Power distribution network harmonic control method and system based on distributed power supply

Abstract

The invention discloses a power distribution network harmonic control method and system based on a distributed power supply, and belongs to the technical field of power distribution network harmonic control. Through power distribution network global topology directed graph modeling and harmonic time-space coupling degree analysis, harmonic source accurate grouping and harmonic emission responsibility quantification of each node are completed, targeted positioning of harmonic compensation is realized, and the problems of fuzzy harmonic source positioning and insufficient management and control pertinence of a traditional scheme are solved; a multi-target collaborative hierarchical control strategy is constructed, dynamic optimal distribution of compensation tasks is realized based on the residual capacity of the distributed power supply, the available capacity of a grid-connected inverter is utilized to the maximum extent, the harmonic compensation efficiency is greatly improved, and the investment of treatment equipment is reduced; the compensation current is accurately output through real-time harmonic impedance perception and a phase correction model, self-adaptive iteration of a control strategy is achieved by combining closed-loop effect verification and active prevention and control of resonance risks, parallel resonance and circulating current risks are effectively avoided, and the power quality and operation stability of a power distribution network are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of distribution network harmonic control technology, and particularly to a method and system for distribution network harmonic control based on distributed generation. Background Technology

[0002] Distributed power sources are connected to the grid via power electronic inverters, whose switching operations generate a large amount of harmonic current. Furthermore, the large-scale application of power electronic equipment such as nonlinear industrial loads and charging piles for new energy vehicles in the distribution network further exacerbates harmonic pollution. Harmonic distortion can lead to increased line losses in the distribution network, inaccurate metering equipment, malfunctioning relay protection, and even parallel resonance, seriously threatening the safe and stable operation of the distribution network.

[0003] Currently, harmonic mitigation in distribution networks primarily employs fixed equipment such as passive and active power filters for single-point compensation, which is ill-suited for distribution network architectures with distributed generation and multiple harmonic sources. Existing harmonic compensation schemes based on distributed generation inverters mainly focus on local single-point harmonic suppression, lacking a systematic analysis of the harmonic transmission characteristics across the entire distribution network. This makes it difficult to accurately quantify the harmonic emission responsibility of each node, easily leading to compensation blind spots and overcompensation issues. Furthermore, when multiple distributed generation sources collaborate in compensation, the spatiotemporal coupling characteristics of harmonics are not considered, easily resulting in imbalances in compensation capacity allocation, resonance and circulating currents caused by multiple parallel connections, and failing to meet the requirements for refined harmonic control across the entire distribution network under high-proportion distributed generation access. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for harmonic control in power distribution networks based on distributed power sources, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: The method for harmonic control in distribution networks based on distributed generation includes the following steps: Acquire basic parameter data of the distribution network, obtain electrical sampling data of each node in the entire distribution network based on the basic parameter data of the distribution network, extract harmonic features of the electrical sampling data, and generate a set of harmonic feature parameters and a directed graph of the topology of the entire distribution network. Based on the directed graph of the distribution network topology and the set of harmonic characteristic parameters, the spatiotemporal coupling degree of harmonic components is calculated and the harmonic source is grouped. The harmonic emission responsibility of each node is quantified, and the harmonic compensation requirement parameters are generated. Based on the harmonic compensation requirement parameters, harmonic hierarchical classification processing is performed, and a multi-objective collaborative optimization strategy is constructed to generate a hierarchical control instruction set, which is then distributed to the corresponding local execution unit. The local grid-connected point's real-time harmonic impedance characteristics are obtained based on the issued hierarchical control instruction set. A harmonic phase correction model is constructed based on the real-time harmonic impedance characteristics. Correction and control are performed based on the harmonic phase correction model, and harmonic compensation current is output. After collecting voltage and current signals from all nodes in the entire distribution network after harmonic compensation, the system performs harmonic control effect verification and resonance risk prevention based on the electrical sampling data of the entire distribution network after harmonic compensation. Finally, the system generates control strategy parameters based on the harmonic control effect verification results and resonance risk coefficient.

[0006] Furthermore, the construction process of the global topology directed graph of the distribution network specifically includes: Based on the basic parameter data of the distribution network, the physical nodes of the entire distribution network are traversed, the coverage area of ​​each node is defined, and a unique identity is assigned to each node. The physical nodes of the entire distribution network are classified according to the type of equipment and electrical function of each node. Based on the classification results, the rated parameters of the distributed power grid-connected unit and the access information of nonlinear loads are bound to the corresponding nodes. The electrical sampling data and the harmonic characteristic parameter set are mapped to each node to form a node set of the entire distribution network. Based on the matching of the topological location information of each node in the distribution network topology and electrical connection relationship, an initial branch set corresponding one-to-one with the physical feeders and electrical connections of the distribution network is constructed, the dominant direction of harmonic propagation of the branch is determined, and the initial branch set is mapped to an initial directed edge set with directional attributes. The harmonic transfer weights of each directed edge in the initial directed edge set are calculated based on the harmonic characteristic parameter set, and the directed edge set is then quantized and calibrated. Based on the global node set of the distribution network and the directed edge set after quantization and calibration, an initial global topology directed graph of the distribution network is constructed, and the global node connectivity of the initial topology directed graph is checked and structural anomalies are corrected. Based on the corrected global topology directed graph of the distribution network, a harmonic transmission correlation matrix is ​​constructed. Combined with the global node set of the distribution network and the quantized and calibrated directed edge set, the standardized encapsulation of the topology directed graph is completed, generating the final global topology directed graph of the distribution network.

[0007] Furthermore, each branch in the initial branch connection set corresponds to the distribution network feeder line and physical electrical connection relationship, and the two ends of the branch correspond to two associated nodes in the global node set. Based on the fundamental power flow direction, phase difference and amplitude attenuation characteristics of the same frequency harmonics of the two associated nodes at the two ends of the branch, the dominant direction of harmonic propagation of the branch is determined, and the branch is mapped as a directed edge. The fundamental impedance parameters and harmonic impedance parameters of the corresponding branch are bound to each directed edge to form the initial directed edge set.

