Network-building converter cooperative fault ride-through and frequency support method

Abstract

The invention provides a cooperative fault ride-through and frequency support method for a following-networking converter, and belongs to the technical field of converter cluster cooperative control. The method comprises the steps of performing modeling and association arrangement on operation state information of a power grid and operation information of a follow-up grid construction converter; based on the operation state information of the power grid after modeling and association arrangement and the operation information of the follow-up network construction converters, a social cooperation network is constructed, a cooperation control instruction is generated according to the social cooperation network, and the cooperation control instruction is used for indicating cooperation control requirements of the follow-up network construction converters; issuing the cooperative control instruction to the corresponding follow-up network construction converter; and each networking converter executes a control action according to the cooperative control instruction to realize cooperative fault ride-through control and frequency support control. According to the method, the decentralized social cooperative network is constructed, so that the network-following converter cluster can be quickly self-organized to form cooperative control when the power grid fails.

Description

Technical Field

[0001] This invention relates to the field of converter cluster collaborative control technology, specifically to a method for collaborative fault ride-through and frequency support of grid-connected converters. Background Technology

[0002] With the increasing penetration of new energy power generation, represented by wind power and photovoltaics, into the power system, power electronic converters, as their main grid connection interface, are replacing traditional synchronous generators on a large scale, becoming a crucial force in power supply. This transformation has profoundly changed the dynamic characteristics of the power system: traditional power systems rely on the inherent rotational inertia, damping characteristics, and strong overcurrent capacity of synchronous generators to maintain stability, while highly electronic new power systems exhibit characteristics such as low inertia, low short-circuit ratio, and fast dynamic response. Against this backdrop, converters mainly adopt two basic control modes: "grid-following" and "grid-connecting." Grid-following converters synchronize with the grid voltage through phase-locked loops, exhibiting constant power source characteristics. However, their output is highly dependent on stable grid voltage support. When severe faults such as short circuits or generator tripping cause instantaneous voltage drops or drastic frequency changes, their output capacity is significantly limited or even disconnected from the grid, making it difficult to actively support the grid. Grid-connecting converters, on the other hand, autonomously establish voltage and frequency references through internal control algorithms, exhibiting voltage source characteristics. They can provide the necessary inertia and short-circuit capacity support for the grid and are considered a key technology for enhancing grid stability. However, typical solutions for achieving coordinated control of converter clusters in existing technologies rely on a centralized decision-making architecture. This architecture requires a central controller to collect global information, perform calculations, and distribute instructions downwards. This results in large system response delays, high requirements for the communication bandwidth of the central node, and the risk of single point of failure, making it difficult to meet the extreme requirements for control speed and reliability under grid fault conditions. Summary of the Invention

[0003] The purpose of this invention is to provide a method for collaborative fault ride-through and frequency support of grid-connected and grid-connected converters. By constructing a decentralized credit cooperation mechanism, grid-connected and grid-connected converter clusters can quickly self-organize to form collaborative control capabilities during grid faults, thereby achieving reliable fault ride-through and rapid frequency support.

[0004] To achieve the above objectives, this invention provides a method for coordinated fault ride-through and frequency support of grid-connected converters, comprising: collecting and preprocessing grid operating status information and operating information of grid-connected converters; modeling and associating the grid operating status information and the operating information of grid-connected converters; constructing a social cooperation network based on the modeled and associated grid operating status information and the operating information of grid-connected converters to characterize the cooperative relationship between different grid-connected converters; generating cooperative control commands according to the social cooperation network, the cooperative control commands indicating the cooperative control requirements of each grid-connected converter; issuing the cooperative control commands to the corresponding grid-connected converters; and each grid-connected converter executing control actions according to the cooperative control commands to achieve coordinated fault ride-through control and frequency support control.

[0005] Optionally, the modeling and association of the power grid operation status information and the operation information of the grid-connected converters includes: each grid-connected converter establishing a periodic communication link with its neighboring grid-connected converters through a local peer-to-peer communication network; in each communication cycle, each grid-connected converter sends a data packet to its corresponding neighboring grid-connected converter, the data packet including at least the real-time operation status, adjustable active power margin, adjustable reactive power margin, and dynamically updated credit score of the grid-connected converter; simultaneously, each grid-connected converter receives and parses data packets of the same type from all its neighboring grid-connected converters; based on the received and parsed neighboring grid-connected converter data, each grid-connected converter locally and distributedly constructs and updates a neighbor information table, the neighbor information table recording at least the device identity information, real-time operation status, power margin, and credit score of all communicating neighbors, so as to realize distributed modeling and parameter association of the local network environment to which the grid-connected converter belongs.

[0006] Optionally, the construction of the social collaboration network includes: assigning a uniform initial credit score to each network converter when the network converter cluster starts collaborative operation or when a new network converter joins the network converter cluster; each converter establishes periodic data interaction with neighboring nodes through a local communication network, and the interaction information includes real-time operating parameters, power regulation margin, and credit score; each converter dynamically maintains and updates its credit score based on the neighbor node data obtained from the interaction and its performance record in historical collaborative events, using a credit update algorithm that includes a forgetting factor.

[0007] Optionally, the step of dynamically maintaining and updating the credit score using a credit update algorithm that includes a forgetting factor includes: calculating the attenuation of the converter's historical credit score using a preset forgetting factor to obtain the attenuated historical credit value; obtaining the converter's performance score on multiple preset evaluation dimensions in the latest collaborative event; weighting the performance score of each dimension with a pre-configured weight coefficient for each dimension, and summing all weighted results to obtain the comprehensive performance score for the current event; and adding the comprehensive performance score to the attenuated historical credit value to obtain the updated dynamic credit score.

[0008] Optionally, the social collaboration network quantifies the collaborative relationship between different grid-connected converters by multiplying the dynamic credit score of each grid-connected converter by its current active power reserve capacity to obtain a comprehensive contribution potential index.

