An ac-dc hybrid distribution network partition autonomous and mutual aid operation method based on flexible multi-state switch

Abstract

This invention belongs to the field of AC / DC hybrid distribution network operation and control technology. A method for autonomous and mutual-assistance operation of AC / DC hybrid distribution networks based on flexible multi-state switches includes the following steps: Step S1: Calculate the comprehensive stability index of each node; Step S2: Monitor the comprehensive stability index in real time; Step S3: In the autonomous mode, the SOP within each zone adopts an AC / DC power decoupling control strategy based on dynamic virtual impedance; Step S4: When the power deficit of a zone exceeds a preset proportion of the zone's rated load and the duration exceeds a preset time threshold, the mutual-assistance mode is activated; Step S5: When a system fault occurs, a fault recovery strategy based on model predictive control is adopted; Step S6: Monitor the communication status between the global coordination unit and the control units of each zone in real time. This achieves safe, stable, efficient, and economical system operation, reduces communication dependence, and improves power supply reliability.

Description

Technical Field

[0001] This invention belongs to the field of AC / DC hybrid distribution network operation and control technology, specifically involving a method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches (SOPs), which is particularly suitable for dynamic control of AC / DC hybrid distribution networks in scenarios with a high proportion of distributed power sources. Background Technology

[0002] With the large-scale integration of distributed power sources (such as photovoltaic and wind power) and the rapid growth of DC loads (such as electric vehicles and data centers), hybrid AC / DC distribution networks have become an important development direction for future distribution networks due to their ability to combine the maturity of AC systems with the efficiency of DC systems. Flexible multi-state switches (SOPs), as the core flexible interconnection equipment in hybrid AC / DC distribution networks, can achieve flexible conversion of AC and DC power and precise control of power flow, providing technical support for the zoned operation and power balance of distribution networks.

[0003] Currently, the zonal control and SOP (Standard Operating Procedure) strategies of AC / DC hybrid distribution networks still have many technical shortcomings, making it difficult to adapt to the source-load fluctuation characteristics brought about by the high proportion of distributed generation access. Specifically, these shortcomings are as follows: Static partitioning has significant limitations: existing partitioning methods are mostly based on static division of the grid structure, with fixed partition boundaries. This cannot adapt to the randomness and volatility of distributed power output and the dynamic changes in load, making it difficult to maintain power balance within the partition, resulting in low SOP capacity utilization and even problems such as power circulating current.

[0004] Insufficient SOP control and partition coordination: When multiple AC / DC subnets are interconnected by multi-terminal SOPs, port power coupling is severe. The existing static virtual impedance control strategy has fixed parameters and cannot be adaptively adjusted according to the system operating status, making it difficult to achieve effective decoupling of AC / DC power and affecting the stability of system operation.

[0005] The autonomous and mutual assistance coordination mechanism is imperfect: most schemes only achieve simple power mutual assistance, lack a hierarchical coordination mechanism, and the triggering conditions and priorities for mutual assistance are not reasonably designed, failing to take into account both the independence of regional autonomy and the economy and reliability of the system as a whole.

[0006] Low fault recovery efficiency: When a system fails, existing recovery strategies mostly rely on SOP power regulation and are not optimized in conjunction with distributed power sources, energy storage and other equipment, resulting in slow fault isolation speed, delayed power restoration in non-faulty areas and difficulty in ensuring power supply reliability.

[0007] High dependence on communication: Existing centralized control solutions rely on global real-time communication, which not only puts a lot of pressure on communication, but also causes the entire control system to crash if communication is interrupted. At the same time, it cannot protect the data privacy of different operating entities.

[0008] In recent years, some improved solutions have emerged in related fields, such as the "decentralized reactive power optimization method for AC / DC hybrid distribution areas based on Nash equilibrium" proposed by State Grid Hebei Electric Power, as well as several patents on SOP energy management strategies and hybrid switching control. However, these solutions still have obvious shortcomings: either they fail to achieve dynamic adaptive adjustment of zones, or they fail to solve the power coupling problem of multi-terminal SOPs, or they lack a collaborative recovery mechanism under fault scenarios. They are fundamentally different from the technical solution of this invention and cannot solve all the above-mentioned technical pain points. Summary of the Invention

[0009] The purpose of this invention is to provide a method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches. Through dynamic zoning, coordinated control, and hierarchical mutual support, the method achieves safe, stable, efficient, and economical system operation, reduces communication dependence, and improves power supply reliability.