[0008] Furthermore, the calculation process for the harmonic transfer weights of each directed edge is as follows: The directed edges in the initial directed edge set are extracted. The starting point of the directed edge in the positive direction is taken as the first node and the ending point in the positive direction is taken as the last node. Based on the harmonic characteristic parameter set, the amplitude and phase measured parameters of the harmonics and interharmonics of the same frequency at the first and last nodes of the corresponding branches of the directed edge are extracted to determine the calculation benchmark of each frequency harmonic of the directed edge. Based on the established harmonic calculation benchmarks for each frequency, the harmonic amplitude ratio and phase difference change rate of the end node and the beginning node of the directed edge at the same frequency are calculated to obtain the initial harmonic transmission coefficient at that frequency. Based on the fundamental impedance parameters and harmonic impedance parameters of the corresponding branches bound by the directed edge, and combined with the branch's line type, physical length, and impedance frequency characteristics, the initial harmonic transmission coefficients of each frequency are corrected to obtain the harmonic transmission weight of the directed edge at the corresponding frequency. The harmonic transfer weights of all frequency harmonics and interharmonics corresponding to the directed edges are combined in ascending order of frequency to form the harmonic transfer weight vector of the directed edges, thus completing the quantization and calibration of the harmonic transfer weights of all directed edges in the initial directed edge set.

[0009] Furthermore, the calculation process for the spatiotemporal coupling degree of the harmonic components is as follows: Based on the global topological directed graph of the distribution network and the set of harmonic characteristic parameters, all node pairs with harmonic propagation reachability in the topological directed graph are traversed. Extract the amplitude and phase time series data of all frequency harmonics and interharmonics corresponding to each node within the preset sliding time window, and determine the calculation frequency interval; Based on the extracted harmonic amplitude time series data of each node, the temporal synchronization degree of the node pair in the same frequency harmonic amplitude change within the sliding time window is calculated, and the temporal correlation data of the dynamic change of harmonic amplitude of the node pair at the corresponding frequency is obtained. Based on the harmonic propagation path between nodes in the directed graph of the entire distribution network topology, the theoretical cumulative phase difference of all directed edges on the harmonic propagation path is calculated. Combined with the measured phase difference of the nodes for the same frequency harmonic, the degree of matching between the harmonic phase propagation characteristics and the inherent propagation law of the topology is calculated. Based on the degree of matching between time correlation data, harmonic phase propagation characteristics and the inherent propagation law of the topology, the spatiotemporal coupling degree of the harmonic components of the node at the corresponding frequency is obtained.

[0010] Furthermore, the harmonic source grouping process includes: Based on the spatiotemporal coupling degree of harmonic components of all node pairs at all frequencies, the comprehensive spatiotemporal coupling degree of each node pair is calculated, and the harmonic coupling degree correlation matrix of all nodes in the distribution network is constructed. Based on the harmonic coupling degree correlation matrix, all harmonic emitting nodes are initially clustered to obtain multiple initial harmonic source clusters. Based on the harmonic transmission matrix and harmonic transmission weight vector of the directed graph of the global topology of the distribution network, the dominant harmonic transmission path in the initial harmonic source cluster is identified, and the nodes on the dominant harmonic transmission path and the associated feeder nodes of the harmonic emitting nodes are taken as the coverage node range of the initial harmonic source cluster. Based on the harmonic component timing characteristics of each node within the coverage area of ​​the initial harmonic source cluster and the harmonic transmission attenuation characteristics of the corresponding dominant transmission path, the harmonic characteristic matching degree between each node and the reference node of the dominant transmission path is calculated, and free nodes with matching degree lower than the preset matching degree threshold are removed. Based on the directed graph of the entire distribution network topology, the initial harmonic clusters with electrical connections and adjacent coverage areas are identified as adjacent clusters. The comprehensive spatiotemporal coupling degree between adjacent clusters is calculated, and adjacent clusters with comprehensive spatiotemporal coupling degree exceeding the preset merging threshold are merged to generate merged harmonic source clusters, thus completing the harmonic source grouping process.

[0011] Furthermore, the construction of the multi-objective collaborative optimization strategy includes: Based on harmonic compensation demand parameters, the directed graph of the distribution network topology, and the set of harmonic characteristic parameters, the local harmonic compensation sensing range of each distributed power grid-connected unit is obtained, and the communication interaction relationship between adjacent distributed power nodes is determined. Each distributed power grid-connected unit generates local initial compensation capacity and virtual impedance matching parameters based on locally collected harmonic data, harmonic source compensation requirements within the sensing range, and the remaining available capacity of the local inverter. Adjacent distributed power grid-connected units exchange local initial compensation capacity, virtual impedance parameters and harmonic sensing data through peer-to-peer communication, and calculate the compensation redundancy and compensation gap between adjacent nodes. Each distributed power grid-connected unit adjusts its local initial compensation capacity based on compensation redundancy and compensation gap, performs constraint verification on the adjusted local compensation capacity, and generates local compensation tasks for each distributed power grid-connected unit. A hierarchical control instruction set is generated based on the local compensation task of each distributed power grid-connected unit.

[0012] Furthermore, the construction of the multi-objective collaborative optimization strategy also includes: The compensation capacity adjustment step size threshold of each distributed power grid-connected unit is preset, the fluctuation amplitude of harmonic compensation demand is obtained, and it is determined whether the fluctuation amplitude is lower than the preset compensation capacity adjustment step size threshold. If the fluctuation range is lower than the preset compensation capacity adjustment step size threshold, the distributed power source will not adjust the local compensation capacity. Acquire harmonic distortion rate data within the sensing range and determine whether the harmonic distortion rate exceeds the preset distortion standard; If the harmonic distortion rate exceeds the preset distortion standard, the comprehensive spatiotemporal coupling degree between each distributed power source and the harmonic source, and the remaining capacity of each distributed power source are obtained. The distributed power source with the highest comprehensive spatiotemporal coupling degree and the most sufficient remaining capacity is selected to undertake the incremental compensation task. Obtain the virtual impedance parameters of adjacent distributed power sources, perform impedance matching and collaborative calibration based on the virtual impedance parameters of adjacent distributed power sources, and update the hierarchical control instruction set.

[0013] This invention provides another technical solution: a distribution network harmonic control system based on distributed power sources, comprising: The harmonic sensing modeling module is used to acquire electrical sampling data of all nodes in the entire distribution network based on the obtained distribution network topology parameters, rated parameters of distributed power grid-connected units, and nonlinear load access information, through the synchronous signal acquisition unit, to extract harmonic features and generate a directed graph of the entire distribution network topology and a set of harmonic feature parameters. The harmonic grouping and quantization module is used to calculate the spatiotemporal coupling degree of harmonic components based on the directed graph of the distribution network topology and the set of harmonic characteristic parameters. Based on the calculation results, it performs harmonic source grouping processing, quantifies the harmonic emission responsibility of each node, and generates harmonic compensation requirement parameters. The hierarchical instruction distribution module is used to construct a hierarchical classification and collaborative allocation strategy for harmonics based on the harmonic compensation requirement parameters generated by the harmonic grouping and quantization module, generate a hierarchical control instruction set with stable boundary constraints, and distribute it to the corresponding execution unit. The local admittance control module is used to obtain the real-time harmonic impedance characteristics of the local grid connection point based on the issued hierarchical control instruction set, construct a harmonic phase correction model and perform correction and control execution, and output harmonic compensation current.