[0009] Optionally, generating collaborative control instructions based on the social collaboration network includes: when a power grid failure occurs, each of the grid-connected converters calculates the leadership score corresponding to all nodes within its communication range based on the information carried by the social collaboration network; each of the grid-connected converters determines the node with the highest leadership score as a temporary leader node through a distributed consensus algorithm; and generates the collaborative control instructions based on the temporary leader node.

[0010] Optionally, the calculation of leadership scores for all nodes within the communication range includes: obtaining the dynamic credit score, current reserve capacity, local fault severity index, and network electrical centrality index for each of the network converters within the communication range; using the dynamic credit score, current reserve capacity, and network electrical centrality index as positive influencing factors, and the local fault severity index as a negative influencing factor; and performing a weighted comprehensive calculation on each influencing factor to generate a leadership score for each network converter.

[0011] Optionally, the step of sending the collaborative control command to the corresponding grid converter includes: the temporary leader node generating a collaborative control command based on the role allocation result of each grid converter in the social collaboration network, the collaborative control command including a specific power reference value and a corresponding control mode; sending each collaborative control command to the corresponding grid converter through a local communication network using multicast or directed unicast communication methods; the temporary leader node monitoring the sending status of the collaborative control command, and if it does not receive a confirmation signal from the corresponding grid converter within a preset time, it resends the same collaborative control command to the grid converter that did not return a confirmation signal.

[0012] Optionally, each grid-connected converter performs control actions according to the coordinated control command, including: each grid-connected converter parses the received coordinated control command and obtains the assigned control mode and power reference value; if the assigned control mode in the coordinated control command is grid-connected control, the converter switches to the grid-connected control mode to stabilize voltage and frequency; if the assigned control mode in the coordinated control command is grid-connected control, the converter enters the power tracking mode to output the power reference value.

[0013] On the other hand, the present invention provides a grid-connected converter collaborative fault ride-through and frequency support system for realizing a grid-connected converter collaborative fault ride-through and frequency support method. The system includes a control module, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to realize the grid-connected converter collaborative fault ride-through and frequency support method.

[0014] The aforementioned technical solution, by constructing a decentralized social collaboration network, enables the grid-connected converter cluster to quickly self-organize and form collaborative control during grid faults. By using a distributed architecture to avoid single points of failure and improve reliability, it combines local decision-making with global coordination to complete fault ride-through and frequency support in a short time, thereby enhancing grid resilience.

[0015] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0016] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the following detailed description to explain the invention, but do not constitute a limitation thereof. In the drawings:

[0017] Figure 1 This is a flowchart of the method for fault ride-through and frequency support in conjunction with the grid converter.

[0018] Figure 2 This is a flowchart of the election and collaborative startup process for temporary leader nodes based on a social collaboration network. Detailed Implementation

[0019] The following is in conjunction with the appendix Figure 1 -Appendix Figure 2 The specific implementation methods of the embodiments of the present invention will be described in detail below. It should be understood that the specific implementation methods described herein are only for illustrating and explaining the embodiments of the present invention, and are not intended to limit the embodiments of the present invention.

[0020] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with the relevant provisions of national laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0021] In the process of realizing this invention, the inventors of this application discovered that the prior art relies on a centralized decision-making architecture, which results in large system response delays, high communication bandwidth requirements, and the risk of single point of failure, and cannot meet the dual high requirements for control speed and reliability when the power grid fails.

[0022] Example 1

[0023] Reference Figures 1-2 This is the first embodiment of the present invention, which provides a method for coordinated fault ride-through and frequency support of a grid-connected converter, comprising:

[0024] S100: Collects and preprocesses power grid operation status information and grid-connected converter operation information, and models and correlates the power grid operation status information and grid-connected converter operation information.

[0025] In the embodiments of this application, the operation status information of the power grid and the operation information of the grid-connected converters (i.e., converters that support both grid-connected and grid-connected control modes, hereinafter referred to as "grid-connected converters") are modeled and correlated. This includes: each grid-connected converter establishes a periodic communication interaction link with its neighboring grid-connected converters that have a direct electrical connection through a local peer-to-peer communication network; in each communication cycle, each grid-connected converter sends a data packet to its corresponding neighboring grid-connected converter, the data packet including at least the real-time operation information of the grid-connected converter itself. The system includes status, adjustable active power margin, adjustable reactive power margin, and dynamically updated credit score. Simultaneously, each network converter receives and parses data packets of the same type from all its neighboring network converters. Based on the received and parsed neighbor network converter data, each network converter locally builds and updates a neighbor information table. This neighbor information table records at least the device identity information, real-time operating status, power margin, and credit score of all communicating neighbors, enabling distributed modeling and parameter association of the local network environment to which the network converter belongs.

[0026] In a preferred embodiment of this application, all converters are initially assigned a base credit score of 80 points (out of 100) upon startup. Newly connected converters initially receive a transitional score of 70 points, which is updated according to normal rules after 30 communication cycles, thus preventing imbalances in collaboration between new and old nodes. The initial scoring weights prioritize stability, followed by accuracy and a balance between speed and strength, allocated as follows: response speed 30%, support strength 30%, control precision 20%, and collaboration compliance 20%. Preferably, in different fault scenarios such as short circuits and tripping, the upper-level dispatcher can adjust the corresponding dimension weights according to preset conditions to adapt to the collaboration requirements of different faults. The local fault severity index is evaluated using a unified standard of 0-1 based on the depth of voltage drop and the magnitude of frequency deviation (0 for no fault, closer to 1 for more severe faults). Network centrality is determined based on the tightness of the electrical connection between the converter and its neighbors; the more core the connection and the wider the impact, the higher the centrality, ensuring that the evaluation of key indicators is uniform and comparable.

