A power distribution network harmonic suppression method and system based on a remaining capacity of a photovoltaic inverter

Abstract

The application discloses a power distribution network harmonic suppression method and system based on a residual capacity of a photovoltaic inverter. The method obtains line impedance and node load impedance of a power distribution network, calculates a compensation coefficient for correcting a remote harmonic compensation component, the coefficient establishes a pre-distortion model based on a circuit shunt principle to offset current attenuation caused by the line impedance, accurately calculates the residual capacity according to a rated capacity, real-time active and reactive power of the photovoltaic inverter, and generates a final current instruction containing photovoltaic grid-connected current, local and remote harmonic compensation components by using the compensation coefficient in combination with the residual capacity. The system preferentially compensates local nodes by using the residual capacity, and distributes the surplus capacity to remote nodes in order from small to large impedance, so that the whole-area energy transmission loss minimization is realized. The application breaks through the physical limitation of line impedance on the remote compensation accuracy, realizes double optimization of asset utilization and system safety, and significantly reduces the overall operation loss of the power distribution network.

Description

Technical Field

[0001] This invention relates to the field of power quality improvement technology for distribution networks, and in particular to a method and system for harmonic suppression in distribution networks based on the remaining capacity of photovoltaic inverters. Background Technology

[0002] With the transformation of the global energy structure, distributed generation (DG), represented by photovoltaic power generation, is increasingly penetrating low-voltage distribution networks. Simultaneously, the load characteristics of distribution networks have also undergone profound changes. The large-scale integration of nonlinear loads such as electric vehicle charging stations, variable frequency air conditioners, and switching power supplies has led to increasingly severe harmonic pollution problems in distribution networks. The injection of harmonic currents not only increases line losses and causes transformer overheating, but in severe cases, it can even induce system resonance, threatening the safe and stable operation of the power grid.

[0003] Traditional power quality management architectures primarily rely on the centralized installation of active power filters (APFs) or passive filters at the main outlet of a distribution substation. While this centralized approach can improve power quality at the point of common coupling (PCC) to some extent, it has significant physical limitations: firstly, dedicated APF equipment is expensive and requires a large area, making it difficult to deploy widely at various end nodes of the distribution network, resulting in severe harmonics at the end-user side within the substation; secondly, centralized management cannot eliminate line losses and voltage distortion caused by harmonic currents during transmission on distribution network lines.

[0004] In light of this, academia and industry have begun exploring solutions that utilize grid-connected photovoltaic (PV) inverters with APF (Active Power Factor) functionality. PV inverters are highly similar to APFs in hardware topology, both employing bridge circuits composed of fully controlled power electronic switching devices (such as IGBTs). During periods of insufficient sunlight or at night, the inverter is idle or lightly loaded, theoretically capable of outputting reactive power or harmonic current. For example, Chinese patent document CN104113081B discloses a PV inverter circuit with reactive power compensation function, which controls the inverter to compensate the grid for reactive current by detecting the relationship between the PV side voltage and the grid connection point voltage. However, this type of solution mainly focuses on reactive power support, without in-depth discussion of harmonic mitigation, which has higher frequencies and more stringent control bandwidth requirements. Furthermore, its control logic is often based on simple voltage threshold judgments, lacking precise quantification of the inverter's real-time current carrying capacity, making it prone to overcurrent tripping during sudden changes in sunlight.

[0005] Further technological evolution has led to the emergence of "photovoltaic grid-connected inverters with adaptive harmonic local mitigation function," as described in Chinese patent application CN104218838A. This solution advocates a "local mitigation" strategy, whereby the inverter only detects and compensates for load harmonics near its installation point. While this strategy is effective in single-unit operation, its limitations become apparent in distribution substations with multiple units operating in parallel. Due to significant variations in load characteristics and lighting conditions at each node, resource mismatches may occur, such as "node A experiencing severe harmonics but the inverter being fully loaded and unable to compensate, while node B has a light load and the inverter is idle." This fragmented control model cannot achieve globally optimized allocation of compensation resources within the distribution substation.

[0006] To address resource misallocation, some research has shifted towards multi-machine cooperative control. However, in the actual physical environment of low-voltage distribution networks, cooperative control faces a more hidden and fatal mechanistic obstacle—the shunting effect of line impedance. Existing literature on cooperative control often simplifies the distribution network to a single-point connection model or assumes zero line impedance. But in real-world low-voltage distribution networks (such as 380V / 220V systems), the ratio of line resistance to reactance (R / X) is relatively large, and the line impedance... This cannot be ignored. When a remote node (such as node B) attempts to inject harmonic compensation current into node A, according to Kirchhoff's Current Law (KCL) and the current shunting principle of parallel circuits, the compensation current generated by the inverter will not all flow to the target node A, but will be affected by the line impedance. It is shunt to the equivalent impedance on the grid side.