[0014] Furthermore, the local admittance control module is also used for: Based on the issued hierarchical control instruction set, real-time voltage and current sampling data of the local grid connection point are obtained, and the real-time harmonic impedance characteristics and background harmonic components of the grid side of the local grid connection point are extracted. The target harmonic phase offset is calculated based on the real-time harmonic impedance characteristics and the reference impedance parameters in the hierarchical control instruction set. The virtual admittance phase angle parameters in the hierarchical control instruction set are corrected based on the target harmonic phase offset to generate the corrected virtual admittance phase angle. The virtual admittance amplitude parameters in the hierarchical control instruction set are dynamically limited and corrected to generate the corrected virtual admittance amplitude. Based on the corrected virtual admittance phase angle and the corrected virtual admittance amplitude, combined with the background harmonic components of the power grid side, the final virtual admittance control parameters are generated. The control loop of the distributed power grid-connected inverter is adjusted based on the final virtual admittance control parameters to generate a total current reference command, which drives the inverter to output the corresponding grid-connected current.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention achieves accurate harmonic source grouping by modeling the entire distribution network topology with a directed graph and quantifying the harmonic propagation weights, combined with the calculation of the spatiotemporal coupling degree of harmonic components. It can accurately quantify the harmonic emission responsibility of each node, distinguish the contribution ratio of native harmonics of distributed generation sources, load harmonics, and background harmonics of the grid, and process the harmonic compensation demand in a hierarchical and classified manner. Combined with the collaborative optimization allocation of the remaining capacity of distributed generation sources, it achieves dynamic balance between supply and demand of compensation capacity through peer-to-peer communication between adjacent nodes. It prioritizes the distributed generation sources with the highest coupling degree and the most sufficient remaining capacity to undertake the compensation task, maximizing the utilization of the remaining capacity of the grid-connected inverters of distributed generation sources while ensuring the compensation effect, and avoiding the investment of additional filtering equipment.

[0016] 2. This invention achieves accurate correction output of compensation current through real-time sensing of local harmonic impedance and harmonic phase correction model; at the same time, through closed-loop verification of the full-domain electrical data after compensation, it realizes adaptive iterative optimization of control strategy, actively identifies and prevents resonance risks, solves the resonance and circulating current problems that are easily caused by parallel compensation of multiple distributed power sources, and ensures stable operation of high-proportion distributed power sources connected to the distribution network across the entire frequency band. Attached Figure Description

[0017] Figure 1 This is a flowchart of the distribution network harmonic control method based on distributed power sources according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 The present invention provides the following technical solutions: The method for harmonic control in distribution networks based on distributed generation includes the following steps: Acquire basic distribution network parameter data, including distribution network topology parameters, rated parameters of distributed power grid-connected units, and nonlinear load access information. Based on the basic distribution network parameter data, acquire electrical sampling data of each node in the entire distribution network through the synchronous signal acquisition unit, including voltage and current time-domain signals, output power, inverter switching frequency, DC bus voltage, and inverter remaining available capacity. Extract harmonic features from the electrical sampling data and generate a set of harmonic feature parameters and a directed graph of the entire distribution network topology. Based on the directed graph of the distribution network topology and the harmonic characteristic parameter set, the spatiotemporal coupling degree of harmonic components is calculated and the harmonic source is grouped. The spatiotemporal coupling degree is a weighted sum of the temporal correlation of amplitude changes and the spatial transmission consistency of phase difference of the same frequency harmonic between different nodes. The harmonic source grouping process corresponds to a unique harmonic dominant transmission path and coverage node range for each cluster, quantifies the harmonic emission responsibility of each node, distinguishes the contribution ratio of native harmonics of distributed power sources, load harmonics, and background harmonics of the power grid, and generates harmonic compensation requirement parameters, including the location of compensation nodes, target compensation frequency, target compensation capacity for a single harmonic, compensation priority ranking, compensation accuracy threshold, and compensation response time limit. Based on the harmonic compensation requirement parameters, harmonic hierarchical classification processing is performed, and a multi-objective collaborative optimization strategy is constructed to generate a hierarchical control instruction set with stable boundary constraints. The hierarchical control instruction set is then distributed to the corresponding local execution unit. The local grid-connected point's real-time harmonic impedance characteristics are obtained based on the issued hierarchical control instruction set. A harmonic phase correction model is constructed based on the real-time harmonic impedance characteristics. Correction and control are performed based on the harmonic phase correction model, and harmonic compensation current is output. After collecting voltage and current signals from all nodes in the entire distribution network after harmonic compensation, the system performs harmonic control effect verification and resonance risk prevention based on the electrical sampling data of the entire distribution network after harmonic compensation. Based on the harmonic control effect verification results and resonance risk coefficient, the system generates control strategy parameters. That is, when the distribution network experiences changes in operating conditions such as the addition of distributed power sources, significant load changes, or topology adjustments, the system updates the topology directed graph, source-network interaction model, and control instruction set to achieve adaptive iterative optimization of distribution network harmonic control.

[0020] In this embodiment, by systematically collecting basic parameters of the distribution network and electrical sampling data of all nodes, the effective extraction of harmonic characteristics and the construction of a directed graph of the distribution network topology are achieved. At the same time, the harmonic emission responsibility of different nodes can be quantitatively distinguished, and the contribution ratio of native harmonics of distributed power sources, load harmonics and background harmonics of the power grid can be clarified. This enables a refined sorting of harmonic compensation requirements and a reasonable planning of priorities. It takes into account the governance needs of different nodes and harmonics of different frequencies in the distribution network, generates corresponding hierarchical control commands, and constructs corresponding harmonic phase correction models to complete the dynamic correction of compensation control. This can adapt to the dynamic changes in impedance characteristics during power grid operation, improve the adaptability of harmonic compensation current output to actual governance needs, and ensure the execution effect of harmonic governance.

