection at a preset frequency point Indicates the first The current increment response caused by this perturbation injection at a preset frequency point. and Both are obtained by subtracting the corresponding frequency domain responses before and after the perturbation injection, and are used to characterize the additional response caused by the perturbation at the grid connection point. Their physical meaning is as follows: It reflects the additional response strength of the voltage side at the grid connection point to the frequency disturbance. The additional response strength of the grid connection point current side to the frequency disturbance is reflected, and the two together determine the identification result of the equivalent impedance at the current frequency. and The reason is that if impedance calculations are performed directly on the original voltage and current frequency domain quantities, background harmonics and original load fluctuations will be included in the results, while the incremental form can more directly approximate the frequency response triggered by the current injection action.

[0121] in, Indicates the first The residual amplitude of background harmonics extracted within a preset frequency band near a preset frequency point is derived from the amplitude statistics of residual non-target components in the frequency band adjacent to the current frequency point after the frequency domain separation and background harmonic cancellation terminal has completed two differential operations. The physical meaning of this parameter is that it characterizes the background spectral energy level near the current frequency point that has not yet been effectively canceled out. The reason is that although most stable background components can be weakened after differential processing before and after perturbation injection, slow-drifting background harmonics, short-time pulsations, or noise leakage may still exist under actual operating conditions. These residual amounts will affect the purity of the incremental response at the current frequency point, therefore, it is necessary to... The effect was quantitatively characterized.

[0122] in, Indicates the first The background harmonic residual factor at a preset frequency point is calculated by the relative ratio of the background harmonic residual amplitude to the effective incremental response. The physical meaning of this parameter is that it reflects the dominance of the "residual background component" relative to the "target perturbation response component" at the current frequency. When When the value is small, it indicates that the current frequency response is mainly caused by this perturbation, and the background pollution is relatively mild; when A larger value indicates that there is still strong background interference in the current frequency response, making it unsuitable for direct impedance identification. The reason for setting this parameter is that even... and However, the fact that none of the values ​​are zero does not automatically mean that the current frequency point is reliable, because the increment may contain residual background data. Therefore, it is necessary to set... As a precursor to credibility decay.

[0123] in, and They represent the first Within the sampling period, the first Voltage increment response and current increment response at a preset frequency point This represents the total number of sampling periods used for continuous comparison. This parameter is primarily used to construct the relationship between adjacent sampling periods. Its physical meaning is that... and Reflecting the continuous response trajectory of the same frequency point in the time dimension, This determines the length of the observation window used to evaluate the stability of the continuous response trajectory. The reason for setting this set of parameters is that the frequency response within a single sampling period may be affected by transient disturbances and exhibit occasional anomalies. Only by comparing the responses of the same frequency point across multiple sampling periods can it be determined whether the frequency response is continuously stable or occurs occasionally.

[0124] in, Indicates the first The adjacent sampling period discrete offset factor of a preset frequency point is used to characterize the discreteness of the incremental response of the same frequency point in multiple consecutive sampling periods. The logical meaning is that this parameter is used to quantify the consistency of the current frequency response over time; the larger the value, the more significant the response shift of the current frequency in different periods, and the worse the stability; the smaller the value, the better the consistency of the current frequency in continuous periods. The reason is that the frequency response on which impedance identification depends must not only be strong, but also stable. If a certain frequency point shows a large response in the current cycle, but fluctuates greatly in the continuous cycle, then the frequency point is more likely to be the result of transient noise triggering, and should not be given high weight in the impedance spectrum construction.

[0125] in, This represents the current discrete offset weighting coefficient, which is used to balance the voltage increment response jump and the current increment response jump. The relative contribution in the calculation. The logic behind this setup is that, since the voltage and current sides of the grid connection point are not entirely sensitive to disturbances under different operating conditions, assuming that their effects are equal might mask abnormal fluctuations on one side. Therefore, by... The appropriate conversion factor can be applied to discrete changes on the current side according to the system application scenario.