[0007] Specifically, if compensation commands are issued directly without considering the influence of line impedance, the amplitude of the compensation current output by the inverter will attenuate and the phase will shift after long-distance transmission. This physical signal distortion will prevent the actual compensation current reaching the target node from forming an equal-amplitude, anti-phase cancellation relationship with the harmonic current generated by the load. Furthermore, at certain frequencies, the phase shift may even cause "compensation" to become "enhancement," thus worsening the harmonic distortion rate of the power grid. Existing technical solutions lack a mechanism that can incorporate line impedance into the control closed loop and perform precise mathematical corrections to remote compensation commands.

[0008] In summary, current power distribution network harmonic mitigation technologies face a dilemma: high cost of dedicated equipment versus low accuracy of reused equipment. A key technological bottleneck to overcome in this field is how to quantify the remaining capacity of each inverter in a complex multi-node, multi-impedance power distribution network in real time, and overcome the shunting and attenuation effects caused by line impedance to achieve high-precision collaborative compensation. Summary of the Invention

[0009] To address the aforementioned technical problems, this invention provides a method and system for suppressing harmonics in distribution networks based on the remaining capacity of photovoltaic inverters. It includes: Obtain the line impedance and node load impedance of the power distribution network; Calculate the compensation coefficients used to correct the long-range harmonic compensation components based on the line impedance and the node load impedance. ; Obtain the remaining capacity of the photovoltaic inverter; Based on the remaining capacity, the final current command is generated using the compensation coefficient.

[0010] Furthermore; the remaining capacity The calculation formula is: in, For rated capacity, For real-time active power, This represents real-time reactive power.

[0011] Furthermore; the final current command The formula for generating it is: ;in, For photovoltaic grid-connected current, For local harmonic compensation current, For the remote harmonic compensation current component allocated to the local inverter, This is a correction coefficient for the remote harmonic compensation component. In the control command generation stage, the method combines this remaining capacity and uses the compensation coefficient to generate the final current command. (Line impedance) For distribution network line impedance, the node load impedance includes the remote node load impedance. and local node load impedance The steps for obtaining the line impedance and node load impedance of the distribution network include: injecting a micro-disturbance current of a specific frequency into the power grid, collecting the response voltage and current, and identifying the line impedance and load impedance using FFT analysis; or reading line parameters from the distribution network management system and calculating the impedance matrix. The amplitude range of the disturbance signal is set to 1%-5% of the rated current to ensure that it does not affect normal operation; the frequency selection avoids the fundamental frequency and its lower harmonics, and multiple frequency points in the range of 200Hz-2kHz are selected for frequency sweeping; the FFT analysis uses the Fast Fourier Transform algorithm; the mathematical model for impedance identification is based on Ohm's law, and the impedance frequency characteristics are established through multi-frequency point measurements. Based on these impedance parameters, compensation coefficients for correcting remote harmonic compensation components are calculated according to the line impedance and node load impedance. The compensation coefficient The calculation formula is: ,in For line impedance, The equivalent impedance of the local harmonic source. This represents the local node load impedance, and ' / / ' indicates parallel operation. The equivalent impedance of the local harmonic source... The harmonics are obtained by measuring the harmonic current and harmonic voltage at the local node and applying Ohm's law. The calculation yielded, where The amplitude of the harmonic voltage. This is the harmonic current amplitude; or, through a load model identification algorithm, the equivalent impedance is calculated based on the nonlinear characteristic curve of the load. This compensation coefficient is used to compensate for the line impedance. The resulting current attenuation, physically represented by the reciprocal of the line impedance shunting ratio, serves as a "pre-distortion gain" to offset the current shunting effect during transmission. Furthermore, the method includes a capacity allocation strategy: prioritizing the use of the remaining capacity of the photovoltaic inverter to compensate for harmonics at local nodes; when there is still remaining capacity after local harmonic mitigation is satisfied, the remaining capacity is allocated to remote nodes. The criteria for satisfying local harmonic mitigation are: after local harmonic current compensation, the total harmonic distortion (THD) of the local node is lower than a preset threshold (e.g., 5%), or the local harmonic compensation current requirement is less than the maximum compensable current corresponding to the remaining capacity. When allocating the remaining capacity to remote nodes, a weighted allocation is performed based on the harmonic severity and electrical distance of each remote node.

[0012] Furthermore, the method also includes a remote allocation principle: when allocating to remote nodes, the remote allocation principle is followed, that is, priority is determined according to the electrical distance or impedance between nodes, and the remaining capacity is allocated in ascending order of impedance. The electrical distance is defined as the magnitude of the line impedance between two nodes. For nodes connected by multiple line segments, the electrical distance is the magnitude of the vector sum of the impedances of each line segment. The allocation in ascending order of impedance means: first, the remaining capacity is allocated to the remote node with the smallest electrical distance. If there is still remaining capacity after the harmonic mitigation requirements of that node are met, it is then allocated to the node with the next smallest electrical distance, and so on.