[0021] In this embodiment, the process of constructing the directed graph of the global distribution network topology specifically includes: Based on the basic parameter data of the distribution network, the physical nodes of the entire distribution network are traversed, and the coverage of each node is defined, including the distribution network bus nodes, feeder key nodes, distributed power grid connection points, nonlinear load access points, and grid common connection points (PCCs). A unique identity is assigned to each node to ensure that the node identity corresponds one-to-one with the physical location. Based on the equipment type and electrical function of each node, the physical nodes of the entire distribution network are classified into categories, including distributed power grid-connected nodes, nonlinear load access nodes, grid common connection nodes, distribution network switch nodes, and passive branch nodes. Based on the classification results, the rated parameters of the distributed power grid-connected units and the nonlinear load access information are bound to the corresponding nodes. The electrical sampling data and harmonic characteristic parameter sets are mapped to each node to form a distribution network node set. For each node, a multi-dimensional attribute set containing topological location information, rated electrical parameters, real-time operating electrical quantities, and harmonic characteristic vectors is generated to complete the unique mapping between the node and all electrical parameters. Based on the matching of the topological location information of each node in the distribution network topology and electrical connection relationship, an initial branch set corresponding one-to-one with the physical feeders and electrical connections of the distribution network is constructed. The dominant direction of harmonic transmission of the branch is determined according to the harmonic transmission characteristics of the nodes at both ends of the branch, and the initial branch set is mapped to an initial directed edge set with direction attributes. The harmonic transfer weights of each directed edge in the initial directed edge set are calculated based on the harmonic characteristic parameter set, and the directed edge set is then quantized and calibrated. Based on the global node set of the distribution network and the quantized and calibrated directed edge set, an initial global topology directed graph of the distribution network is constructed. The initial topology directed graph is then subjected to full node connectivity verification and structural anomaly correction. Specifically: Using the grid common connection point, the grid connection point of the distributed power generation unit, and the nonlinear load access point as core anchor points, a full-domain traversal search algorithm is used to perform full node connectivity verification on the initial directed graph, and topological island nodes and directed edge direction conflicts are identified. For the identified topological island nodes, corresponding directed edges are added based on the switch status and branch connection relationship in the distribution network topology parameters, and the island nodes are included in the global connectivity domain to ensure that all physical nodes connected to the distribution network are included in the connectivity system of the topological directed graph; for directed edges with abnormal directional conflicts, the dominant direction of the directed edge and the harmonic transmission weight vector are recalibrated based on the real-time harmonic characteristic parameters and fundamental power flow direction of the nodes at both ends of the corresponding branch to eliminate directional conflicts and ensure that the direction of the directed edge matches the actual harmonic transmission characteristics. After anomaly correction, the reachability of harmonic propagation of all nodes in the directed graph is verified to ensure that there is at least one computable directed path for harmonic propagation between any two connected nodes, providing a complete topological basis for subsequent global harmonic propagation characteristic analysis. Based on the corrected directed graph of the global distribution network topology, a node adjacency matrix is ​​constructed. The dimension of the adjacency matrix is ​​consistent with the number of nodes in the global node set. The element value of the adjacency matrix is ​​the harmonic transmission weight vector of the directed edge between two corresponding nodes. The element value of the element with no direct directed connection is a zero vector. The digital mapping of the topology directed graph is completed through the adjacency matrix. The harmonic transmission correlation matrix is ​​calculated and generated based on the adjacency matrix. Combined with the global node set of the distribution network and the quantized and calibrated directed edge set, the standardized encapsulation of the topology directed graph is completed, generating the final global topology directed graph of the distribution network.

[0022] In this embodiment, each branch of the initial branch connection set corresponds to the distribution network feeder line and the physical electrical connection relationship, and the two ends of the branch correspond to two associated nodes in the global node set; For each branch, based on the fundamental power flow direction of the two associated nodes at both ends of the branch, and the phase difference and amplitude attenuation characteristics of the harmonics of the same frequency, the dominant direction of harmonic propagation for that branch is determined, and the branch is mapped as a directed edge, wherein the positive direction of the directed edge is strictly marked according to the actual direction of harmonic propagation: During harmonic propagation in a branch, the amplitude gradually attenuates along the propagation direction. Therefore, the opposite direction of harmonic amplitude attenuation is used as the positive direction reference. Simultaneously, the phase gradually lags along the propagation direction during harmonic propagation. Therefore, the positive direction points from the leading-phase node to the lagging-phase node. These two calibration rules jointly constrain and ensure that the positive direction of the directed edge perfectly matches the actual harmonic propagation direction, while retaining the reverse directed edge to cover the reverse harmonic propagation scenario. Each directed edge is bound to the fundamental impedance parameter and the impedance parameters of each harmonic of the corresponding branch to form an initial set of directed edges.

[0023] In this embodiment, by traversing all physical nodes of the distribution network, assigning unique identifiers, and classifying them, a precise and unique mapping between nodes and their physical locations and all electrical parameters is achieved. Combined with the dual constraints of harmonic amplitude attenuation and phase propagation characteristics at both ends of a branch, the mapping from branch to directed edges is completed and the direction is calibrated. Simultaneously, branch impedance parameters are bound, effectively solving the problem that traditional undirected topologies cannot reflect the directionality of harmonic propagation, accurately reflecting the harmonic propagation characteristics of distribution network branches. Through core node anchoring, global connectivity verification, and anomaly correction, it is ensured that all physical nodes are included in the connected system and that the direction of directed edges matches the actual harmonic propagation characteristics, guaranteeing consistency between the directed topology graph and the actual operating conditions of the power grid. By constructing a node adjacency matrix, digital mapping and standardized encapsulation of the topology are completed, enabling the directed topology graph to be directly adapted to the subsequent harmonic analysis and control process, providing standardized and computable topological support for harmonic control across the entire distribution network.

[0024] In this embodiment, the harmonic transfer weight calculation process for each directed edge is as follows: The directed edges in the initial directed edge set are extracted. The starting point of the directed edge in the positive direction is taken as the first node and the ending point in the positive direction is taken as the last node. Based on the harmonic characteristic parameter set, the amplitude and phase measured parameters of the harmonics and interharmonics of the same frequency at the first and last nodes of the corresponding branches of the directed edge are extracted to determine the calculation benchmark of each frequency harmonic of the directed edge. Based on the established harmonic calculation benchmarks for each frequency, the harmonic amplitude ratio and phase difference rate of change of the end node and the beginning node of the directed edge are calculated at the same frequency. Based on the harmonic frequency level and the priority of harmonic control in the distribution network, corresponding weighting coefficients are assigned to the amplitude ratio and phase difference rate of change. The weighted amplitude ratio and phase difference rate of change are summed to obtain the initial harmonic transmission coefficient at that frequency. The amplitude ratio is calculated using the following formula: In the formula, is the amplitude carryover ratio of the h-th harmonic, which characterizes the amplitude attenuation characteristics of the harmonic propagating along the positive direction of the directed edge; As the first node, For end nodes; This represents the measured amplitude at the first node; This represents the measured amplitude at the end node; The rate of change of phase difference is calculated using the following formula: In the formula, is the phase difference change rate of the h-th harmonic; L is the physical length of the branch corresponding to the directed edge, which characterizes the phase shift characteristics of the harmonic propagation along the positive direction of the directed edge. The measured phase of the end node; The measured phase of the first node; The initial harmonic transmission coefficient is calculated using the following formula: In the formula, Let be the initial harmonic transmission coefficient of the h-th harmonic; These are the weighting coefficients corresponding to the amplitude ratio; These are the weighting coefficients corresponding to the phase difference change rate; the values ​​of the weighting coefficients are adaptively adjusted according to the harmonic frequency level and the priority of harmonic control in the distribution network. Based on the fundamental impedance parameters and harmonic impedance parameters of the corresponding branches bound by the directed edge, and combined with the branch's line type, physical length, and impedance frequency characteristics, the initial harmonic transmission coefficients of each frequency are corrected to eliminate the deviation of the inherent impedance characteristics of the branch in the calculation of harmonic transmission capacity, and the harmonic transmission weight of the directed edge at the corresponding frequency is obtained. The corrected harmonic carry weight is calculated using the following formula: In the formula, The harmonic transfer weight for the h-th harmonic; This is the correction factor for the transmission coefficient of the h-th harmonic; It is calculated based on the harmonic impedance frequency characteristics of the branch, the line type coefficient, and the physical length, and is used to correct the inherent influence of the branch's inherent impedance on the harmonic transmission amplitude attenuation and phase shift. The harmonic transfer weights corresponding to all frequency harmonics and interharmonics of the directed edges are combined in ascending order of frequency to form the harmonic transfer weight vector of the directed edges. The harmonic transfer weight vectors of all directed edges in the entire distribution network are uniformly normalized to eliminate the difference in amplitude of harmonics of different frequencies, and the harmonic transfer weights of all directed edges in the initial directed edge set are quantitatively calibrated.