[0010] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: a method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches, characterized by comprising the following steps: Step S1: Construct a three-dimensional dynamic partitioning index system that includes SOP port power fluctuation characteristics, distributed power output uncertainty and power supply reliability, and calculate the comprehensive stability index of each node. Step S2: Monitor the comprehensive stability index in real time. When the comprehensive stability index of any node exceeds the preset first threshold, trigger partition reconstruction. Use the improved Louvain community discovery algorithm to re-divide the partition boundary. The node weight is dynamically adjusted according to the distributed power penetration rate and the SOP (flexible multi-state switch) port capacity. And it is mandatory to constrain different ports of the same SOP to be located in different partitions. Step S3: In the autonomous partition mode, each partition's SOP adopts an AC / DC power decoupling control strategy based on dynamic virtual impedance, which suppresses power circulating current between multi-terminal SOPs by adjusting the dynamic virtual impedance value. Step S4: When the power deficit of a certain zone exceeds the preset proportion of the rated load of that zone and the duration exceeds the preset time threshold, the zone mutual assistance mode is activated. A mutual assistance priority sequence is generated based on the reserve capacity margin and voltage stability index of adjacent zones, and power is allocated between zones through SOP. Step S5: When a system failure occurs, a fault recovery strategy based on model predictive control (MPC) is adopted to optimize the power of the SOP port and the output of the distributed power supply within a preset prediction time window, so as to achieve fault isolation and rapid power restoration of non-faulty areas.

[0011] Step S6: Monitor the communication status between the global coordination unit and each partition control unit in real time. When the communication delay exceeds the preset communication threshold or the communication link is interrupted, the system immediately switches to fully autonomous mode, and each partition independently performs local control. After the communication is restored to stability, it automatically switches back to normal operation mode.

[0012] Furthermore, the comprehensive stability index is obtained by weighted summation of the SOP port power fluctuation coefficient, the distributed power output fluctuation coefficient, and the power supply reliability index, with each weight coefficient preset according to the system operation requirements.

[0013] Furthermore, the first threshold is preset based on the maximum allowable stability deviation of the system, and is used to determine whether the system operating state has changed significantly, thereby triggering partition reconstruction.

[0014] Furthermore, the execution steps of the improved Louvain community detection algorithm include: Initialization: Assign each node in the distribution network to an independent community and calculate the initial modularity; Node weight update: The weight of each node is calculated based on the distributed power penetration rate, node voltage stability, and power supply reliability; Local move: For each node, try to move it to the adjacent community in turn, calculate the modularity increment, select the community corresponding to the largest modularity increment, and if the modularity increment is greater than 0, move the node to that community. Community aggregation: Treat each community as a super node, repeat the above steps of node weight update and local movement until the modularity no longer increases; Partition constraint check: Check if there are multiple ports of the same SOP (Flexible Multistate Switch) located in the same partition. If so, adjust the partition affiliation of the port of that SOP and recalculate the modularity until the constraint conditions are met. Output the final partitioning results.

[0015] Furthermore, corresponding to step S3, the dynamic virtual impedance value is the virtual equivalent impedance connected in series between the voltage outer loop and the current inner loop in the SOP AC / DC power decoupling control loop. Its value increases exponentially with the system running time and eventually converges to a preset steady-state value. The calculation formula for the dynamic virtual impedance is: ,in, This is the value of the dynamic virtual impedance. The preset maximum steady-state virtual impedance, t is the time constant, which is preset according to the dynamic response characteristics of the system; e is the duration of the system entering the current operating condition; and e is the natural constant (approximately 2.71828). When the system starts up or the operating condition changes abruptly, the virtual impedance value is small to ensure the system's rapid response capability. When the system is running in steady state, the virtual impedance value increases to enhance the decoupling effect.

[0016] Furthermore, corresponding to step S4, the power deficit threshold is the ratio threshold of the real-time power deficit within the partition to the rated load of the partition, and the duration threshold is the minimum time threshold for the power deficit to continuously exceed the power deficit threshold. The power deficit threshold is usually set to 5% to 15% of the rated load of the partition, and the duration threshold is usually set to 10s to 30s. Both are preset according to the system power supply reliability level, distributed power penetration rate and load characteristics, which can effectively filter out false triggering caused by short-term fluctuations in photovoltaic and wind power output or instantaneous changes in load, and ensure that the partition mutual assistance mode is only activated when the partition is indeed unable to maintain power balance through its own resources.