[0013] Furthermore, obtaining the remaining capacity of the photovoltaic inverter includes: real-time acquisition of the rated capacity of the photovoltaic inverter. Real-time active power and real-time reactive power The real-time active power and real-time reactive power are obtained by real-time acquisition of the voltage and current at the grid connection point through current transformers and voltage transformers, and calculated through coordinate transformation.

[0014] Furthermore; line impedance The impedance of the distribution network lines includes the node load impedance, which includes the remote node load impedance. and local node load impedance Simultaneously, the method obtains the remaining capacity of the photovoltaic inverter. The specific steps for obtaining the remaining capacity include real-time acquisition of the rated capacity of the photovoltaic inverter. Real-time active power and real-time reactive power .

[0015] Furthermore, it also includes parallel management of multiple harmonics: extracting each harmonic component through FFT analysis; independently calculating the line impedance and compensation coefficient at the corresponding frequency for each harmonic; allocating the total remaining capacity according to the weight of each harmonic content; and superimposing the corrected current commands of each harmonic in the time domain to generate the final control command.

[0016] Furthermore, it also includes a dynamic parameter update step: real-time monitoring of changes in load impedance and photovoltaic power generation, dynamic adjustment of the compensation coefficient based on impedance fluctuations, and dynamic updating of remaining capacity and compensation capacity based on capacity fluctuations.

[0017] This invention also discloses a distribution network harmonic suppression system, comprising: a compensation capacity calculation module configured to calculate the remaining capacity of a photovoltaic inverter; the compensation capacity calculation module is configured to execute a formula. To calculate the remaining capacity; The compensation capacity allocation module is configured to calculate the compensation coefficient based on the line impedance and load impedance, and to formulate a capacity allocation strategy. The logical configuration of the compensation capacity allocation module is as follows: priority is given to meeting local governance needs, and if there are any surpluses, they are allocated to remote nodes according to electrical distance priority. The current command tuning module is configured to generate a final control signal by combining the remaining capacity and the compensation coefficient.

[0018] Furthermore, the system also includes multiple distributed photovoltaic grid-connected inverters, and the compensation capacity calculation module, the compensation capacity allocation module, and the current command setting module are integrated into the controller of the distributed photovoltaic grid-connected inverters. This can be implemented as software code stored in memory and executed by a DSP or MCU, or as hardware logic circuitry in an FPGA. This system is applied to distribution network areas containing multiple photovoltaic inverter nodes, where line impedance exists between the nodes.

[0019] Furthermore, the system is applied to a distribution network area containing multiple photovoltaic inverter nodes, and the line impedance exists between each node.

[0020] Furthermore, the system is also configured to: detect the harmonic current components of the local load in real time, and receive harmonic current data and support requests from remote nodes through a communication network; the communication network may employ power line carrier, wireless communication, industrial Ethernet, or fiber optic communication, etc.

[0021] The present invention, through the above-described technical solution, has produced significant beneficial effects in the management of power quality in power distribution networks: 1. Overcame the physical limitations of line impedance on the accuracy of remote compensation: Unlike existing technologies that ignore line impedance, this invention introduces technologies that include line impedance. Compensation coefficient for load impedance parameters A precise circuit shunting compensation model was established. This coefficient is essentially a "pre-distortion" gain, capable of offsetting the shunting effect of current during transmission in low-voltage distribution network lines. After applying this correction coefficient, the current waveform distortion rate at the point of connection (PCC) and remote nodes is significantly reduced, and the waveform sinusoidality is greatly improved, solving the phase deviation problem in traditional collaborative control that is either a temporary fix or exacerbates the problem. It also addresses the issue that existing collaborative control strategies neglect the influence of distribution network line impedance on current transmission, leading to shunting and phase shift of the remote compensation current during transmission, resulting in inaccurate offsetting of harmonics at the target node and a significant decrease in mitigation effect with increasing electrical distance. 2. It achieves dual optimization of asset utilization and system security: Through real-time calculation This invention defines precise dynamic boundaries for harmonic mitigation. This not only allows photovoltaic inverters to function as virtual APFs during periods of low sunlight or at night, maximizing the use of expensive power electronic assets and saving the huge investment required for separate APF installations on the grid side; but also ensures that harmonic compensation is always in a "residual" state, never crowding out the active power channel of photovoltaic power generation, and preventing inverter damage due to overload, thus cleverly protecting both power generation revenue and equipment safety. It solves the problem of lacking precise geometric calculations of the inverter's real-time remaining capacity, which leads to either wasted compensation capacity due to excessive margins or overcurrent protection triggered by overshoot during periods of fluctuating sunlight, failing to maximize equipment potential while ensuring active power generation.