[0025] In this embodiment, by accurately extracting the measured harmonic parameters of the first and last nodes, and constructing the initial transmission coefficient by combining the amplitude ratio and the phase difference change rate, and by making targeted corrections based on the inherent impedance characteristics of the branches and the line parameters, the deviation of the inherent characteristics of the branches in the calculation of harmonic transmission capacity is effectively eliminated. By allocating weighting coefficients according to harmonic frequency and control priority, combining weight vectors and performing interval normalization, the difference in amplitude magnitude of harmonics of different frequencies is eliminated, and the accurate quantitative calibration of the harmonic transmission weight of the directed edges is realized. This can truly reflect the harmonic transmission capacity of each directed edge at different frequencies, and improve the scientificity and rationality of the harmonic analysis and control of the entire distribution network.

[0026] In this embodiment, the calculation process for the spatiotemporal coupling degree of the harmonic components is as follows: Based on the global topology directed graph and harmonic characteristic parameter set of the distribution network, with distributed power grid connection nodes, nonlinear load access nodes, and grid common connection points (PCCs) as core computing nodes, all node pairs with harmonic propagation reachability in the topology directed graph are traversed. Extract the amplitude and phase time series data of all frequency harmonics and interharmonics corresponding to each node within the preset sliding time window, and synchronously match the harmonic propagation path of the corresponding node pair in the topological directed graph and the harmonic propagation weight vector data of each directed edge to determine the calculation frequency interval. Based on the extracted harmonic amplitude time series data of each node, the temporal synchronization degree of the node pair in the same frequency harmonic amplitude change within the sliding time window is calculated, and the temporal correlation data of the dynamic change of harmonic amplitude of the node pair at the corresponding frequency is obtained. Based on the harmonic propagation path between nodes in the directed graph of the entire distribution network topology, the theoretical cumulative phase difference of all directed edges on the harmonic propagation path is calculated. Combined with the measured phase difference of the nodes for the same frequency harmonic, the degree of matching between the harmonic phase propagation characteristics and the inherent propagation law of the topology is calculated. Based on the degree of matching between time correlation data, harmonic phase propagation characteristics and the inherent propagation law of the topology, the spatiotemporal coupling degree of the harmonic components of the node at the corresponding frequency is obtained.

[0027] In this embodiment, the harmonic source grouping processing includes: Based on the spatiotemporal coupling degree of harmonic components of all node pairs at all frequencies, the comprehensive spatiotemporal coupling degree of each node pair is calculated, and the harmonic coupling degree correlation matrix of all nodes in the distribution network is constructed. The matrix element value is the comprehensive spatiotemporal coupling degree between the corresponding two nodes, and the element value of the node pair with no harmonic transmission reachability is 0. Based on the harmonic coupling degree correlation matrix, all harmonic emitting nodes are initially clustered to obtain multiple initial harmonic source clusters, including distributed power grid-connected nodes and nonlinear load access nodes. After the initial clustering is completed, the comprehensive spatiotemporal coupling degree between nodes within the same cluster is higher than the preset coupling degree threshold, while the comprehensive spatiotemporal coupling degree between nodes in different clusters is lower than the preset isolation threshold. Based on the harmonic transmission matrix and harmonic transmission weight vector of the directed graph of the global topology of the distribution network, the dominant harmonic transmission path in the initial harmonic source cluster is identified, and the nodes on the dominant harmonic transmission path and the associated feeder nodes of the harmonic emitting nodes are taken as the coverage node range of the initial harmonic source cluster. Based on the harmonic component timing characteristics of each node within the coverage area of ​​the initial harmonic source cluster and the harmonic transmission attenuation characteristics of the corresponding dominant transmission path, the harmonic characteristic matching degree between each node and the reference node of the dominant transmission path is calculated, and free nodes with matching degree lower than the preset matching degree threshold are removed. Based on the directed graph of the entire distribution network topology, the initial harmonic clusters with electrical connections and adjacent coverage areas are identified as adjacent clusters. The comprehensive spatiotemporal coupling degree between adjacent clusters is calculated, and adjacent clusters with comprehensive spatiotemporal coupling degree exceeding the preset merging threshold are merged to generate merged harmonic source clusters. This ensures that each final cluster corresponds to a unique harmonic dominant transmission path and an independent coverage node range, thus completing the harmonic source grouping process.

[0028] In this embodiment, the calculation of the spatiotemporal coupling degree of harmonic components is based on core nodes. Combining harmonic time series data and topological transmission path information within a sliding time window, the temporal correlation and phase transmission matching degree of harmonics of the same frequency between node pairs are accurately quantified. This comprehensively reflects the spatiotemporal distribution and transmission correlation characteristics of harmonic components, providing scientific quantitative support for harmonic source clustering. The harmonic source clustering process completes the initial clustering based on the harmonic coupling degree correlation matrix. Combining topological transmission characteristics, the dominant path and coverage of the cluster are defined. By eliminating detached nodes and merging adjacent clusters, it is ensured that each final cluster corresponds to a unique dominant transmission path and independent coverage. This effectively clarifies the influence range and transmission rules of different harmonic sources, accurately distinguishes the propagation paths of various harmonics, and provides a clear cluster division basis for subsequent quantification of node harmonic emission responsibility, clarification of the contribution ratio of various harmonics, and formulation of precise harmonic compensation strategies. This further improves the pertinence and effectiveness of harmonic governance in the distribution network.