[0027] First, each converter collects core operating data in real time from the grid-connected point, including voltage, frequency, output power, adjustable active and reactive power reserve capacity, and equipment temperature. The collected data is preprocessed to remove instantaneous spikes or obvious errors in voltage and frequency data, retaining only valid data. Then, the converter's own data is compared with historical normal data and relevant data from neighboring converters for consistency verification. Data with excessive deviations (real-time voltage deviation from historical data > ±2% of rated value, or deviation from neighboring data > ±1.5% of rated value; frequency deviation from historical data > ±0.05Hz, or deviation from neighboring data > ±0.03Hz; power deviation from historical data > ±5% of rated value, or deviation from neighboring data > ±3% of rated value; meeting any one of these conditions is sufficient) is marked as "data to be verified." If a node fails consistency verification three times consecutively, or if the reported reserve capacity exceeds its rated capacity, it is marked as a "low-reliability node," thus ensuring the reliability of data transmission and use.

[0028] Furthermore, each converter prioritizes establishing a primary communication link with 3-5 directly electrically connected neighbors, while additionally connecting 1-2 nearby nodes as backup links (e.g., using a wireless private network to back up the wired link), thus building a stable local peer-to-peer communication network. Then, periodic checks are performed by sending a heartbeat packet every 50 milliseconds. If no heartbeat response is received from a neighbor node for three consecutive times, the primary link is considered faulty, and the system immediately switches to the backup link. Simultaneously, communication latency and packet loss rate are monitored in real time. If these metrics exceed preset thresholds, the link is reconnected or switched, and the communication reliability of the neighbor node is recorded. This comprehensive approach ensures stable communication links and prevents communication failures from affecting collaboration.

[0029] In each communication cycle (100 milliseconds under normal conditions, shortened to 20 milliseconds during a fault), each converter sends a data packet to its neighbor containing the device's unique identifier, real-time operating status, standby capacity, current credit score, communication reliability, and a data checksum. Upon receiving the data packet, the checksum is first verified. If the data is incorrect, the sender is immediately requested to retransmit (up to 2 times). Then, the validity of the timestamp is verified. If the timestamp deviates from the local time by more than 50 milliseconds, it is considered invalid data and discarded. At the same time, communication anomalies of the sending node are recorded to ensure accurate data transmission.

[0030] Preferably, the communication reliability score adopts a percentage system. Within the scoring period that is consistent with the data interaction period, the scores of the four dimensions, namely link connectivity (40% weight), communication delay (25% weight), data packet loss rate (25% weight), and data verification pass rate (10% weight), are calculated according to their respective quantification rules and then summed. The final score is ≥80 points and is judged as reliable communication.

[0031] Furthermore, each converter constructs a neighbor information table containing neighbor device ID, real-time operating status, reserve capacity, dynamic credit score, local fault severity index, network centrality, communication reliability, and trustworthiness identifier (high / medium / low), as shown in Table 1. Among them, the real-time operating status, reserve capacity, credit score, and communication reliability data are updated synchronously every communication cycle, and the network centrality is reassessed every 10 communication cycles. If a low-trustworthiness node has normal data for 10 consecutive cycles, it can be restored to medium-trustworthiness. Through structured storage and dynamic updates, accurate neighbor node information is provided to support distributed modeling.

[0032] Table 1. Neighbor Information Table for Network Converters

[0033] Field Name Data types Update frequency Neighbor device unique ID String (e.g., CONV-001) Fix during initial connection Electrical connection relationship Enumeration (direct / indirect) Determined upon initial connection, updated upon link change. Real-time voltage at grid connection point Value (kV) Each communication cycle (100ms normally / 20ms in case of failure) Real-time frequency of grid connection point Value (Hz) Each communication cycle Real-time output of active power Value (MW) Each communication cycle Real-time output of reactive power Numerical value (Mvar) Each communication cycle Active reserve capacity Value (MW) Each communication cycle Reactive power reserve capacity Numerical value (Mvar) Each communication cycle Dynamic credit score Value [0-100 points] Each communication cycle Local Fault Severity Index Values ​​[0-1] The value is fixed at 0 when there is no fault, and updated in real time when a fault occurs. Network centrality Level (High / Medium / Low) Reassess every 10 communication cycles Communication reliability score Value [0-100 points] Each communication cycle Credibility Identifier Enumerate (high / medium / low) Real-time updates Data update timestamp Time format (YYYY-MM-DDHH:MM:SS:mmm) Each communication cycle Remark text Update as needed

[0034] Preferably, each converter, based on its neighbor information table, performs multi-dimensional data association and structured integration to complete the modeling and sorting of the local network environment. On the one hand, it verifies the correlation between the voltage and frequency data collected by itself from the grid and the data of the same type from neighboring nodes, and simultaneously binds the power margin of each node with the grid status (such as associating the reactive power support capacity that the corresponding node can output when the grid voltage drops), clarifying the matching relationship between the local grid demand and the converter's support capacity. On the other hand, it establishes a corresponding association of "operating status, support capacity, and credit reputation" by combining the dynamic credit score, communication reliability score, and credibility identifier of the neighbors (see Table 2). It determines the cooperation priority based on the tightness of electrical connection and the credit score (such as nodes with direct electrical connection and a credit score ≥80 are listed as core cooperation objects), forming a structured information set of the local network centered on itself. This set not only includes the identity, status, capability, credit, and other core data of all valid neighbors, but also clarifies the correlation logic between the data, realizing a complete modeling of the grid operating status and converter operating information.

[0035] Table 2 Credibility Identifier Classification Table

[0036] Credibility Identifier Core judgment criteria (100% corresponding to the original solution) high 1. No data anomalies: The collected core data such as voltage / frequency / power are valid after removing spikes and obvious errors; 2. Consistency check passed: The deviation between the data and "its own historical normal data + neighboring data of the same type" is reasonable and there are no three consecutive failures; 3. Reserve capacity complies with regulations: The reported active / reactive reserve capacity does not exceed its own rated capacity; 4. No "data to be verified" records. middle 1. No serious data errors: core data is valid, but there are occasional minor deviations (marked as "data to be verified", which can be corrected later); 2. No low confidence conditions were triggered (no three consecutive verification failures, capacity is compliant); 3. Can be recovered from the low confidence node's "10 consecutive communication cycles of normal data". Low The following conditions must be met: 1. The consistency check fails three times consecutively; 2. The reported standby capacity significantly exceeds its rated capacity. Credibility level Core judgment criteria (100% corresponding to the original solution)

[0037] The aforementioned solution, through periodic heartbeat detection, seamless primary / backup link switching, and a rigorous data verification process, effectively mitigates system reliability degradation caused by communication interruptions or data anomalies. Furthermore, the dynamically updated neighbor information table not only achieves distributed and accurate modeling of the local network environment but also provides crucial criteria for selecting subsequent collaborative partners by introducing dimensions such as "trustworthiness identifiers" and "communication reliability scores."