[0022] (3) Reduced the overall operating losses of the power distribution network: The proximity-priority, impedance-based power allocation strategy proposed in this invention conforms to the principle of minimum action in power transmission. This is because the local compensation current does not flow through the line impedance. This avoids the line Heat loss and voltage drop, therefore, prioritizing the solution of local harmonics can significantly improve system energy efficiency. Supporting neighboring nodes only when local capacity is sufficient, this orderly cooperation mode avoids the additional network losses caused by forced compensation over long distances, and achieves optimal energy efficiency at the distribution area level. (4) Improved the adaptability of the distribution network to complex operating conditions: This system can determine the load impedance. Dynamic adjustment of compensation coefficient This means that the system can update control parameters in real time, regardless of impedance fluctuations caused by load switching on the user side or capacity fluctuations caused by changes in illumination. This dynamic adaptive characteristic makes this invention particularly suitable for modern active distribution networks with highly random source-load characteristics. It solves the problem that existing distribution strategies are mostly based on fixed ratios or simple equal-sharing principles, failing to consider differences in electrical distance (impedance) between nodes, resulting in high overall system losses and an inability to cope with uneven spatiotemporal distribution of load and illumination. (See attached figures for details.) Figure 1 This is a schematic diagram of the overall topology and control principle of the distribution network harmonic suppression system based on the remaining capacity of photovoltaic inverters according to the present invention. Figure 2 This is a schematic diagram of the logic determination and execution flow of the cooperative harmonic suppression method in this invention; Figure 3 This is a schematic diagram of the current waveforms at each node of a dual-inverter dual-nonlinear load system without compensation measures. Figure 4 The output current waveform and harmonic spectrum analysis diagram of the photovoltaic inverter at node 1 after implementing the present invention; Figure 5 This is a schematic diagram of the current waveform at the grid connection point PCC when harmonic redistribution is not performed. Figure 6 A schematic diagram of the current waveform at the grid connection point PCC after harmonic redistribution using the present invention. Figure 7 This is a performance comparison chart of the total harmonic distortion (THD) and system line loss power at the grid connection point (PCC) under different load conditions for embodiments of the present invention and various comparative examples.

[0023] in Figure 7 In the diagram: Curve A represents the performance trajectory of Embodiment 1 of the present invention, Curve B represents the performance trajectory of Comparative Example 1 after removing the impedance correction coefficient, and Curve C represents the performance trajectory of Comparative Example 2 using the traditional average distribution strategy. Detailed Implementation

[0024] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The following specific examples illustrate the implementation of this disclosure, and those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0025] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The drawings only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex. Example

[0026] This embodiment constructs a collaborative power quality management environment for low-voltage distribution network areas, aiming to resolve the contradiction between harmonic pollution caused by nonlinear loads and the idle resources of photovoltaic inverters in scenarios with high penetration of distributed photovoltaics. Figure 1 As shown, the physical system mainly consists of multiple distributed photovoltaic grid-connected inverters operating in parallel, local nonlinear loads, distribution network transmission lines, and a point of common coupling (PCC). At the physical level, each inverter node is connected via a specific line impedance. The cable connects to the distribution network bus, and this distributed parameter characteristic inevitably causes current shunting and attenuation effects during transmission. The photovoltaic inverter selected in this embodiment adopts a three-phase full-bridge topology and is connected to the grid through an LCL filter. Its core controller adopts a high-performance digital signal processor (DSP) combined with a field-programmable gate array (FPGA) architecture to meet the stringent computing power requirements of high-frequency harmonic control.