[0029] In this embodiment, the construction of a multi-objective collaborative optimization strategy includes: Based on harmonic compensation demand parameters, the directed graph of the distribution network topology, and the set of harmonic characteristic parameters, the local harmonic compensation sensing range of each distributed power grid-connected unit is obtained, and the communication interaction relationship between adjacent distributed power nodes is determined. Each distributed power grid-connected unit generates local initial compensation capacity and virtual impedance matching parameters based on locally collected harmonic data, harmonic source compensation requirements within the sensing range, and the remaining available capacity of the local inverter. Adjacent distributed power grid-connected units exchange local initial compensation capacity, virtual impedance parameters and harmonic sensing data through peer-to-peer communication, and calculate the compensation redundancy and compensation gap between adjacent nodes. Each distributed power grid-connected unit adjusts its local initial compensation capacity based on compensation redundancy and compensation gap to achieve a balance between supply and demand of compensation capacity within the sensing range; and based on the active power output priority rules of distributed power sources and the stability boundary conditions of full-band harmonic control, it performs constraint verification on the adjusted local compensation capacity and generates local compensation tasks for each distributed power grid-connected unit. A hierarchical control instruction set is generated based on the local compensation task of each distributed power grid-connected unit.

[0030] In this embodiment, the construction of the multi-objective collaborative optimization strategy further includes: The compensation capacity adjustment step size threshold of each distributed power grid-connected unit is preset, the fluctuation amplitude of harmonic compensation demand is obtained, and it is determined whether the fluctuation amplitude is lower than the preset compensation capacity adjustment step size threshold. If the fluctuation amplitude is lower than the preset compensation capacity adjustment step size threshold, the distributed power source will not adjust the local compensation capacity, thus reducing the number of compensation actions. Acquire harmonic distortion rate data within the sensing range and determine whether the harmonic distortion rate exceeds the preset distortion standard; If the harmonic distortion rate exceeds the preset distortion standard, the comprehensive spatiotemporal coupling degree of each distributed power source and the harmonic source, and the remaining capacity of each distributed power source are obtained. The distributed power source with the highest comprehensive spatiotemporal coupling degree and the most sufficient remaining capacity is selected first to undertake the incremental compensation task, so as to ensure the speed of harmonic suppression and the optimal capacity utilization. The system acquires the virtual impedance parameters of adjacent distributed power sources, performs impedance matching and collaborative calibration based on these parameters, ensures that the output impedance of multiple distributed power sources in parallel compensation meets the stability boundary conditions with the grid-side impedance, eliminates resonance risk and circulating current risk, and updates the hierarchical control instruction set.

[0031] In this embodiment, the constructed multi-objective collaborative optimization strategy, by clarifying the compensation sensing range and communication interaction relationship of distributed power sources, generates local initial compensation parameters in combination with the remaining capacity of local inverters, and then adjusts the compensation capacity through peer-to-peer communication interaction, achieves the supply and demand balance of compensation capacity within the sensing range; by preset the compensation capacity adjustment step size threshold, unnecessary compensation actions are reduced, improving compensation efficiency; for cases where harmonic distortion rate exceeds the standard, distributed power sources with high comprehensive spatiotemporal coupling and sufficient remaining capacity are prioritized to undertake incremental compensation tasks, balancing the speed of harmonic suppression with optimal capacity utilization; at the same time, through the collaborative calibration of virtual impedance of adjacent distributed power sources, impedance matching during parallel compensation is ensured, effectively eliminating resonance and circulating current risks, ensuring the stability of the harmonic compensation process, and the finally generated hierarchical control instruction set can accurately adapt to the compensation tasks of each distributed power source, improving the accuracy, stability, and efficiency of distribution network harmonic compensation.

[0032] To better demonstrate the implementation of harmonic control in distribution networks based on distributed generation, this invention provides a harmonic control system for distribution networks based on distributed generation, comprising: The harmonic sensing modeling module is used to acquire electrical sampling data of all nodes in the entire distribution network based on the obtained distribution network topology parameters, rated parameters of distributed power grid-connected units, and nonlinear load access information, through the synchronous signal acquisition unit, to extract harmonic features and generate a directed graph of the entire distribution network topology and a set of harmonic feature parameters. The harmonic grouping and quantization module is used to calculate the spatiotemporal coupling degree of harmonic components based on the directed graph of the distribution network topology and the set of harmonic characteristic parameters. Based on the calculation results, it performs harmonic source grouping processing, quantifies the harmonic emission responsibility of each node, and generates harmonic compensation requirement parameters. The hierarchical instruction distribution module is used to construct a hierarchical classification and collaborative allocation strategy for harmonics based on the harmonic compensation requirement parameters generated by the harmonic grouping and quantization module, generate a hierarchical control instruction set with stable boundary constraints, and distribute it to the corresponding execution unit. The local admittance control module is used to obtain the real-time harmonic impedance characteristics of the local grid connection point based on the issued hierarchical control instruction set, construct a harmonic phase correction model and perform correction and control execution, and output harmonic compensation current.

[0033] In this embodiment, the local admittance control module is further configured to: Based on the issued hierarchical control instruction set, real-time voltage and current sampling data of the local grid connection point are obtained, and the real-time harmonic impedance characteristics and background harmonic components of the grid side of the local grid connection point are extracted. Based on the real-time harmonic impedance characteristics and the reference impedance parameters in the hierarchical control instruction set, the target harmonic phase offset caused by grid impedance fluctuations is calculated; the virtual admittance phase angle parameters in the hierarchical control instruction set are corrected based on the target harmonic phase offset to generate the corrected virtual admittance phase angle, ensuring that the local output compensation current is opposite to the target harmonic phase. Based on the preset full-band stability boundary conditions, the virtual admittance amplitude parameters in the hierarchical control instruction set are dynamically limited and corrected to generate the corrected virtual admittance amplitude, ensuring that the output impedance of the distributed power grid-connected inverter and the grid-side input impedance meet the amplitude and phase frequency requirements without resonance risk in the full compensation frequency band. Based on the corrected virtual admittance phase angle and the corrected virtual admittance amplitude, combined with the background harmonic components of the power grid side, the final virtual admittance control parameters are generated. The control loop of the distributed power grid-connected inverter is adjusted based on the final virtual admittance control parameters to generate a total current reference command containing the fundamental active component and harmonic compensation component, which drives the inverter to output the corresponding grid-connected current, thereby realizing local suppression and coordinated compensation of harmonics.