[0017] Furthermore, the generation rule for the mutual assistance priority sequence is as follows: adjacent partitions with larger reserve capacity margins and better voltage stability have higher mutual assistance priority, and priority is given to providing power support to partitions with power shortages.

[0018] Furthermore, the prediction time window and control cycle of the model predictive control are preset according to the system response speed requirements. By predicting the system operating status in the future, the power of the SOP port and the output of the distributed power supply are planned in advance to achieve rapid fault recovery.

[0019] Furthermore, corresponding to step S6, the communication delay is the bidirectional data transmission delay between the global coordination unit and each partition control unit. The preset communication threshold is set according to the real-time requirements of system control, typically 50ms to 200ms. When the system detects in real time that the communication delay exceeds the preset communication threshold, or detects a communication link interruption, regardless of whether the system is currently in partition mutual aid mode or fault recovery mode, it will immediately and automatically switch to fully autonomous mode. Each partition will perform independent control based solely on locally collected operating data, and the status information exchange and power mutual aid between partitions will cease. After the communication delay recovers to below the threshold and the communication link has been running stably for more than 30 seconds, the system will automatically switch back to the normal partition autonomous and mutual aid operation mode.

[0020] Compared with the prior art, the present invention has the following advantages: 1. This invention solves the problem that existing static partitioning methods cannot adapt to the dynamic fluctuations of high proportions of distributed power sources: By constructing a three-dimensional dynamic partitioning index system that includes the power fluctuation characteristics of SOP ports, the output uncertainty of distributed power sources, and the reliability of power supply, this invention can reflect changes in the system's operating status in real time. When the system's operating status changes significantly, it automatically triggers partition reconstruction, ensuring that the partition boundaries always match the current source-load distribution and SOP operating characteristics. This avoids the waste of SOP capacity and the decrease in system stability caused by unreasonable partitioning, and significantly improves the rationality of partitioning and the adaptability of the system.

[0021] 2. Solves the problem of severe power coupling and easy circulation current between multi-terminal SOPs: The present invention adopts an AC / DC power decoupling control strategy based on dynamic virtual impedance. The virtual impedance value is adaptively adjusted with the running time, taking into account both the decoupling effect and the dynamic response speed of the system. It effectively suppresses the power circulation current between multi-terminal SOPs, improves the operating efficiency of SOPs and the operating stability of the system, and overcomes the defects of the existing static virtual impedance control parameters being fixed and the decoupling effect being limited.

[0022] 3. This invention solves the problems of slow speed and large voltage fluctuation in existing fault recovery methods: The present invention adopts a fault recovery strategy based on model predictive control, which can simultaneously optimize the power of SOP port and the output of distributed power sources within the prediction time window, plan the fault recovery path in advance, realize the rapid isolation of faults and the rapid power supply restoration of non-faulty areas, effectively shorten the power outage time, reduce voltage fluctuations during the recovery process, and improve the reliability of system power supply.

[0023] 4. This invention solves the problems of high communication dependence and poor data privacy in existing centralized control methods: The invention adopts a hierarchical and autonomous and mutually supportive architecture, in which most control functions are completed locally within the partition, and only a small amount of status information needs to be exchanged between partitions. This significantly reduces the system's dependence on communication, improves the system's ability to resist communication interruptions, and protects the data privacy of different operating entities.

[0024] 5. Overcoming existing technical biases and achieving synergistic technical benefits: This invention overcomes the existing technical bias that "different ports of the same SOP should be located in the same partition". It forces different ports of the same SOP to be located in different partitions, so that the SOP can give full play to its flexible interconnection function. Combined with the synergistic effect of the three core technical features of three-dimensional dynamic partitioning, dynamic virtual impedance decoupling and model prediction fault recovery, it produces unexpected technical effects and significantly improves the overall operation performance of AC / DC hybrid distribution networks. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0026] Figure 1 This is a topology diagram of an AC / DC hybrid distribution network system based on SOP in an embodiment of the present invention.

[0027] Figure 2 This is a flowchart of the improved Louvain community discovery algorithm in this embodiment of the invention.

[0028] Figure 3 This is a block diagram of the SOP dynamic virtual impedance decoupling control in an embodiment of the present invention.