[0027] During the initialization or periodic calibration phases of system operation, the control system first needs to establish an accurate power grid impedance sensing model. This is because the distribution network line impedance... It is a key physical quantity affecting the accuracy of remote compensation. The system injects a micro-disturbance current of a specific frequency (such as a frequency sweep signal of 200Hz-2kHz) into the power grid through an inverter, and uses the vector relationship between the response voltage and current to identify the line impedance through FFT analysis. and the load impedance of each node (including the load impedance of remote nodes) and local node load impedance Specifically, the amplitude range of the disturbance signal is set to 1%-5% of the rated current to ensure that it does not affect normal operation; the frequency selection is based on avoiding the fundamental frequency and its lower harmonics, and multiple frequency points in the range of 200Hz-2kHz are selected for frequency sweeping; the FFT analysis adopts a 1024-point Fast Fourier Transform algorithm, and the sampling frequency is set to 10kHz; the mathematical model for impedance identification is based on Ohm's law. Impedance-frequency characteristic curves are established through multi-frequency measurements. Alternatively, for distribution networks with relatively fixed topologies, the system can directly read line length and cable type parameters from the distribution network GIS / PMS system database via a communication interface, and calculate the impedance matrix by looking up a table. This process establishes a physical benchmark for subsequent mathematical corrections, ensuring that the control algorithm can perceive the real power grid topology environment. Based on the identified impedance parameters, the system calculates the compensation coefficient. According to the principle of circuit current shunting, when inverter j injects compensation current into remote node i, due to line impedance... With local impedance branch When a parallel shunt circuit is formed, the actual current reaching node i will decrease. To compensate for this decrease, a compensation factor is used. The calculation formula is: ,in For line impedance, The equivalent impedance of the local harmonic source. This represents the local node load impedance, and ' / / ' indicates parallel operation. The equivalent impedance of the local harmonic source... The harmonics are obtained by measuring the harmonic current and harmonic voltage at the local node and applying Ohm's law. The calculation yielded, where The amplitude of the harmonic voltage. This represents the harmonic current amplitude. The physical meaning of this coefficient is: by pre-amplifying the output command, the actual current reaching the target node after being shunted by the line impedance is exactly equal to the required compensation current. For example, if... , ,but This means the inverter needs to output 1.25 times the command current to ensure the target node receives 1 times the compensation effect. After entering the real-time control loop, as... Figure 2 As shown, the system first performs the "remaining capacity audit" action (corresponding to...). Figure 2 (Step S101 in the text). Figure 2Step S101 corresponds to the remaining capacity calculation process, step S102 corresponds to the task allocation and correction process, and step S103 corresponds to the current command setting process. The current transformers and voltage transformers of the photovoltaic inverter collect the voltage at the grid connection point in real time at a sampling frequency of 10kHz or higher. and current The current real-time active power is calculated through coordinate transformation. and real-time reactive power To ensure that harmonic mitigation does not encroach on the primary function of photovoltaic power generation, and to strictly adhere to the thermal stability limits of power electronic devices, the controller performs geometric subtraction operations within the vector space of apparent power. Specifically, the system follows the formula... Dynamically extract the remaining capacity of the inverter This calculation process physically defines a dynamic safety boundary: when the light intensity makes... Approaching rated capacity At that time, remaining capacity When the IGBT automatically converges to near zero, the system automatically locks out the harmonic compensation function to prevent overheating and damage. Conversely, in low light conditions or at night, the system releases a large amount of energy. The space transforms the inverter into a virtual active power filter (APF). This remaining capacity is then converted into a physically constrained maximum compensable harmonic current limit. After determining the "capabilities," the system enters the "task allocation and adjustment" phase (corresponding to...). Figure 2 (Step S102 in the original text). At this time, if there is a harmonic demand in the local load, the system prioritizes using the remaining capacity for local compensation. The criterion for satisfying local governance is: after local harmonic current compensation, the total harmonic distortion rate (THD) of the local node is lower than a preset threshold (e.g., 5%), or the local harmonic compensation current demand is less than the maximum compensable current corresponding to the remaining capacity. However, the core innovation of this embodiment lies in handling the working condition of "local underutilization and remote underutilization". When there is still remaining capacity after local governance is satisfied, the system starts the remote coordination mechanism. To address the problem of compensation failure caused by ignoring line impedance in traditional methods, the controller introduces a compensation coefficient based on the circuit shunting principle. Calculate the compensation coefficient based on the principle of current shunting in circuits. This coefficient is used to compensate for line impedance. This results in current attenuation. It should be noted that the above impedance parameters and compensation coefficient calculations are performed separately for specific harmonic frequencies (such as the 5th, 7th, 11th, and 13th harmonics), meaning the controller calculates each harmonic independently in the frequency domain. And superimposed. From a physical perspective, due to the line impedance... With local impedance branch This forms a parallel shunt circuit, meaning only a portion of the current generated by the inverter reaches the remote node. This coefficient The value is typically greater than 1, and its function is equivalent to a "predistortion gain," artificially amplifying the inverter's output command. This mathematical amplification precisely cancels out the line impedance during physical transmission. The resulting attenuation and shunting losses reduce the final speed at which the signal reaches the target node. The compensation current is precisely matched with the harmonic current at that point in both amplitude and phase (equal amplitude and opposite phase).

[0028] Finally, in the current command setting stage (corresponding to) Figure 2 (In step S103 of the process), the controller will transfer the photovoltaic grid-connected fundamental current. Local harmonic compensation current and after coefficient Corrected remote harmonic compensation components Vector superposition is performed to generate the final current control command. This instruction drives the space vector pulse width modulation (SVPWM) module to control the switching of the inverter's power switches, thereby outputting a current waveform containing precise harmonic components on the physical circuit.

[0029] To verify the effectiveness of this embodiment, a dual-inverter dual-nonlinear load simulation platform was built in the MATLAB / Simulink environment. The system parameters were set as follows: rated capacity of the photovoltaic inverter... The mains voltage is 380V, and the line impedance is... The experiment was conducted with node 1 lightly loaded and well-lit (with remaining capacity), and node 2 heavily loaded and experiencing severe harmonics. Without enabling the strategy of this invention (e.g....), Figure 3 and Figure 5 As shown in the figure, due to the limited capacity of the inverter at node 2, it is unable to fully compensate for the local harmonics, and node 1 does not participate in the coordination, resulting in severe distortion of the current waveform at the grid connection point PCC. The measured total harmonic distortion (THD) rate is as high as 18.4%, and the waveform shows an obvious "M"-shaped collapse.