[0126] in, Indicates the first The noise baseline amplitude corresponding to each preset frequency point is derived from the average noise amplitude extracted from non-target frequency bands near the current frequency point, or from a period without injection reference. The physical meaning of this parameter is that it characterizes the fundamental noise level in the environment in which the current frequency response is located. The reason is that the absolute value of the incremental response is not directly equivalent to "discriminability." For example, in high-noise frequency bands, even... or A relatively large value may still be insufficient to demonstrate a clear and reliable response at that frequency point; therefore, further analysis is needed. Perform normalized evaluation.

[0127] in, Indicates the first The incremental response significance factor at a preset frequency point is used to characterize the prominence of the effective incremental response relative to the noise baseline. The physical meaning of this parameter is that it measures whether "the target response is sufficiently higher than the background noise." When A higher value indicates that the current frequency response is significantly more prominent than the background noise, possessing high identification value; when A lower value indicates that the current frequency response may be submerged in noise. (Settings) The reason is that Example 2 requires not only frequency points with low background interference and high periodicity, but also sufficiently significant target perturbation response itself. This constitutes an item that enhances credibility.

[0128] in, , , These represent the background residual suppression coefficient, the discrete offset suppression coefficient, and the significance amplification coefficient, respectively. These three parameters are all adjustment parameters in the credibility model, and their logical meaning is as follows: The degree to which background contamination affects credibility decay is determined. The degree to which periodic discrete fluctuations are sensitive to credibility decay is determined. The degree to which incremental significance amplifies the improvement in credibility is determined. The reason for setting the above parameters is that the relative influences of background harmonic intensity, sampling stability, and target response intensity are not entirely consistent under different grid connection scenarios. Therefore, adjustable parameters are needed to enable the credibility model to adapt to different application scenarios.

[0129] in, Indicates the first The reliability index of the perturbation response at a preset frequency point is composed of the background harmonic residual factor. Discrete offset factor between adjacent sampling periods and incremental response significance factor Determined jointly. The aforementioned The logical implication is that this parameter is a comprehensive gating quantity for determining whether the current frequency is suitable as an impedance identification input. A higher value indicates that the current frequency simultaneously meets the three conditions of "lower background pollution," "more stable periodic response," and "stronger target response"; a lower value indicates that the current frequency identification basis is unreliable. The reason is that this embodiment does not simply include all frequency points equally in the impedance calculation, but first identifies the quality of the frequency points and then identifies the impedance, thereby improving the quality of the impedance spectrum.

[0130] in, Indicates the first The historical equivalent impedance parameters retained at each preset frequency point in the previous update cycle. This represents the first period obtained under the credibility constraint. The equivalent impedance parameters at a preset frequency point. The physical meaning of this parameter is that it reflects the reference impedance level at the current frequency point under historical stable conditions; This is the current period impedance estimate after integrating current measurement results and historical reference results. The reason for setting these two parameters is that impedance identification relying entirely on current period observations would be too sensitive to single-period anomalies, while... By integrating the current impedance estimation results, the impedance update can be made both real-time and continuous.

[0131] in, This represents the local curvature of the impedance spectrum, which characterizes the level of local curvature at the current frequency point in the impedance spectrum. The physical meaning of this parameter is that it reflects the first... The degree of deviation of the impedance value at each frequency point relative to two adjacent frequency points; if A larger value indicates a significant jump in the spectrum at the current frequency point. The reason is that although the true impedance spectrum may change nonlinearly, if there are excessively sharp bends at local points, it is more likely caused by measurement noise, transient disturbances, or frequency identification errors. Therefore, it is necessary to... Identify this type of local anomaly.