[0038] S200: Based on the grid operation status information and the operation information of the grid converter after modeling and correlation processing, a social cooperation network is constructed to represent the collaborative relationship between different grid converters.

[0039] In the embodiments of this application, constructing a social collaboration network includes: assigning a uniform initial credit score to each grid converter when the grid converter cluster starts collaborative operation or when a new grid converter joins the grid converter cluster; each converter establishes periodic data interaction with neighboring nodes through a local communication network, and the interaction information includes real-time operating parameters, power regulation margin, and credit score; each converter dynamically maintains and updates its credit score based on the neighbor node data obtained from the interaction and its performance record in historical collaborative events, using a credit update algorithm that includes a forgetting factor.

[0040] In a preferred embodiment of this application, when the network converter cluster starts collaborative operation or a new converter is added, an initial credit score is allocated according to a preset rule. All existing converters are uniformly set to a base credit score of 80 points (out of 100 points). Newly connected converters initially use a transitional score of 70 points, and after running for 30 communication cycles, they are switched to the normal update rule to ensure fair collaboration between new and old nodes. Each converter, based on the stable communication link built by S100, continuously exchanges key information with neighboring nodes according to a communication cycle of 100ms under normal conditions and 20ms under fault conditions. This includes pre-processed real-time operating parameters (output power, voltage, and frequency data), power regulation margin (taken from the neighbor information table), and current dynamic credit score, as well as real-time records of four dimensions of performance: response speed (such as fault detection response time, command execution delay, etc.), support strength (such as standby capacity utilization rate, maximum output value, etc.), control accuracy (such as power tracking error, power fluctuation amplitude, etc.), and cooperation compliance (such as command execution status, data reporting consistency, etc.). The corresponding fields in the neighbor information table are updated synchronously to maintain the dynamic connectivity of the social cooperation network.

[0041] Furthermore, each converter, based on the stable communication link built by S100, continuously exchanges key information with neighboring nodes according to a communication cycle of 100ms under normal conditions and 20ms under fault conditions. This includes pre-processed real-time operating parameters (output power, voltage data, frequency data), power regulation margin (taken from the neighbor information table), current dynamic credit score, and real-time recorded data of four-dimensional performance: response speed, support strength, control accuracy, and cooperative compliance. The corresponding fields in the neighbor information table are updated synchronously to maintain the dynamic connectivity of the social cooperation network. Then, each converter extracts the core information of the four-dimensional performance from local monitoring data, neighbor feedback data, and communication interaction records and quantifies and scores it (each with a maximum score of 10 points). For example, response speed is based on fault detection response time and command execution delay (e.g., response time ≤ 5ms gets 10 points, > 20ms gets 3 points); support strength is based on standby capacity utilization and continuous support duration (e.g., utilization ≥ 90% gets 10 points, < 50% gets 3 points); control accuracy is based on power tracking error and voltage and frequency control deviation (e.g., power tracking error ≤ 3% gets 10 points, > 10% gets 3 points); cooperation compliance is based on command execution compliance rate and data reporting authenticity (full compliance gets 10 points, one violation gets 5 points).

[0042] Furthermore, the credit score is dynamically maintained and updated using a credit update algorithm that incorporates a forgetting factor. This includes: calculating the attenuation of the converter's historical credit score using a preset forgetting factor to obtain the attenuated historical credit value; obtaining the converter's performance score on multiple preset evaluation dimensions in the latest collaborative event; weighting each dimension's performance score with a pre-configured weight coefficient for each dimension, and summing all weighted results to obtain the comprehensive performance score for this event; and adding the comprehensive performance score to the attenuated historical credit value to obtain the updated dynamic credit score.

[0043] In a preferred embodiment of this application, each converter updates its credit score based on historical collaboration records and a four-dimensional quantitative score, according to the following rules: First, the historical credit score is attenuated using a preset forgetting factor γ (slightly less than 1) to emphasize recent performance weight. Then, the four-dimensional quantitative score is weighted and summed with its corresponding weights (initial response speed 30%, support strength 30%, control accuracy 20%, and collaboration compliance 20%) to obtain the current comprehensive performance score. Finally, the attenuated historical credit score is added to the comprehensive performance score to obtain the updated dynamic credit score. Under different fault scenarios such as short circuits and tripping, the upper-level scheduler adjusts the weights of each dimension according to preset conditions to adapt to the scenario requirements.

[0044] The formula for updating credit scores is as follows:

[0045]

[0046] in, This represents the updated dynamic credit score of node i. γ represents the historical credit score of node i before the update, γ represents the forgetting factor, which is a constant slightly less than 1, such as 0.95 or 0.98; k represents the evaluation dimension index. This represents the performance score of node i in the k-th dimension during this collaborative event.

[0047] In a preferred embodiment of this application, the social collaboration network quantifies the collaborative relationship between different grid converters by multiplying the dynamic credit score of each grid converter by its current active power reserve capacity to obtain a comprehensive contribution potential index.

[0048] It should be noted that each converter calculates its comprehensive contribution potential index (i.e., the product of dynamic credit score and current active power reserve capacity) based on the dynamic credit score and active power reserve capacity in the neighbor information table.