[0030] After activating the method described in this invention, node 1 identifies the remaining capacity and calculates the compensation coefficient based on the line impedance. (Example value) Actively inject the corrected compensation current into node 2. For example... Figure 4 As shown, the output current of node 1 contains a significant subharmonic component superimposed on it. The waveform finally observed at point PCC is ( Figure 6The system exhibits highly smooth sinusoidal characteristics, effectively filtering out glitches and distortions. Instrumental analysis showed that the THD of the current at the PCC dropped sharply from 18.4% to 3.2%, and the content of each harmonic was lower than the requirements of the national standard GB / T14549. Further analysis of the correlation between electrical distance and compensation effect revealed that as the electrical distance (line impedance magnitude) between nodes increases, without the compensation coefficient correction of this invention, the amplitude error of the remote compensation current reaching the target node shows a non-linear increasing trend; however, with the application of this invention, this error is effectively suppressed at a constant low level and does not significantly worsen with increasing distance. This empirical data demonstrates that, through precise impedance correction and capacity synergy, this system successfully overcomes the physical barrier of line impedance and achieves precise harmonic mitigation across nodes. Example 2 This embodiment, building upon Embodiment 1, further elaborates on a modularly designed distribution network harmonic suppression system, focusing on its capacity allocation strategy and system interaction logic in a complex multi-node network. Unlike Embodiment 1, which emphasizes the control loop of a single unit, this embodiment focuses on coordinated scheduling at the distribution area level.

[0031] This system is integrated into the control firmware of the distributed photovoltaic inverter and mainly includes a compensation capacity calculation module, a compensation capacity allocation module, and a current command setting module. These modules can be represented as software code stored in memory and executed by a digital signal processor (DSP) or microcontroller (MCU), or as logic gates in an FPGA. In actual operation, the communication modules of each node construct an information exchange network. This communication network is not limited to PLC power line carrier or wireless communication; it can also be any communication method capable of data exchange, such as industrial Ethernet, fiber optic communication, or 5G network slicing. When the harmonic content of a node (denoted as node K) in the distribution area exceeds the limit and its own capacity is exhausted, it broadcasts a "harmonic support request" to the network. This request includes its node location and the required harmonic current spectrum data. Other nodes receiving the request are first processed by the compensation capacity calculation module. The calculation checks the current remaining "ammunition". If it exists... Then the compensation capacity allocation module intervenes. At this time, the system does not simply distribute the task equally, but executes an impedance sorting optimization strategy. The system reads the pre-stored or identified impedance matrix and calculates the electrical distance (i.e., line impedance magnitude) between its own node and the requesting node K. The system logic is set as follows: the closer the electrical distance (...). The smaller the value, the lower the transmission loss and the higher the compensation efficiency, therefore the higher the priority.

[0032] For example, in a system containing nodes A, B, and C, if C requests support, the impedance between A and C is... The impedance between B and C is The system will prioritize scheduling the remaining capacity of node A. Only when the remaining capacity of node A is completely exhausted and still insufficient to completely suppress the harmonics at point C will node B be activated to participate in compensation. After the task allocation is determined, nodes A and B respectively calculate their correction coefficients based on the circuit shunting principle using their respective line impedance parameters with point C. This coefficient is the reciprocal of the line impedance shunt ratio, used to offset the line impedance. The resulting current attenuation.

[0033] This impedance-based hierarchical allocation mechanism minimizes energy transmission loss across the entire distribution area at the physical level. Compared to Example 1, this example avoids the high line losses caused by forced compensation over long distances by introducing networked interaction and sorting logic. This ensures that while improving power quality, the economic operating indicators of the distribution network are maintained. Experimental data shows that in multi-node collaborative scenarios, the impedance sorting strategy reduces the overall line loss of the system by approximately 12.5% ​​compared to the traditional average distribution strategy, demonstrating the engineering practicality of this system architecture in complex topologies.

[0034] Description of Parallel Harmonic Mitigation Implementation: In actual power distribution networks, multiple harmonics of various frequencies, such as the 5th, 7th, and 11th harmonics, often coexist. To address this complex operating condition, this system supports parallel mitigation of multiple harmonics. Specifically, the system processes each harmonic separately in the frequency domain: First, it extracts each harmonic component through FFT analysis; second, for each harmonic (e.g., the 5th harmonic), it independently calculates its corresponding line impedance. And the load impedance, and substitute them into the formula to calculate the compensation coefficient at that frequency. At the same time, the total remaining capacity The harmonics are allocated according to their weights or preset priorities, and the maximum compensable limit for each harmonic is calculated. Finally, the corrected current commands calculated for each harmonic are superimposed in the time domain to generate the final wideband control command. This parallel processing mechanism ensures that the system can not only manage single-frequency harmonics but also effectively handle nonlinear loads with complex spectra. Comparative Example 1 To further verify the "compensation coefficient" in claims 1 and 2 "This core feature highlights the physical necessity of cross-node harmonic mitigation. To illustrate this, a specific experimental environment was constructed in this comparative example. Unlike the closed-loop control model in Example 1, which included line impedance correction, this comparative example maintains the photovoltaic inverter hardware topology and remaining capacity calculation logic (..." Under the premise that the basic current control loop parameters (such as PI parameters and switching frequency) are completely consistent with those in Example 1, the line impedance-based control algorithm was artificially removed. A module for calculating the compensation coefficient relationship with load impedance. Specifically, when generating remote harmonic compensation commands, the control system forcibly sets... This means assuming the distribution network is an ideal "copper plate" connection, ignoring the distributed parameter effects of the transmission lines, and directly sending the reverse-phase value of the harmonic current detected by the remote node as an instruction to the local inverter.