[0132] in, This represents the bending smoothing adjustment coefficient. Indicates the bending sensitivity index. This represents the smoothed impedance parameter. Used to control the overall impact of localized bending on smoothness strength. This is used to control the nonlinear amplification trend of this effect as the bending degree increases, and This represents the actual impedance spectrum input value after local smoothing. The reason for setting this set of parameters is that Example 2 does not uniformly average all frequency points, but rather preserves realistic details as much as possible when local bends are small, and increases the smoothing intensity when local bends are significant. Therefore, through... and The combination of these elements can achieve a layered effect of "light bending with weak treatment and strong bending with strong treatment".

[0133] in, This represents the total number of frequency points involved in impedance spectrum construction. This represents the historical offset penalty coefficient. This represents the penalty coefficient for local bending. This indicates the stability index of the impedance spectrum. The physical or logical meaning of this parameter is that it is used to measure, from a holistic perspective, the stability and continuity of the current impedance spectrum relative to historical impedance spectra. If A higher value indicates that the current impedance spectrum shows neither significant overall drift nor excessive local jumps; if... A lower value indicates a weaker overall reliability of the current impedance spectrum. (Settings) The reason is that the compensation decision does not only depend on the quality of a single frequency point, but also on the overall reliability of the entire impedance spectrum. Therefore, a comprehensive index is needed to constrain the update range of subsequent compensation priority quantities.

[0134] in, Indicates harmonic distortion index, This indicates a voltage fluctuation index. Indicates a negative-order imbalance index. This represents the reactive power fluctuation index. The above parameters are all obtained through further analysis of real-time sampling data collected by the grid connection point voltage and current acquisition terminal, and correspond to different disturbance dimensions in the power quality state. Their physical meaning is as follows: Reflects the degree of harmonic distortion. It reflects the degree to which the voltage amplitude fluctuates over time. It reflects the prominence of the negative sequence component in a three-phase voltage or current imbalance state. This reflects the degree of reactive power fluctuation over time. The reason for setting the above parameters is that the compensation decision in this embodiment is not only based on the impedance state, but also on the external manifestation of the current disturbance. Therefore, it is necessary to establish characterization quantities for the current-side disturbance and the voltage-side disturbance respectively.

[0135] in, This represents the set of frequency bands associated with parallel compensation. This represents the set of frequency bands associated with series compensation. Both sets are logical divisions of the frequency index range in the impedance spectrum, where... Used to characterize the frequency band range more relevant to parallel compensation This is used to characterize the frequency range more relevant to series compensation. The reason for setting this set of parameters is that parallel compensation and series compensation are usually more sensitive to different types of disturbances and impedance changes at different frequency bands, respectively. Without the division of frequency bands, it is difficult to establish a clear relationship between impedance spectrum characteristics and compensation path selection.

[0136] in, , , These represent the bias amplification factor, reactive power conversion factor, and negative order conversion factor, respectively. This indicates the dominant bias of the composite disturbance. The logical implication is that this parameter is used to comprehensively reflect whether the current operating condition is more suitable for parallel compensation or series compensation. When At higher values, it indicates a stronger relative impact from harmonic distortion and reactive power fluctuations, and a more pronounced response of the impedance spectrum in the parallel compensation correlated frequency band; when At lower values, it indicates that the effects of voltage fluctuations and negative sequence imbalances are stronger, and the impedance spectrum response is more prominent in the series compensation correlated frequency band. The reason is that this embodiment does not rely solely on experience to determine the compensation direction based on a single power quality indicator, but rather couples the disturbance appearance with impedance sensitivity to form a direction discrimination quantity.

[0137] in, to This is the priority adjustment coefficient. Indicates the priority of parallel compensation. This indicates the priority of series compensation. and The logical implication is that these two parameters are used to measure the priority of compensation intensity allocated to parallel and series compensation channels in the current cycle, respectively. The larger the value, the higher the priority the corresponding compensation path should be. The reason for setting this set of parameters is that compensation decisions need to move from "determining which type of disturbance is more dominant" to "assigning a certain priority to the corresponding compensation path," thus requiring... and Transform the direction decision variable into an executable priority variable.