[0049] Furthermore, by combining the comprehensive contribution potential, communication reliability score, trustworthiness indicator, and four-dimensional performance score of each node, collaboration priorities are determined, with nodes having a comprehensive contribution potential value ≥25.5× ( Nodes with a rated active capacity of the converter, a communication reliability score ≥80, a trustworthiness rating of "high," and an average four-dimensional performance score ≥8 are designated as core collaborative nodes. Nodes with a dynamic credit score <60, a trustworthiness rating of "low," or an average four-dimensional performance score <5 are designated as backup collaborative nodes. Every 5 communication cycles, based on updated credit scores, communication link status, and quantitative four-dimensional performance data, the node collaboration priority is dynamically adjusted, and the correlation strength between nodes within the social collaboration network is updated synchronously to ensure that the network always reflects the real-time capabilities and reputation of each node, accurately representing the dynamic collaborative relationships between converters.

[0050] The aforementioned scheme, by creating a dynamic credit score system and a comprehensive contribution potential index, concretizes the abstract inter-converter collaborative relationship into a quantifiable social cooperation network, successfully overcoming the limitations of existing technologies that rely solely on instantaneous capacity for decision-making. By calculating comprehensive contribution potential, the optimal collaborative node can be quickly and accurately identified in emergency situations. Crucially, the credit score update algorithm, incorporating a forgetting factor, establishes a long-term positive incentive and negative penalty mechanism, driving each node to continuously optimize its response speed, control accuracy, and command compliance. This achieves a performance leap for the entire cluster from passive control to active collaboration, enhancing the self-organization and adaptive capabilities of the distributed system.

[0051] S300: Generates collaborative control instructions based on the social collaboration network. These instructions are used to indicate the collaborative control requirements of each grid converter.

[0052] In the embodiments of this application, generating collaborative control instructions based on a social collaboration network includes: when a power grid failure occurs, each grid converter calculates the leadership score corresponding to all nodes within its communication range based on the information carried by the social collaboration network; each grid converter determines the node with the highest leadership score as a temporary leader node through a distributed consensus algorithm; and generates collaborative control instructions based on the temporary leader node.

[0053] In a preferred embodiment of this application, each converter determines the grid operating status in real time based on the local fault severity index and grid connection point voltage and frequency monitoring data. When the grid connection point voltage deviation exceeds ±5% of the rated value or the frequency deviation exceeds ±0.2Hz, and the local fault severity index is ≥0.3, a collaborative response mechanism is immediately triggered. Each converter broadcasts the fault trigger signal and its own real-time operating data to neighboring nodes through a local peer-to-peer communication link. Simultaneously, based on the constructed social cooperation network, it selects effective neighboring nodes with a communication reliability score ≥80 and forms a temporary collaborative cluster. The temporary collaborative cluster refers to a temporary collaborative organization composed of the fault trigger node and grid-connected converter nodes with which it has a direct electrical connection and reliable communication status. All member nodes within the cluster are collaborative nodes, and a distributed collaborative decision-making process is initiated.

[0054] In the embodiments of this application, the leadership score corresponding to all nodes within the communication range is calculated, including: obtaining the dynamic credit score, current reserve capacity, local fault severity index, and network electrical centrality index of each grid converter within the communication range; using the dynamic credit score, current reserve capacity, and network electrical centrality index as positive influencing factors, and the local fault severity index as a negative influencing factor; and performing a weighted comprehensive calculation on each influencing factor to generate the leadership score corresponding to each grid converter.

[0055] In a preferred embodiment of this application, each cooperating node calculates its leadership score based on information from all nodes within its communication range, including obtaining the dynamic credit score of each grid converter within the communication range. Current active power reserve capacity (Normalization process) / , The network electrical centrality index (E) is defined as the rated active power capacity, the network electrical centrality index (0-1 range, with higher values ​​for higher connectivity; E ≥ 0.8 is high, 0.5 ≤ E < 0.8 is medium, and E < 0.5 is low), and the local fault severity index (F) is defined as the local fault severity index (0-1 range, a negative factor). The formula for E is as follows:

[0056]

[0057] in, The number of directly connected neighbors, i.e., the total number of neighboring nodes directly electrically connected to the converter, when When ≥4, The score is 1.0; 2≤ When ≤3, The score is 0.75; When =1, The score is 0.25; When =0, The score is 0; Percentage of core neighbors =Number of core neighbors ÷ (Core neighbors refer to direct neighbors with a dynamic credit score ≥ 80 and a communication reliability score ≥ 80); when When ≥50%, Score: 1.0; 30% ≤ When <50%, The score is 0.67; 10% ≤ When <30%, The score was 0.33; when When <10%, The score is 0.

[0058] The formula for calculating F is as follows:

[0059]

[0060] in, Indicates the real-time voltage at the grid connection point. Indicates the rated voltage of the power grid. This represents the real-time frequency of the grid connection point. When calculating F, simply substitute the real-time data; the result will automatically fall within the 0-1 range, and if it exceeds the range, it will be 0 or 1.

[0061] Then, a fusion calculation is performed to determine the leadership score. The calculation formula is as follows:

[0062]

[0063] Leadership (scores are normalized to a 0-1 range, with higher scores indicating stronger leadership at the node).

[0064] Furthermore, within the temporary collaborative cluster, each node determines a temporary leader node through a distributed consensus algorithm, and each collaborative node then submits its calculated leadership score. The node's overall contribution potential value is synchronized to all nodes in the cluster via a 20ms fault communication cycle. Leader node election is then conducted based on this overall contribution potential value, following the priority order:

[0065] 1. Prioritize selection based on comprehensive contribution potential value ≥ 36× ( The optimal high-potential node (within the rated active capacity of the converter) is selected as the candidate leader node;

[0066] 2. If there is no optimal high-potential node, then select 36× >Comprehensive contribution potential value ≥ 25.5 × High-quality, high-potential nodes were selected as candidates;

[0067] 3. Perform a second screening of candidate nodes, selecting nodes with "high" network centrality and an average four-dimensional performance score of ≥8.5.