[0034] In the above embodiments, the local admittance control module can accurately capture the dynamic changes in the grid operating status by collecting local grid-connected electrical data in real time and extracting harmonic impedance characteristics and grid-side background harmonic components. By calculating the phase offset caused by grid impedance fluctuations, the virtual admittance phase angle parameter is corrected to ensure that the compensation current is opposite in phase to the target harmonic, thus improving the accuracy of harmonic compensation. Based on the full-band stability boundary conditions, the virtual admittance amplitude is dynamically limited and corrected to ensure that the inverter output impedance matches the grid-side input impedance, effectively avoiding the resonance risk in the full compensation frequency band. The control parameters are generated by combining the corrected virtual admittance parameters and background harmonic components to adjust the inverter control loop and output a grid-connected current containing fundamental and harmonic compensation components, realizing local suppression and coordinated compensation of harmonics. This further improves the harmonic control closed loop, enhances the real-time performance, adaptability, and reliability of distribution network harmonic control, and ensures the continuous and stable harmonic control effect.

[0035] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for harmonic control in distribution networks based on distributed generation, characterized in that, Includes the following steps: Acquire basic parameter data of the distribution network, obtain electrical sampling data of each node in the entire distribution network based on the basic parameter data of the distribution network, extract harmonic features of the electrical sampling data, and generate a set of harmonic feature parameters and a directed graph of the topology of the entire distribution network. Based on the directed graph of the distribution network topology and the set of harmonic characteristic parameters, the spatiotemporal coupling degree of harmonic components is calculated and the harmonic source is grouped. The harmonic emission responsibility of each node is quantified, and the harmonic compensation requirement parameters are generated. Based on the harmonic compensation requirement parameters, harmonic hierarchical classification processing is performed, and a multi-objective collaborative optimization strategy is constructed to generate a hierarchical control instruction set, which is then distributed to the corresponding local execution unit. The local grid-connected point's real-time harmonic impedance characteristics are obtained based on the issued hierarchical control instruction set. A harmonic phase correction model is constructed based on the real-time harmonic impedance characteristics. Correction and control are performed based on the harmonic phase correction model, and harmonic compensation current is output. After collecting voltage and current signals from all nodes in the entire distribution network after harmonic compensation, the system performs harmonic control effect verification and resonance risk prevention based on the electrical sampling data of the entire distribution network after harmonic compensation. Finally, the system generates control strategy parameters based on the harmonic control effect verification results and resonance risk coefficient.

2. The distribution network harmonic control method based on distributed power sources as described in claim 1, characterized in that, The process of constructing the global topology directed graph of the distribution network specifically includes: Based on the basic parameter data of the distribution network, the physical nodes of the entire distribution network are traversed, the coverage area of ​​each node is defined, and a unique identity is assigned to each node. The physical nodes of the entire distribution network are classified according to the type of equipment and electrical function of each node. Based on the classification results, the rated parameters of the distributed power grid-connected unit and the access information of nonlinear loads are bound to the corresponding nodes. The electrical sampling data and the harmonic characteristic parameter set are mapped to each node to form a node set of the entire distribution network. Based on the matching of the topological location information of each node in the distribution network topology and electrical connection relationship, an initial branch set corresponding one-to-one with the physical feeders and electrical connections of the distribution network is constructed, the dominant direction of harmonic propagation of the branch is determined, and the initial branch set is mapped to an initial directed edge set with directional attributes. The harmonic transfer weights of each directed edge in the initial directed edge set are calculated based on the harmonic characteristic parameter set, and the directed edge set is then quantized and calibrated. Based on the global node set of the distribution network and the directed edge set after quantization and calibration, an initial global topology directed graph of the distribution network is constructed, and the global node connectivity of the initial topology directed graph is checked and structural anomalies are corrected. Based on the corrected global topological directed graph of the distribution network, a harmonic transmission correlation matrix is ​​constructed. Combined with the global node set of the distribution network and the quantized and calibrated directed edge set, the standardized encapsulation of the topological directed graph is completed, generating the final global topological directed graph of the distribution network.

3. The distribution network harmonic control method based on distributed power sources as described in claim 2, characterized in that, Each branch in the initial branch connection set corresponds to the distribution network feeder line and physical electrical connection relationship. The two ends of the branch correspond to two associated nodes in the global node set. Based on the fundamental power flow direction, phase difference and amplitude attenuation characteristics of the same frequency harmonics of the two associated nodes at the two ends of the branch, the dominant direction of harmonic propagation of the branch is determined, and the branch is mapped as a directed edge. The fundamental impedance parameter and harmonic impedance parameters of the corresponding branch are bound to each directed edge to form the initial directed edge set.

4. The distribution network harmonic control method based on distributed power sources as described in claim 3, characterized in that, The calculation process for the harmonic transfer weights of each directed edge is as follows: The directed edges in the initial directed edge set are extracted. The starting point of the directed edge in the positive direction is taken as the first node and the ending point in the positive direction is taken as the last node. Based on the harmonic characteristic parameter set, the amplitude and phase measured parameters of the harmonics and interharmonics of the same frequency at the first and last nodes of the corresponding branches of the directed edge are extracted to determine the calculation benchmark of each frequency harmonic of the directed edge. Based on the established harmonic calculation benchmarks for each frequency, the harmonic amplitude ratio and phase difference change rate of the end node and the beginning node of the directed edge at the same frequency are calculated to obtain the initial harmonic transmission coefficient at that frequency. Based on the fundamental impedance parameters and harmonic impedance parameters of the corresponding branches bound by the directed edge, and combined with the branch's line type, physical length, and impedance frequency characteristics, the initial harmonic transmission coefficients of each frequency are corrected to obtain the harmonic transmission weight of the directed edge at the corresponding frequency. The harmonic transfer weights of all frequency harmonics and interharmonics corresponding to the directed edges are combined in ascending order of frequency to form the harmonic transfer weight vector of the directed edges, thus completing the quantization and calibration of the harmonic transfer weights of all directed edges in the initial directed edge set.

5. The distribution network harmonic control method based on distributed power sources as described in claim 1, characterized in that, The calculation process for the spatiotemporal coupling degree of the harmonic components is as follows: Based on the global topological directed graph of the distribution network and the set of harmonic characteristic parameters, all node pairs in the topological directed graph are traversed. Extract the amplitude and phase time series data of all frequency harmonics and interharmonics corresponding to each node within the preset sliding time window, and determine the calculation frequency interval; Based on the extracted harmonic amplitude time series data of each node, the temporal synchronization degree of the node pair in the same frequency harmonic amplitude change within the sliding time window is calculated, and the temporal correlation data of the dynamic change of harmonic amplitude of the node pair at the corresponding frequency is obtained. Based on the harmonic propagation path between nodes in the directed graph of the entire distribution network topology, the theoretical cumulative phase difference of all directed edges on the harmonic propagation path is calculated. Combined with the measured phase difference of the nodes for the same frequency harmonic, the degree of matching between the harmonic phase propagation characteristics and the inherent propagation law of the topology is calculated. Based on the degree of matching between time correlation data, harmonic phase propagation characteristics and the inherent propagation law of the topology, the spatiotemporal coupling degree of the harmonic components of the node at the corresponding frequency is obtained.