[0138] in, and These represent the priority quantities for parallel compensation and series compensation in the previous cycle, respectively. , , These represent the parallel change penalty coefficient, the series change penalty coefficient, and the stability penalty coefficient, respectively. This represents the compensation switching vibration suppression coefficient. The logical implication is that this parameter controls the update speed of the current cycle's compensation priority relative to the previous cycle's priority. A smaller value indicates a stronger suppression of the update action in the current cycle to avoid abrupt changes in the compensation amount. The reason for setting this parameter is that even if grid connection point disturbances and impedance states change, it may not be suitable to adjust the compensation strategy instantaneously and fully, especially when impedance spectrum stability is insufficient or priority changes are large. Therefore, it is necessary to use [a more nuanced approach]. Buffer the switching process.

[0139] in, This indicates the amount of the modified parallel compensation command. This represents the corrected series compensation command quantity. The two parameters are the final motion control quantities output by the entire calculated chain in Embodiment 2, which are directly fed into the parallel compensation section and the series compensation section of the collaborative compensation execution terminal, respectively. Their logical meaning is that... and The current disturbance state, current impedance spectrum state, historical compensation state, and switching smoothing requirements have all been considered simultaneously, making it a more suitable actual execution quantity for engineering applications. The reason for setting these two parameters is that all calculation models in Example 2 must ultimately be grounded in specific actions. If only intermediate indicators are calculated without forming execution quantities, the closed loop from identification to compensation cannot be completed.

[0140] Furthermore, the aforementioned parameters do not exist in isolation, but rather function sequentially along a chain relationship: "frequency response extraction, frequency quality determination, impedance parameter construction, impedance spectrum stability evaluation, disturbance dominance bias determination, compensation priority generation, compensation switching smoothing correction, and compensation execution output." , , , , , , Parameters that constitute the response extraction layer; , , , , , , , , Parameters constituting the response quality evaluation layer; , , , , , Parameters constituting the impedance spectrum construction and stability evaluation layer; , , , , , , , , The parameters that constitute the compensation dominant bias and priority quantity generation layer; , , , , These parameters constitute the compensation switching and execution output layer parameters.

[0141] Example 3: Based on Examples 1 and 2, this example further explains the electronic device implementation, program execution implementation, and computer-readable storage medium implementation of the above-mentioned grid impedance adaptive power quality compensation system based on perturbation injection, so that the aforementioned system structure, processing flow, and parameterized calculation process have clearer hardware support relationships and program implementation relationships.

[0142] Specifically, this embodiment provides an electronic device for implementing the above-mentioned grid impedance adaptive power quality compensation system based on perturbation injection. The electronic device can be any of a central control device, an edge control device, or a distributed collaborative control device. The electronic device includes a processor, a memory, a communication interface, a data acquisition interface, an injection control interface, and a compensation drive interface. The processor is connected to the memory, the communication interface, the data acquisition interface, the injection control interface, and the compensation drive interface, respectively. The memory stores program instructions and running data. The processor is used to call the program instructions stored in the memory to execute the processing steps described in Embodiment 1, including grid connection point voltage and current acquisition, perturbation injection, frequency domain separation and background harmonic cancellation, online grid impedance identification, compensation instruction generation, and collaborative compensation execution. Furthermore, it executes the perturbation response reliability calculation, impedance spectrum stability calculation, composite perturbation dominance bias determination, compensation priority generation, and compensation switching vibration suppression correction processes described in Embodiment 2.

[0143] The communication interface is used to establish communication connections with the grid-connected point voltage and current acquisition terminal, the perturbation injection terminal, and the collaborative compensation execution terminal to receive real-time sampling data from the grid-connected point, send perturbation injection control information, and output compensation execution commands. The necessity of setting up this communication interface lies in the fact that, in this embodiment, each terminal can be set on the sampling side, injection side, compensation side, and central processing side according to actual deployment requirements. Without a communication interface, the status information, control information, and execution information between the terminals cannot form a closed-loop transmission, which is detrimental to achieving collaborative control of online identification and adaptive compensation.