[0068] Nodes within the cluster vote on the candidate nodes after the second round of screening. The node with a vote rate of ≥50% is elected as the temporary leader node. If multiple nodes meet the criteria, the node with the highest leadership score is selected.

[0069] The temporary leader node calculates the total active power deficit of the local power grid based on the grid connection point voltage, frequency data, and active power reserve capacity of each node in the neighbor information table; then, according to the defined cooperation priority, it allocates power support tasks according to the proportion of comprehensive contribution potential value. The specific allocation rules are as follows:

[0070] 1. Core collaborative nodes (comprehensive contribution potential ≥ 25.5 × PN, communication reliability score ≥ 80 points, and credibility "high", i.e., credibility is marked as high and four-dimensional performance ≥ 8 points) shall undertake 70% of the power deficit task, and the power value allocated to a single core node shall not exceed 90% of its own active power reserve capacity;

[0071] 2. The backup collaborative node undertakes the remaining 30% of the power deficit task, and the replacement mechanism is only activated when the core node's output is insufficient;

[0072] 3. The temporary leader node sends the task allocation instruction (i.e., the collaborative control instruction) to each collaborative node. The instruction content includes the target output value, control mode and execution time limit. The communication cycle is 20ms under fault conditions.

[0073] The aforementioned scheme can quickly elect a temporary leader node when a power grid failure occurs, using a leadership scoring system based on multi-factor weighted calculation and a distributed consensus algorithm. This effectively avoids the single-point-of-failure risk of traditional centralized control and the low decision-making efficiency that may result from fully distributed negotiation. The temporary leader node scientifically allocates power support tasks according to preset, transparent priority rules, ensuring that the generated collaborative control commands are not only responsive but also have a reasonable task allocation scheme.

[0074] S400: Sends coordinated control commands to the corresponding grid converters.

[0075] The process of issuing coordinated control commands to the corresponding grid converters includes: the temporary leader node generating coordinated control commands based on the role allocation results of each grid converter in the social cooperation network, the coordinated control commands including specific power reference values ​​and corresponding control modes; issuing each coordinated control command to the corresponding grid converter through the local communication network using multicast or directed unicast communication methods; the temporary leader node monitoring the issuance status of the coordinated control commands, and if it does not receive an acknowledgment signal from the corresponding grid converter within a preset time, it resends the same coordinated control command to the grid converter that did not return an acknowledgment signal.

[0076] In a preferred embodiment of this application, the temporary leader node first combines the role allocation results of each node (core cooperating node / standby cooperating node) to break down the unified collaborative control command into dedicated command packets. The command packet of the core cooperating node includes the target output value (≤90% of its own active power reserve capacity), control mode (voltage and frequency control for network type, power control for follow-up network type), and execution time limit; the command packet of the standby cooperating node specifies the replacement trigger condition and the target output value. Subsequently, based on the node communication addresses in the neighbor information table, a one-to-one mapping between command packets and target nodes is completed to ensure that commands can be transmitted in a targeted manner. Relying on the local peer-to-peer communication link, the temporary leader node uses a differentiated strategy of "targeted unicast + multicast" to issue commands. Targeted unicast is used for core cooperating nodes to ensure transmission reliability, and multicast is used for standby cooperating nodes to improve efficiency, maintaining a 20ms communication cycle throughout the process in the event of a fault. All commands are appended with a CRC (Cyclic Redundancy Check) check code to prevent loss or tampering during transmission and ensure that commands safely arrive at the target node.

[0077] Furthermore, upon receiving the aforementioned command, the temporary leader node immediately initiates status monitoring, with a preset confirmation timeout of 40ms (two fault communication cycles). Each network converter must return a confirmation signal containing the unique ID of the neighboring device, a data update timestamp, and its local execution readiness status within 20ms of receiving the command. The temporary leader node temporarily stores the feedback status in its local cache (without writing it to the neighbor information table) and synchronously verifies the feedback node ID against the bound node list to ensure no omissions.

[0078] If no acknowledgment signal is received from a node within 40ms, the temporary leader node determines that its communication is abnormal and immediately performs two operations: First, the instruction is resent, with a maximum of 2 retries, and the "Remarks" field of the node's Table 1 is rewritten to record "Number of instruction resentment: X, Time: XXX"; Second, replacement and score adjustment: if the retries are unsuccessful, a node with "sufficient active backup capacity" and "high credibility" is selected from the backup collaborating nodes to replace the node, and the data in the abnormal node's Table 1 is updated, including a communication reliability score of -5, and a dynamic credit score calculated iteratively (deducting points from the "collaboration compliance" dimension; 3 points are deducted for the first instruction timeout without acknowledgment, 8 points are deducted for two resentments without acknowledgment, and 10 points are deducted for failure to execute the instruction as required after receiving it), and the "data update timestamp" is refreshed synchronously after the update.

[0079] Preferably, when all target nodes return confirmation signals, or when an abnormal node completes its replacement, the temporary leader node completes the determination instruction and updates the neighbor information table simultaneously.

[0080] The above scheme, by employing a differentiated communication strategy combining directional unicast and multicast, and adding a cyclic redundancy check code, significantly improves the accuracy, reliability, and efficiency of control command transmission. The established acknowledgment timeout and limited retransmission mechanisms provide effective protection against transient network disturbances. In particular, when an abnormal situation occurs where a command is not acknowledged, the grid-connected converter collaborative control system can automatically trigger a backup mechanism based on standby collaborative nodes and degrade the abnormal node, enhancing the task success rate and overall reliability of the grid-connected converter collaborative control system under fault conditions.

[0081] S500: Each grid converter executes control actions according to the coordinated control command to achieve coordinated fault ride-through control and frequency support control.

[0082] In the embodiments of this application, each grid-connected converter executes control actions according to the cooperative control command, including: each grid-connected converter parses the received cooperative control command and obtains the assigned control mode and power reference value; if the assigned control mode in the cooperative control command is grid-connected control, the converter switches to the grid-connected control mode to stabilize voltage and frequency; if the assigned control mode in the cooperative control command is grid-connected control, the converter enters the power tracking mode to output the power reference value.