[0029] Figure 4 This is a state machine diagram showing the switching between autonomous and mutual assistance operation modes in this embodiment of the invention. Detailed Implementation

[0030] The following will refer to the accompanying drawings in the embodiments of the present invention ( Figures 1 to 4 The technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0031] Example This embodiment provides a method for SOP collaborative control and adaptive zoning in AC / DC hybrid distribution networks based on multi-dimensional dynamic indicators, applicable to... Figure 1 The AC / DC hybrid distribution network shown here, which includes multiple AC subgrids, DC subgrids, multi-terminal SOPs, distributed power sources, energy storage systems, and loads, is implemented in the following steps: Step S1: Construct a three-dimensional dynamic zoning index system and calculate the comprehensive stability index. Based on the operating characteristics of AC / DC hybrid distribution networks, a three-dimensional dynamic partitioning index system is constructed, which includes the power fluctuation characteristics of SOP ports, the output uncertainty of distributed power sources, and the power supply reliability. The three dimensions of the index are: SOP port power fluctuation coefficient, distributed power source output fluctuation coefficient, and power supply reliability index.

[0032] Among them, the SOP port power fluctuation coefficient is used to reflect the dynamic change of SOP port power, and is calculated based on the deviation between the real-time power and the average power of SOP port; the distributed power output fluctuation coefficient is used to reflect the randomness and volatility of distributed power output such as photovoltaic and wind power, and is calculated based on the deviation between the real-time output and the predicted output of distributed power; the power supply reliability index is used to reflect the power supply quality within the zone, and is characterized by derived indexes of the system average outage related indexes.

[0033] The above three dimensions of indicators are weighted and summed to obtain the comprehensive stability index of each node. Each weight coefficient is preset according to the system operation requirements (such as power supply reliability priority and economic priority) to ensure that the comprehensive stability index can fully and accurately reflect the operating status of the node and the system.

[0034] Step S2: Implement dynamic partition reconstruction based on the improved Louvain algorithm The comprehensive stability index of each node is collected in real time and compared with a preset first threshold. When the comprehensive stability index of any node exceeds the first threshold, it indicates that the system operating status has changed significantly and the original partition boundary is no longer reasonable, triggering the partition reconstruction process.

[0035] An improved Louvain community detection algorithm is used for partition boundary reconstruction. The specific execution process of this algorithm is as follows: Figure 2 As shown, the detailed steps are as follows: Initialization: Each node in the distribution network (including AC subnet nodes, DC subnet nodes, and SOP port nodes) is assigned to an independent community. The initial modularity is calculated. The modularity is used to measure the rationality of the partitioning results. The larger the modularity, the better the partitioning effect.

[0036] Node weight update: The weight of each node is calculated based on the distributed power penetration rate, voltage stability, and power supply reliability of the node. Nodes with higher distributed power penetration rate, better voltage stability, and higher power supply reliability have greater weights, ensuring that the partitioning results can take into account both source and load distribution and power supply quality.

[0037] Local move: For each node, try to move it to the adjacent community in turn, calculate the modularity increment after each move. If the modularity increment is greater than 0, it means that moving the node can improve the rationality of the partition, so move the node to the corresponding community; if the modularity increment is less than or equal to 0, keep the original community affiliation of the node.

[0038] Community aggregation: Treat each community as a super node, repeat the above steps of node weight update and local movement until the modularity no longer increases, at which point the preliminary partitioning result is obtained.

[0039] Partition constraint check: Check if there are multiple ports of the same SOP in the preliminary partitioning results. If so, adjust the partition assignment of the port of that SOP and assign it to different adjacent partitions. Recalculate the modularity until the constraint condition that "different ports of the same SOP must be located in different partitions" is met.

[0040] Output the final partitioning results to complete the partition reconstruction.

[0041] The improved Louvain algorithm described above enables dynamic adaptive adjustment of partition boundaries, ensuring that the partitioning results always match the current operating state of the system.

[0042] Step S3: SOP dynamic virtual impedance decoupling control in partitioned autonomous mode When the overall stability index of each partition is within the preset threshold range, and the power in each partition can balance itself and operate stably, the system enters the partition autonomy mode. The control functions in each partition are completed independently by the partition control unit without the need for global coordination.