[0144] The acquisition interface is used to receive three-phase voltage signal sampling data and three-phase current signal sampling data output by the grid-connected point voltage and current acquisition terminal. After calling program instructions, the processor performs time alignment, buffer storage, and preprocessing operations on the received three-phase voltage signal sampling data and three-phase current signal sampling data to form real-time sampling data corresponding to the same sampling moment. The necessity of setting up the acquisition interface is that the frequency domain separation processing, incremental response extraction, and impedance identification described in Embodiment 1 are all based on the time-synchronized voltage and current data. If the sampling data cannot enter the processor through a unified interface, the subsequent processing chain will be difficult to execute stably.

[0145] The injection control interface is used to output control parameters corresponding to a preset frequency, preset amplitude, preset injection period, and preset injection window to the perturbation injection terminal, enabling the perturbation injection terminal to inject multi-frequency small-amplitude perturbation signals when the preset injection conditions are met. After calling program instructions, the processor can read pre-stored frequency configuration data, amplitude limit data, and injection timing data from the memory, and send corresponding control commands to the perturbation injection terminal through the injection control interface. The necessity of setting up the injection control interface lies in the fact that this application does not passively rely on natural disturbances at the grid connection point for impedance estimation, but rather obtains frequency response information through active perturbation injection; therefore, a controllable output channel for the injection terminal is required.

[0146] The compensation drive interface is used to output parallel compensation instructions and series compensation instructions to the collaborative compensation execution terminal to drive the terminal to perform parallel compensation and / or series compensation. After calling the program instructions, the processor generates the parallel compensation priority quantity, series compensation priority quantity, parallel compensation instruction quantity, and series compensation instruction quantity output by the terminal according to the compensation instructions, forming corresponding compensation drive data, and sends it to the collaborative compensation execution terminal through the compensation drive interface. The necessity of setting up the compensation drive interface lies in the fact that the entire identification and decision-making process in this embodiment ultimately needs to be implemented in the compensation execution action. Without the drive connection with the compensation execution device, a complete closed loop from sampling, identification to execution cannot be formed.

[0147] Furthermore, the memory can be divided into a sampling data storage area, a frequency configuration storage area, a historical impedance spectrum storage area, a compensation priority storage area, and a program instruction storage area. The sampling data storage area is used to cache the three-phase voltage signal and three-phase current signal data of the grid connection point in each sampling period, as well as the corresponding frequency domain transformation results; the frequency configuration storage area is used to store preset frequency points, preset angular frequencies, preset amplitudes, preset injection periods, and injection window information; the historical impedance spectrum storage area is used to store historical equivalent impedance parameters, smoothed impedance parameters, and impedance spectrum stability indices in the previous update period or multiple update periods; the compensation priority storage area is used to store parallel compensation priority, series compensation priority, compensation switching damping coefficient, and compensation instruction quantity in the previous compensation period or multiple compensation periods; the program instruction storage area is used to store program instructions for implementing the processing procedures of Embodiment 1 and Embodiment 2. The necessity of adopting the above-mentioned partitioned storage method is that the frequency configuration data, real-time sampling data, historical impedance data and historical compensation data involved in this application are different in terms of their target and update time. If they are not stored in categories, the reading relationship will be easily confused when the processor calls them, which is not conducive to maintaining the continuity of the impedance identification and compensation update process.