[0083] In a preferred embodiment of this application, after receiving the issued collaborative control command, each grid converter first performs a complete parsing and validity verification of the command, extracting the control mode, power reference value, and execution time limit, and simultaneously confirming the command's directional matching by associating it with its own "neighbor device unique ID". Next, it retrieves the active power reserve capacity, reactive power reserve capacity, and dynamic credit score fields from the neighbor information table to verify whether the power reference value allocated by the command is within its own 90% reserve capacity limit. Then, combining the data update timestamp and CRC check result attached to the command, it eliminates expired, tampered, or mismatched abnormal commands. If the CRC check fails, the converter immediately reports the abnormality to the temporary leader node, synchronously updates the neighbor information table's remarks field to record the cause of the fault, and maintains normal operation mode while awaiting correction commands; if the CRC check passes, it enters the control mode directional switching stage.

[0084] Furthermore, based on the analyzed control mode, the converter completes targeted mode switching and regulation. If the command is assigned to the grid-connected control mode, the converter immediately exits the normal operation mode and switches to the grid-connected control mode. With the core objective of stabilizing the local grid voltage and frequency, it collects real-time voltage and frequency data from the grid connection point in Table 1 according to a 20ms communication cycle under fault conditions. Based on the reference values ​​of ±2% of rated voltage and ±0.2Hz of frequency, it adjusts the output parameters in real time. When the voltage or frequency deviation exceeds the threshold, it automatically increases the regulation intensity to ensure that the voltage is stable within ±3% of the rated value and the frequency is stable within ±0.1Hz of the rated value, achieving the fault ride-through control requirements. After the switch is completed, it synchronously updates the neighbor information table data update timestamp and reports the successful mode switch and regulation status to the temporary leader node.

[0085] If the instruction is assigned to grid-following control mode, the converter switches to power tracking mode, focusing on accurately outputting the instruction power reference value. It associates the real-time output active power, reactive power, and active reserve capacity fields in the neighbor information table, adjusts the output at 20ms intervals, controls the deviation between the actual output active power and the reference value to be ≤±3%, and compensates reactive power as needed to avoid oscillation. After the switch is completed, it also updates the data update timestamp of the neighbor information table and feeds back the operating status.

[0086] Furthermore, in the dynamic control operations conducted by each grid-connected converter based on the parsed collaborative control commands and combined with real-time operating data from the grid connection point, the converters maintain dynamic control and data linkage at all times. Every 20ms, core field data from the neighbor information table is collected, including grid connection point voltage and frequency, output power, and equipment temperature. Based on this, control parameters are fine-tuned to optimize operating performance. In grid-connected mode, voltage and frequency stability is enhanced; in grid-following mode, power tracking deviation is corrected. Simultaneously, real-time operating status, parameter values, and deviations are uploaded to the temporary leader node. If execution is normal, the dynamic credit score and communication reliability score are each slightly increased by 1 point.

[0087] Preferably, boundary protection is carried out simultaneously, and the equipment temperature and output limit are monitored in real time. Once the safety threshold is exceeded, the output is immediately reduced and feedback is sent to the temporary leader node to trigger the backup node replacement mechanism. At the same time, the abnormal details are recorded in the remarks field of the neighbor information table.

[0088] Furthermore, if a control mode switching failure occurs, such as the grid-connected mode failing to stabilize voltage or the grid-following mode having a power deviation exceeding ±10%, the converter will immediately switch to the backup protection mode to maintain basic output and prevent equipment damage or escalation of grid disturbances. Within one communication cycle, the reason for the switching failure will be reported to the temporary leadership node. Upon receiving the report, the temporary leadership node will deduct 5 points from the node's dynamic credit score in its neighbor information table and 3 points from its communication reliability score. The credit score will be iteratively updated using a calculation formula, simultaneously refreshing the data update timestamp and remarks fields to ensure that the score accurately reflects the node's execution capabilities.

[0089] In a preferred embodiment of this application, every 5 fault communication cycles (100ms), the converter and the temporary leader node collaboratively verify the control effect. In the grid-connected mode, the focus is on verifying the voltage and frequency stability accuracy, while in the grid-following mode, the focus is on power tracking deviation, confirming whether the fault ride-through and frequency support targets have been achieved. If there is a small deviation, the local control parameters are optimized based on real-time data from the neighbor information table. If the deviation is too large (e.g., voltage deviation exceeding ±3% in the grid-connected mode, power deviation exceeding ±5% in the grid-following mode), a dynamic rescheduling adjustment command parameter is triggered. The temporary leader node uses the task allocation rules set in S300, combined with the active power reserve capacity and comprehensive contribution potential value of each node in the neighbor information table, to adjust the target output allocation. The core cooperating node still undertakes the main tasks according to the proportion of comprehensive contribution potential value, and the output of a single node does not exceed 90% of its own reserve capacity. If necessary, a backup cooperating node is activated to fill the gap. The corrected command is reissued to the corresponding node. After receiving the corrected command, the converter repeats the above parsing, switching, and execution process to ensure that the control effect meets the target, and the verification and rescheduling results are synchronously written into the remarks field of the neighbor information table.

[0090] The above scheme ensures that each converter performs validity verification and capacity matching before executing commands, effectively preventing equipment overload risks caused by command errors or exceeding limits. Furthermore, it precisely switches to "grid-connecting" or "grid-following" control modes according to command requirements, focusing on stabilizing grid voltage and frequency and accurately outputting active and reactive power, directly and efficiently meeting the core control requirements of fault ride-through and frequency support. In addition, by introducing a closed-loop management mechanism that includes dynamic regulation, safety boundary monitoring, and periodic effect verification, and can trigger dynamic rescheduling, a feedback loop is formed, ensuring high-quality achievement of coordinated control objectives and rapid and stable recovery after grid faults.