[0043] In the partitioned autonomous mode, each partition's SOPs employ an AC / DC power decoupling control strategy based on dynamic virtual impedance to address power coupling issues between multi-terminal SOPs and suppress power circulating current. The control block diagram is shown below. Figure 3 As shown, the value of the dynamic virtual impedance increases exponentially with the system running time, eventually converging to a preset steady-state value. In the initial stage of system startup or during sudden changes in operating conditions, the virtual impedance value is small, ensuring the system's rapid response capability and enabling it to quickly track power changes within the partition, maintaining system voltage and frequency stability. After the system enters steady-state operation, the virtual impedance value gradually increases, enhancing the decoupling effect of AC and DC power, effectively suppressing power circulating current between multiple SOPs, and improving the operating efficiency of SOPs and system stability.

[0044] Depend on Figure 3 As can be seen, the SOP control architecture includes voltage outer loop control and current inner loop control. The dynamic virtual impedance module is connected in series between the voltage outer loop and the current inner loop. According to the system running time and running status, the virtual impedance value is adjusted in real time to achieve power decoupling control.

[0045] Step S4: Power distribution control in zone mutual assistance mode The system operating mode switching logic is as follows: Figure 4 As shown, the power balance status of each partition is monitored in real time. When a partition experiences a power deficit, and the power deficit exceeds the preset proportion of the partition's rated load and the duration exceeds the preset time threshold, it indicates that the partition cannot achieve power balance through its own resources, triggering the partition mutual assistance mode, and the system switches from autonomous mode to mutual assistance mode.

[0046] After the partitioned mutual assistance mode is activated, the global coordination unit collects the reserve capacity margin and voltage stability indicators of each adjacent partition and generates a mutual assistance priority sequence according to preset rules: the adjacent partition with the larger the reserve capacity margin and the better the voltage stability has the higher the mutual assistance priority, and prioritizes providing power support to the power shortage partition.

[0047] Power exchange between zones is achieved through SOP. According to the mutual assistance priority sequence, power is delivered from high-priority zones to power-deficient zones. At the same time, the voltage and power status of each zone are monitored in real time, and the power delivery amount is dynamically adjusted to ensure the stable operation of each zone during the mutual assistance process and avoid system fluctuations caused by power mutual assistance.

[0048] Once the power balance of the power deficit partition returns to normal and the duration reaches the preset stabilization time, the system exits the partition mutual assistance mode and switches back to the partition autonomous mode (e.g., Figure 4 (As shown).

[0049] Step S5: Model Prediction-Based Fast Recovery Control in Fault Scenario When a system failure occurs (such as a line fault, SOP failure, or distributed power supply failure), the system immediately switches to fault recovery mode (such as...). Figure 4 As shown in the figure, a fast recovery strategy based on model predictive control (MPC) is adopted to achieve rapid power restoration in fault isolation and non-faulty areas.

[0050] The core of model predictive control is to predict the system's operating status (including voltage, power, frequency, etc.) within a preset prediction time window based on the system's mathematical model. The optimization goal is to optimize the power of the SOP port and the output of distributed power sources with the aim of "fastest fault isolation speed, most timely power restoration in non-faulty areas, and minimal voltage fluctuations".

[0051] The specific execution process is as follows: After a fault occurs, the fault detection module first locates the fault location, isolates the fault, and cuts off the connection between the faulty area and the non-faulty area. Then, the global coordination unit collects the operating status information of the non-faulty area and inputs it into the model predictive control module. Within a preset prediction time window, the model predictive control module generates multiple control schemes for SOP port power and distributed power output, and calculates the optimization target value of each control scheme. The optimal control scheme is selected and sent to each partition control unit and device to execute the control command. The system operating status is monitored in real time. If the recovery target is not achieved, the above optimization process is repeated until the non-faulty area restores normal power supply.

[0052] Step S6: Monitor the communication status between the global coordination unit and each partition control unit in real time. When the communication delay exceeds the preset communication threshold or the communication link is interrupted, the system immediately switches to fully autonomous mode (e.g., Figure 4 As shown), only local control within the partition is retained, and each partition executes local control independently to avoid communication interruptions affecting system operation. Once communication is restored to a stable state, it will automatically switch back to normal operation mode (re-switching to normal autonomous and mutual assistance mode).