[0148] Specifically, after invoking the program instructions, the processor can perform the following processing procedure:

[0149] First, the three-phase voltage signal sampling data and three-phase current signal sampling data output by the grid connection point voltage and current acquisition terminal are read. Time alignment processing is performed on the read data, and real-time sampling data corresponding to the same sampling time is generated. Then, based on the real-time sampling data, it is determined whether the current grid connection point meets the preset injection conditions. If the preset injection conditions are met, the perturbation injection terminal is controlled through the injection control interface to perform multi-frequency small-amplitude perturbation signal injection within the preset injection window. Before and after perturbation injection, frequency domain transformation and corresponding frequency point extraction are performed on the grid connection point response data to obtain the frequency domain response before perturbation injection and the frequency domain response after perturbation injection at each preset frequency point. Based on the above frequency domain response, the voltage increment response and current increment response are calculated, and the background harmonic residual factor, adjacent sampling period discrete offset factor, and increment response significance factor at each preset frequency point are further calculated to obtain the perturbation response reliability index at each preset frequency point. Based on the perturbation response reliability index and the current... Using frequency response data from the current period and historical periods, the equivalent impedance parameters, local bend of the impedance spectrum, and smoothed impedance parameters under reliable constraints are calculated. The grid impedance spectrum at the grid connection point is constructed according to frequency order, and an impedance spectrum stability index is generated. Further, combining the grid impedance spectrum, impedance spectrum stability index, and harmonic distortion index, voltage fluctuation index, negative sequence imbalance index, and reactive power fluctuation index obtained from the real-time sampling data analysis, a composite disturbance dominance bias index is calculated. Based on this composite disturbance dominance bias index, parallel compensation priority quantities and series compensation priority quantities are generated. Then, combining the priority quantity data corresponding to the previous compensation period, the compensation switching vibration suppression coefficient is calculated, and the corrected parallel compensation command quantities and series compensation command quantities are output. Finally, the parallel compensation command quantities and series compensation command quantities are sent to the collaborative compensation execution terminal through the compensation drive interface to drive the collaborative compensation execution terminal to perform parallel compensation, series compensation, or a collaborative compensation of parallel and series compensation.

[0150] Furthermore, in some embodiments, the processor in the electronic device can be a single processor or a processing platform composed of multiple processors working together; the memory can be a single physical memory or a hierarchical storage structure composed of multiple memories working together; the grid connection point voltage and current acquisition terminal, perturbation injection terminal, frequency domain separation and background harmonic cancellation terminal, grid impedance online identification terminal, compensation instruction generation terminal, and collaborative compensation execution terminal can be deployed on different hardware nodes, or can be implemented within the same electronic device as software functional modules or hardware logic modules. The necessity of setting up the above-mentioned distributed or centralized deployment methods lies in the fact that different application scenarios have different requirements for real-time performance, installation conditions, and communication structures. In some scenarios, sampling and injection devices are more suitable for being set up close to the grid connection point, while impedance identification and compensation decisions are more suitable for being completed uniformly by a central control device; in other scenarios, all processing can be completed centrally by an integrated control device. Therefore, it is necessary to clarify through Embodiment 3 that the solution of this application has multiple deployment adaptability.

[0151] Furthermore, in a more specific implementation, the processor can call the program instructions again after each compensation cycle to read the voltage signal sampling data and current signal sampling data of the grid connection point after compensation, recalculate the updated real-time power quality indicators and the updated grid impedance spectrum, and compare the recalculated impedance spectrum stability index, parallel compensation priority, series compensation priority, and compensation switching damping coefficient with the results of the previous cycle. When the comparison results indicate that the current impedance state or disturbance state has changed, the processor updates the historical data stored in the historical impedance spectrum storage area and the compensation priority storage area, and outputs new compensation instruction quantities. The necessity of setting up this continuous update process lies in the fact that the grid state of the grid connection point targeted by this application is not fixed, but dynamically changes with external power supply conditions, load conditions, and the compensation process. Therefore, it is necessary to realize online impedance identification and online update of compensation strategies through program loop calls.