[0035] Experimental observations show that under ideal operating conditions with light loads or extremely short lines, this comparative ratio can maintain a certain level of mitigation effect. However, under the typical distribution network conditions described in Example 1 (line impedance...), the effect diminishes. After that, the system performance exhibited a significant nonlinear degradation. For example... Figure 7 As shown by curve B, with the increase of the remote nonlinear load rate, the total harmonic distortion (THD) at the PCC exhibits an uncontrollable upward trend, eventually stabilizing at around 8.7%, far exceeding the national standard requirement of 5%. In-depth analysis of its physical failure mechanism reveals that due to the lack of… Gain correction, the compensation current generated by the inverter flows through the line impedance A severe physical current shunting effect occurred. According to Kirchhoff's Current Law (KCL), the injected current was forcibly divided by the line impedance branch and the local load parallel branch according to their impedance ratios, resulting in a limited actual current reaching the remote target node. The compensation current amplitude is significantly attenuated, and due to the presence of line inductance, the phase of the arriving current is also shifted. This physical "command-response" mismatch means that although the inverter consumes the remaining capacity output current, it cannot achieve perfect cancellation with equal amplitude and opposite phase at the target point. Furthermore, at certain frequency points, the phase lag can even lead to harmful harmonic amplification. This reverse empirical process irrefutably proves that in a non-ideal distribution network environment, simply releasing capacity without accurate correction based on the impedance model (i.e., characteristic) is insufficient. This will directly lead to the logical paralysis of the remote collaborative governance mechanism, making it impossible to achieve the technical goal of reducing the harmonic content of the entire network. Comparative Example 2 This comparative example aims to simulate common multi-machine equal-capacity or capacity proportional allocation strategies in the prior art, to demonstrate the high level of inventiveness of the allocation strategy based on electrical distance (impedance) priority described in claim 7 of this invention. In this comparative example, the system hardware architecture, remaining capacity calculation, and single-machine control accuracy are strictly consistent with Embodiment 1. The only variable is the scheduling logic of the compensation capacity allocation module. The system is configured not to perceive the line impedance differences between nodes, but instead adopts a seemingly fair capacity-weighted averaging algorithm: when harmonic mitigation is required in the network, the central controller mechanically distributes the compensation task equally to all capable nodes according to the remaining capacity ratio of each inverter, without considering the physical distance between the nodes and the pollution source.

[0036] exist Figure 7 In the corresponding simulation test, node 2 was set as a severe harmonic source, while nodes 1 (near end, low impedance) and 3 (far end, high impedance) both had sufficient remaining capacity. Experimental data show that when running using the strategy of Comparative Example 2, the overall line loss of the system ( Compared to Example 1, it increased by approximately 35% (e.g. Figure 7 As shown in curve C), node 3 frequently triggers DC bus overvoltage protection. The root cause lies in the fact that this allocation strategy violates the principle of minimum resistance in power transmission. When the system forcibly schedules node 3, which is electrically distant, to undertake a large amount of compensation work, in order to overcome the impedance of long-distance lines… The resulting huge voltage drop means that node 3 must output a higher terminal voltage. This not only causes the inverter's modulated wave to enter the over-modulation region, causing clipping distortion of the output waveform, but also causes a large amount of valuable active / reactive energy to be dissipated as heat on the transmission line.

[0037] This physical-level thermodynamic failure and voltage stress concentration directly led to the collapse of the system's dynamic stability when dealing with high-frequency, large-amplitude load fluctuations. In contrast, Example 1, through impedance sorting logic, prioritizes the use of "electrical neighbors" for mitigation, essentially finding the "geodesic" (shortest path) for energy transmission, thus avoiding the high losses and voltage exceedance risks associated with long-distance transmission at the physical level. The waveform distortion (caused by inverter saturation) and sharp drop in energy efficiency observed in Comparative Example 2 under extreme conditions profoundly reveal the limitations of existing technologies in the absence of topology awareness, conversely confirming the decisive contribution of the impedance weighted allocation strategy introduced in this invention to maintaining the economical operation and safe stability of the distribution network.

[0038] Comprehensive comparison, for example Figure 7 The evolution trajectories of curves A (Example 1 of the present invention), B (Comparative Example 1, without impedance correction), and C (Comparative Example 2, without impedance sorting) clearly demonstrate the "feature-mechanism-effect" logical closed loop of the technical solution of the present invention. From Figure 7 It can be seen that curve A consistently maintains a low THD level below 3.2% across the entire load range, and the line loss remains at its lowest level. This is not a linear superposition of a single technical feature, but rather the result of the synergistic effect of "precise correction" and "optimal scheduling".