6. The distribution network harmonic control method based on distributed power sources as described in claim 5, characterized in that, The harmonic source grouping process includes: Based on the spatiotemporal coupling degree of harmonic components of all node pairs at all frequencies, the comprehensive spatiotemporal coupling degree of each node pair is calculated, and the harmonic coupling degree correlation moment of all nodes in the distribution network is constructed. Based on the harmonic coupling degree correlation matrix, all harmonic emitting nodes are initially clustered to obtain multiple initial harmonic source clusters. Based on the harmonic transmission matrix and harmonic transmission weight vector of the directed graph of the global topology of the distribution network, the dominant harmonic transmission path in the initial harmonic source cluster is identified, and the nodes on the dominant harmonic transmission path and the associated feeder nodes of the harmonic emitting nodes are taken as the coverage node range of the initial harmonic source cluster. Based on the harmonic component timing characteristics of each node within the coverage area of ​​the initial harmonic source cluster and the harmonic transmission attenuation characteristics of the corresponding dominant transmission path, the harmonic characteristic matching degree between each node and the reference node of the dominant transmission path is calculated, and free nodes with matching degree lower than the preset matching degree threshold are removed. Based on the directed graph of the entire distribution network topology, the initial harmonic clusters with electrical connections and adjacent coverage areas are identified as adjacent clusters. The comprehensive spatiotemporal coupling degree between adjacent clusters is calculated, and adjacent clusters with comprehensive spatiotemporal coupling degree exceeding the preset merging threshold are merged to generate merged harmonic source clusters, thus completing the harmonic source grouping process.

7. The distribution network harmonic control method based on distributed power sources as described in claim 1, characterized in that, The construction of the multi-objective collaborative optimization strategy includes: Based on harmonic compensation demand parameters, the directed graph of the distribution network topology, and the set of harmonic characteristic parameters, the local harmonic compensation sensing range of each distributed power grid-connected unit is obtained, and the communication interaction relationship between adjacent distributed power nodes is determined. Each distributed power grid-connected unit generates local initial compensation capacity and virtual impedance matching parameters based on locally collected harmonic data, harmonic source compensation requirements within the sensing range, and the remaining available capacity of the local inverter. Adjacent distributed power grid-connected units exchange local initial compensation capacity, virtual impedance parameters and harmonic sensing data through peer-to-peer communication, and calculate the compensation redundancy and compensation gap between adjacent nodes. Each distributed power grid-connected unit adjusts its local initial compensation capacity based on compensation redundancy and compensation gap, performs constraint verification on the adjusted local compensation capacity, and generates local compensation tasks for each distributed power grid-connected unit. A hierarchical control instruction set is generated based on the local compensation task of each distributed power grid-connected unit.

8. The distribution network harmonic control method based on distributed power sources as described in claim 7, characterized in that, The construction of the multi-objective collaborative optimization strategy also includes: The compensation capacity adjustment step size threshold of each distributed power grid-connected unit is preset, the fluctuation amplitude of harmonic compensation demand is obtained, and it is determined whether the fluctuation amplitude is lower than the preset compensation capacity adjustment step size threshold. If the fluctuation range is lower than the preset compensation capacity adjustment step size threshold, the distributed power source will not adjust the local compensation capacity. Acquire harmonic distortion rate data within the sensing range and determine whether the harmonic distortion rate exceeds the preset distortion standard; If the harmonic distortion rate exceeds the preset distortion standard, the comprehensive spatiotemporal coupling degree between each distributed power source and the harmonic source, and the remaining capacity of each distributed power source are obtained. The distributed power source with the highest comprehensive spatiotemporal coupling degree and the most sufficient remaining capacity is selected to undertake the incremental compensation task. Obtain the virtual impedance parameters of adjacent distributed power sources, perform impedance matching and collaborative calibration based on the virtual impedance parameters of adjacent distributed power sources, and update the hierarchical control instruction set.

9. A distribution network harmonic control system based on distributed power sources, applied in the distribution network harmonic control method based on distributed power sources as described in claim 1, characterized in that, include: The harmonic sensing modeling module is used to acquire electrical sampling data of all nodes in the entire distribution network based on the obtained distribution network topology parameters, rated parameters of distributed power grid-connected units, and nonlinear load access information, through the synchronous signal acquisition unit, to extract harmonic features and generate a directed graph of the entire distribution network topology and a set of harmonic feature parameters. The harmonic grouping and quantization module is used to calculate the spatiotemporal coupling degree of harmonic components based on the directed graph of the distribution network topology and the set of harmonic characteristic parameters. Based on the calculation results, it performs harmonic source grouping processing, quantifies the harmonic emission responsibility of each node, and generates harmonic compensation requirement parameters. The hierarchical instruction distribution module is used to construct a hierarchical classification and collaborative allocation strategy for harmonics based on the harmonic compensation requirement parameters generated by the harmonic grouping and quantization module, generate a hierarchical control instruction set with stable boundary constraints, and distribute it to the corresponding execution unit. The local admittance control module is used to obtain the real-time harmonic impedance characteristics of the local grid connection point based on the issued hierarchical control instruction set, construct a harmonic phase correction model and perform correction and control execution, and output harmonic compensation current.

10. The distribution network harmonic control system based on distributed power sources as described in claim 9, characterized in that, The local admittance control module is also used for: Based on the issued hierarchical control instruction set, real-time voltage and current sampling data of the local grid connection point are obtained, and the real-time harmonic impedance characteristics and background harmonic components of the grid side of the local grid connection point are extracted. Based on the real-time harmonic impedance characteristics and the reference impedance parameters in the hierarchical control instruction set, the target harmonic phase offset is calculated. Based on the target harmonic phase offset, the virtual admittance phase angle parameters and virtual admittance amplitude parameters in the hierarchical control instruction set are dynamically limited and corrected to generate the corrected virtual admittance phase angle and corrected virtual admittance amplitude. Based on the corrected virtual admittance phase angle and the corrected virtual admittance amplitude, combined with the background harmonic components of the power grid side, the final virtual admittance control parameters are generated. The control loop of the distributed power grid-connected inverter is adjusted based on the final virtual admittance control parameters to generate a total current reference command, which drives the inverter to output the corresponding grid-connected current.

Images