[0152] Furthermore, this embodiment also provides a computer program product, which includes program instructions. When executed by a processor, the program instructions are used to perform the following operations: receiving three-phase voltage signal sampling data and three-phase current signal sampling data at the grid connection point; controlling a perturbation injection terminal to inject multi-frequency small-amplitude perturbation signals according to a preset frequency, preset amplitude, and preset injection period; performing differential processing on the frequency domain response before and after perturbation injection to obtain voltage increment response and current increment response; calculating the perturbation response reliability index, equivalent impedance parameter under reliability constraints, impedance spectrum local bending degree, and smoothed impedance parameter based on the voltage increment response and the current increment response; constructing the grid impedance spectrum at the grid connection point and generating an impedance spectrum stability index; combining the grid impedance spectrum, the impedance spectrum stability index, and real-time power quality index to generate parallel compensation priority quantity, series compensation priority quantity, and compensation switching vibration suppression coefficient; and outputting the corrected parallel compensation command quantity and series compensation command quantity to drive the collaborative compensation execution terminal to perform corresponding compensation actions. The necessity of setting up the aforementioned computer program product lies in the fact that the core processing flow in this application solution can essentially be solidified in the control platform through program instructions and executed by the processor during runtime, thereby giving this application solution a clear software implementation form.

[0153] Furthermore, this embodiment also provides a computer-readable storage medium storing program instructions. When the program instructions are executed by a processor, the processor performs the system workflow described in Embodiment 1 and the parameterized calculation process described in Embodiment 2. Specifically, this includes: acquiring and time-aligning the grid connection point voltage and current signals; controlling the perturbation injection terminal to perform multi-frequency small-amplitude perturbation signal injection; performing differential processing on the frequency domain response before and after injection; calculating the perturbation response reliability index at each preset frequency point; constructing equivalent impedance parameters and smoothed impedance parameters under reliable constraints, and generating a grid impedance spectrum and impedance spectrum stability index; generating a composite perturbation dominant bias index, parallel compensation priority quantity, series compensation priority quantity, and compensation switching vibration suppression coefficient based on real-time power quality index and grid impedance spectrum; and outputting the parallel compensation command quantity and series compensation command quantity to the collaborative compensation execution terminal.

[0154] Furthermore, in some embodiments, the memory may also store log data for recording the system's operating status. This log data includes at least the perturbation injection time, injection frequency number, injection amplitude, injection cycle number, perturbation response reliability index, impedance spectrum stability index, composite perturbation dominance bias index, parallel compensation priority, series compensation priority, and compensation command output records. The necessity of setting up log data storage lies in the fact that the system needs to track the identification and compensation results of each cycle during long-term operation, so as to conduct retrospective analysis on abnormal operating conditions, impedance change conditions, or compensation switching conditions. It also helps in subsequent optimization of frequency configuration parameters, vibration suppression parameters, and priority adjustment parameters.

[0155] Furthermore, in some other embodiments, the electronic device may also include a display interface or a human-machine interface, which is used to output the real-time power quality status of the current grid connection point, the current grid impedance spectrum, the current compensation mode, the current parallel compensation priority and series compensation priority, and to receive adjustment commands for preset frequency points, injection amplitudes, injection windows, compensation thresholds, and vibration suppression parameters. The necessity of setting up the display interface or human-machine interface lies in the fact that, in actual engineering applications, operators may need to adjust some configuration parameters according to the on-site operating status, or view the current impedance spectrum and compensation mode, thereby improving the operability of the system in engineering deployment.