[0039] First, addressing the "long-range attenuation" problem revealed by curve B, this invention introduces a method that incorporates line impedance. and the compensation coefficient of the load impedance parameter A pre-distortion model is constructed mathematically. This model precisely compensates for current shunting and phase shift losses during transmission at the physical level, ensuring that the compensation current acts like a "scalpel" precisely on the affected area. This mathematical compensation for the physical transmission characteristics allows this invention to break through the spatial limitations of traditional methods that can only manage local or near-end harmonics, achieving full coverage without dead zones across the entire distribution area. Secondly, addressing the "high loss and easy saturation" defects exposed by curve C, this invention utilizes the remaining capacity... The joint optimization mechanism with electrical distance solves the resource allocation problem from a global perspective of energy flow. By locking the compensation task to the node with the closest electrical distance, the system effectively restricts harmonic currents to flow within the smallest physical area, avoiding secondary pollution of the power grid by ineffective circulating currents.

[0040] In summary, this invention establishes the capacity boundary by calculating the remaining capacity of the photovoltaic inverter in real time, and uses an impedance correction coefficient. Transmission accuracy was solved, and transmission efficiency was optimized through an impedance sequencing strategy. These three aspects are interconnected, overcoming the dual technical obstacles of "inaccurate treatment" (due to impedance shunting) and "high losses" (due to blind dispatching) faced by photovoltaic inverters in power quality management in existing technologies. Experimental data demonstrates that this invention has significant engineering practical value and unexpected technical effects without increasing the cost of additional APF hardware.

Claims

1. A method for suppressing harmonics in a distribution network based on the remaining capacity of a photovoltaic inverter, characterized in that, include: Obtain the line impedance and node load impedance of the power distribution network; Calculate the compensation coefficients used to correct the long-range harmonic compensation components based on the line impedance and the node load impedance. ; Obtain the remaining capacity of the photovoltaic inverter; Based on the remaining capacity, the final current command is generated using the compensation coefficient.

2. The method for suppressing harmonics in a distribution network based on the remaining capacity of a photovoltaic inverter according to claim 1, characterized in that, The remaining capacity The calculation formula is: ; in, For rated capacity, For real-time active power, This represents real-time reactive power.

3. The method for suppressing harmonics in a distribution network based on the remaining capacity of a photovoltaic inverter according to claim 1, characterized in that, The final current command The formula for generating it is: ; in, For photovoltaic grid-connected current, For local harmonic compensation current, For the remote harmonic compensation current component allocated to the local inverter, This is the correction coefficient for the remote harmonic compensation component.

4. The method for suppressing harmonics in a distribution network based on the remaining capacity of a photovoltaic inverter according to claim 1, characterized in that, The method also includes a capacity allocation strategy: prioritizing the use of the remaining capacity of the photovoltaic inverter to compensate for harmonics at local nodes; and allocating the remaining capacity to remote nodes when local governance is satisfied.

5. The method for suppressing harmonics in a distribution network based on the remaining capacity of a photovoltaic inverter according to claim 1, characterized in that, The method also includes a remote allocation principle: priority is determined based on the electrical distance or impedance between nodes, and the remaining capacity is allocated in order of increasing impedance.

6. The method for harmonic suppression in distribution networks based on the remaining capacity of photovoltaic inverters according to claim 1, characterized in that, The process of obtaining the remaining capacity of the photovoltaic inverter includes: real-time acquisition of the rated capacity of the photovoltaic inverter. Real-time active power and real-time reactive power .

7. The method for suppressing harmonics in a distribution network based on the remaining capacity of a photovoltaic inverter according to claim 3, characterized in that, Line impedance The impedance of the distribution network lines includes the node load impedance, which includes the remote node load impedance. and local node load impedance .

8. A power distribution network harmonic suppression system, characterized in that, include: The compensation capacity calculation module is configured to calculate the remaining capacity of the photovoltaic inverter. The compensation capacity calculation module is configured to execute the formula. To calculate the remaining capacity The compensation capacity allocation module is configured to calculate the compensation coefficient based on the line impedance and load impedance, and to formulate a capacity allocation strategy. The logical configuration of the compensation capacity allocation module is as follows: prioritize meeting local governance needs, and if there is any surplus, allocate it to remote nodes according to electrical distance priority. The current command tuning module is configured to generate a final control signal by combining the remaining capacity and the compensation coefficient.

9. The power distribution network harmonic suppression system according to claim 8, characterized in that, The system also includes multiple distributed photovoltaic grid-connected inverters, and the compensation capacity calculation module, the compensation capacity allocation module, and the current command setting module are integrated into the controller of the distributed photovoltaic grid-connected inverters.

10. The power distribution network harmonic suppression system according to claim 8, characterized in that, The system is applied to a distribution network area containing multiple photovoltaic inverter nodes, and the line impedance exists between each node.

Images