[0091] The present invention also provides a grid-connected converter collaborative fault ride-through and frequency support system for realizing a grid-connected converter collaborative fault ride-through and frequency support method. The system includes a control module, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to realize the grid-connected converter collaborative fault ride-through and frequency support method.

[0092] This invention provides a storage medium storing a program that, when executed by a processor, implements a method for coordinated fault ride-through and frequency support with grid-connected converters.

[0093] This invention provides a processor for running a program, wherein the program executes a method for coordinated fault ride-through and frequency support of grid converters during runtime.

[0094] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements a method for coordinated fault ride-through and frequency support with a grid-connected converter. The device described herein can be a server, PC, tablet, mobile phone, etc.

[0095] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing a method for coordinated fault ride-through and frequency support with a grid converter.

[0096] Those skilled in the art will understand that embodiments of this application can provide methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0097] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0098] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0099] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0100] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0101] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0102] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0103] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0104] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for coordinated fault ride-through and frequency support with grid-connected converters, characterized in that, include: Collect and preprocess power grid operation status information and grid-connected converter operation information, and model and associate the power grid operation status information and grid-connected converter operation information; Based on the power grid operation status information and the operation information of the grid converter after modeling and correlation processing, a social cooperation network is constructed to characterize the cooperative relationship between different grid converters. The social collaboration network generates collaborative control instructions, which are used to indicate the collaborative control requirements of each grid converter. The coordinated control command is sent to the corresponding grid converter; Each grid converter executes control actions according to the aforementioned coordinated control instructions to achieve coordinated fault ride-through control and frequency support control.

2. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 1, characterized in that, The modeling and correlation organization of the power grid operation status information and the operation information of the grid converter includes: Each grid converter establishes a periodic communication link with its neighboring grid converters that have a direct electrical connection through a local peer-to-peer communication network. During each communication cycle, each grid converter sends a data packet to its corresponding neighboring grid converter. The data packet includes at least the grid converter's real-time operating status, adjustable active power margin, adjustable reactive power margin, and dynamically updated credit score. At the same time, each grid converter receives and parses the same type of data packets from all its neighboring grid converters. Each network converter, based on the received and parsed data from its neighboring network converters, locally builds and updates a neighbor information table. The neighbor information table records at least the device identity information, real-time operating status, power margin, and credit score of all communicating neighbors, so as to realize distributed modeling and parameter association of the local network environment to which the network converter belongs.

3. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 1, characterized in that, The construction of the social collaboration network includes: When the grid-connected converter cluster starts collaborative operation or when a new grid-connected converter joins the grid-connected converter cluster, a uniform initial credit score is assigned to each grid-connected converter. Each converter establishes periodic data interaction with adjacent nodes through a local communication network. The interaction information includes real-time operating parameters, power regulation margin, and credit score. Each converter dynamically maintains and updates its credit score based on neighbor node data obtained through interaction and its performance records in historical collaborative events, using a credit update algorithm that includes a forgetting factor.

4. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 3, characterized in that, The method of dynamically maintaining and updating credit scores through a credit update algorithm that includes a forgetting factor includes: The historical credit score of the converter is attenuated by a preset forgetting factor to obtain the attenuated historical credit value. Obtain the performance scores of the converter across multiple preset evaluation dimensions in the latest collaborative event; The performance scores for each dimension are weighted by the pre-configured weight coefficients for each dimension, and the sum of all weighted results is obtained to obtain the overall performance score for this event. The overall performance score is added to the attenuated historical credit score, and the result is the updated dynamic credit score.

5. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 1, characterized in that, The social collaboration network quantifies the collaborative relationship between different grid-connected converters by multiplying the dynamic credit score of each grid-connected converter by its current active power reserve capacity to obtain a comprehensive contribution potential index.

6. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 1, characterized in that, The step of generating collaborative control instructions based on the social collaboration network includes: When a power grid failure occurs, each of the aforementioned grid converters calculates the leadership score corresponding to all nodes within the communication range based on the information carried by the social collaboration network. Each of the aforementioned grid converters uses a distributed consensus algorithm to determine the node with the highest leadership score as the temporary leader node; The collaborative control instructions are generated based on the temporary leader node.

7. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 6, characterized in that, The leadership scores for all nodes within the computational communication range include: Obtain the dynamic credit score, current standby capacity, local fault severity index, and network electrical centrality index of each of the aforementioned grid converters within the communication range; The dynamic credit score, current reserve capacity, and network electrical centrality index are used as positive influencing factors, while the local fault severity index is used as a negative influencing factor. The leadership score for each of the aforementioned grid converters is generated by merging and calculating the various influencing factors using a weighted comprehensive method.

8. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 1, characterized in that, The step of sending the coordinated control command to the corresponding grid converter includes: Based on the role allocation results of each grid converter in the social collaboration network, the temporary leader node generates collaborative control instructions, which include specific power reference values ​​and corresponding control modes. Through a local communication network, using multicast or directional unicast communication methods, the aforementioned collaborative control commands are respectively sent to the corresponding grid converters. The temporary leader node monitors the issuance status of the collaborative control command. If it does not receive a confirmation signal from the corresponding grid converter within a preset time, it resends the same collaborative control command to the grid converter that did not return a confirmation signal.

9. The method for coordinated fault ride-through and frequency support of grid-connected converters according to claim 1, characterized in that, Each grid converter executes control actions according to the coordinated control command, including: Each grid converter parses the received cooperative control commands and obtains the assigned control mode and power reference value; If the control mode assigned in the coordinated control command is grid-based control, the converter switches to grid-based control mode to stabilize voltage and frequency. If the control mode assigned in the coordinated control command is grid-following control, the converter enters power tracking mode to output the power reference value.

10. A grid-connected converter-coordinated fault ride-through and frequency support system, characterized in that, The system includes a control module, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the grid-connected converter cooperative fault ride-through and frequency support method according to any one of claims 1-9.

Images