[0156] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A power quality compensation system based on perturbation injection and adaptive power grid impedance, characterized in that, This includes a grid connection point voltage and current acquisition terminal, a perturbation injection terminal, a frequency domain separation and background harmonic cancellation terminal, a grid impedance online identification terminal, a compensation command generation terminal, and a collaborative compensation execution terminal; The grid connection point voltage and current acquisition terminal is used to acquire the voltage and current signals of the grid connection point and output real-time sampling data; The perturbation injection terminal is connected to the grid connection point voltage and current acquisition terminal and is used to inject multi-frequency small-amplitude perturbation signals into the grid connection point according to a preset frequency, preset amplitude and preset injection period, so that the grid connection point can form voltage and current response data containing perturbation response information. The frequency domain separation and background harmonic cancellation terminal is connected to the grid connection point voltage and current acquisition terminal and the micro-perturbation injection terminal, respectively. It is used to perform frequency domain separation processing on the real-time sampling data and the voltage and current response data, and to cancel the background harmonic influence based on the frequency domain difference of at least two sampling results, so as to obtain the frequency band response data corresponding to the micro-perturbation signal. The online grid impedance identification terminal is connected to the frequency domain separation and background harmonic cancellation terminal, and is used to calculate the equivalent impedance parameters of the grid connection point in each frequency band based on the frequency band response data, and to construct the grid impedance spectrum. The compensation instruction generation terminal is connected to the grid connection point voltage and current acquisition terminal and the grid impedance online identification terminal, respectively. It is used to generate compensation instructions based on the grid impedance spectrum and the real-time power quality indicators obtained from the real-time sampling data. The real-time power quality indicators include at least one or more of the following: harmonic distortion indicators, voltage fluctuation indicators, negative sequence imbalance indicators, and reactive power fluctuation indicators. The collaborative compensation execution terminal is connected to the compensation instruction generation terminal and is used to perform parallel compensation and / or series compensation according to the compensation instruction to adaptively compensate for harmonic current, reactive power fluctuation, voltage distortion and three-phase imbalance at the grid connection point.

2. The adaptive power quality compensation system for grid impedance based on perturbation injection according to claim 1, characterized in that, The grid connection point voltage and current acquisition terminal includes a voltage sampling module, a current sampling module, and a timing synchronization module; The voltage sampling module is used to collect the three-phase voltage signal of the grid connection point, the current sampling module is used to collect the three-phase current signal of the grid connection point, and the timing synchronization module is used to perform time alignment processing on the collected three-phase voltage signal and three-phase current signal to obtain real-time sampling data corresponding to the same sampling time.

3. The adaptive power quality compensation system for grid impedance based on perturbation injection according to claim 1, characterized in that, The perturbation injection terminal is used to inject small-amplitude sinusoidal perturbation signals into the grid connection point sequentially or simultaneously according to multiple preset frequency points within a preset injection window, and to limit the injection amplitude corresponding to each preset frequency point so that the injection process of the perturbation signal does not change the normal power supply state of the grid connection point.

4. The adaptive power quality compensation system for grid impedance based on perturbation injection according to claim 1, characterized in that, The frequency domain separation and background harmonic cancellation terminal is used to acquire the first frequency domain sampling result before perturbation injection and the second frequency domain sampling result after perturbation injection, and to perform differential operation on the first frequency domain sampling result and the second frequency domain sampling result to separate the incremental response component caused by the perturbation signal from the mixed spectrum, thereby obtaining the frequency segment response data corresponding to each preset frequency point.

5. The adaptive power quality compensation system for grid impedance based on perturbation injection according to claim 1, characterized in that, The online grid impedance identification terminal is used to calculate the equivalent impedance parameters corresponding to each preset frequency point based on the voltage increment response and current increment response at each preset frequency point, and arrange the equivalent impedance parameters in frequency order to construct a grid impedance spectrum that reflects the frequency response characteristics of the grid connection point.

6. The adaptive power quality compensation system for grid impedance based on perturbation injection according to claim 1, characterized in that, The compensation instruction generation terminal is used to determine the priority of parallel compensation and the priority of series compensation based on the impedance characteristics of different frequency bands in the power grid impedance spectrum and the real-time power quality index, and to generate corresponding compensation instructions. The collaborative compensation execution terminal is used to perform parallel compensation when harmonic current and reactive power fluctuations are dominant, to perform series compensation when voltage distortion and three-phase imbalance are dominant, or to perform collaborative compensation of parallel and series compensation under combined disturbance conditions.

Images