[0053] In this embodiment, the three-dimensional dynamic zoning index system, the dynamic virtual impedance decoupling control strategy, and the model prediction fault recovery strategy are not independent of each other, but are organically combined and work synergistically: the three-dimensional dynamic zoning index system provides a scientific basis for zoning reconstruction and ensures the rationality of zoning boundaries; the dynamic virtual impedance decoupling control strategy ensures the stable operation of multi-terminal SOPs within the zoning and suppresses power circulating current; and the model prediction fault recovery strategy enables rapid recovery when a fault occurs. Together, these three constitute a complete technical solution, effectively solving many defects of existing technologies and improving the operating performance of AC / DC hybrid distribution networks.

Claims

1. A method for AC / DC hybrid distribution network partition autonomous and mutual aid operation based on flexible multi-state switch, characterized in that, Includes the following steps: Step S1: Construct a three-dimensional dynamic partitioning index system that includes SOP port power fluctuation characteristics, distributed power output uncertainty and power supply reliability, and calculate the comprehensive stability index of each node. Step S2: Monitor the comprehensive stability index in real time. When the comprehensive stability index of any node exceeds the preset first threshold, trigger partition reconstruction. Use the improved Louvain community discovery algorithm to re-divide the partition boundary. The node weight is dynamically adjusted according to the distributed power penetration rate and SOP port capacity. And the different ports of the same SOP must be located in different partitions. Step S3: In the autonomous partition mode, each partition's SOP adopts an AC / DC power decoupling control strategy based on dynamic virtual impedance, which suppresses power circulating current between multi-terminal SOPs by adjusting the dynamic virtual impedance value. Step S4: When the power deficit of a certain zone exceeds the preset proportion of the rated load of that zone and the duration exceeds the preset time threshold, the zone mutual assistance mode is activated. A mutual assistance priority sequence is generated based on the reserve capacity margin and voltage stability index of adjacent zones, and power is allocated between zones through SOP. Step S5: When a system failure occurs, a fault recovery strategy based on model predictive control is adopted to optimize the power of the SOP port and the output of the distributed power supply within a preset prediction time window, so as to achieve fault isolation and rapid power restoration of non-faulty areas. Step S6: Monitor the communication status between the global coordination unit and each partition control unit in real time. When the communication delay exceeds the preset communication threshold or the communication link is interrupted, the system immediately switches to fully autonomous mode, and each partition independently performs local control. After the communication is restored to stability, it automatically switches back to normal operation mode.

2. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, The comprehensive stability index is obtained by weighted summation of the SOP port power fluctuation coefficient, the distributed power output fluctuation coefficient, and the power supply reliability index. Each weight coefficient is preset according to the system operation requirements.

3. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, The first threshold is preset based on the maximum allowable stability deviation of the system, and is used to determine whether the system operating state has changed significantly, thereby triggering partition reconstruction.

4. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, The execution steps of the improved Louvain community detection algorithm include: Initialization: Assign each node in the distribution network to an independent community and calculate the initial modularity; Node weight update: The weight of each node is calculated based on the distributed power penetration rate, node voltage stability, and power supply reliability; Local move: For each node, try to move it to the adjacent community in turn, calculate the modularity increment, select the community corresponding to the largest modularity increment, and if the modularity increment is greater than 0, move the node to that community. Community aggregation: Treat each community as a super node, repeat the above steps of node weight update and local movement until the modularity no longer increases; Partition constraint check: Check if there are multiple ports of the same SOP located in the same partition. If so, adjust the partition affiliation of the port of that SOP and recalculate the modularity until the constraint conditions are met. Output the final partitioning results.

5. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, Corresponding to step S3, the dynamic virtual impedance value is the virtual equivalent impedance connected in series between the voltage outer loop and the current inner loop in the SOP AC / DC power decoupling control loop. Its value increases exponentially with the system running time and eventually converges to a preset steady-state value. The calculation formula for the dynamic virtual impedance is: ,in, This is the value of the dynamic virtual impedance. The preset maximum steady-state virtual impedance, t is the time constant, which is preset according to the dynamic response characteristics of the system; t is the duration for which the system enters the current operating condition.

6. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, Corresponding to step S4, the power deficit threshold is the ratio threshold of the real-time power deficit in the partition to the rated load of the partition, and the duration threshold is the minimum time threshold for the power deficit to continuously exceed the power deficit threshold. The power deficit threshold is usually set to 5% to 15% of the rated load of the partition, and the duration threshold is usually set to 10s to 30s. Both are preset according to the system power supply reliability level, distributed power penetration rate and load characteristics.

7. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, The rule for generating the mutual assistance priority sequence is as follows: adjacent partitions with larger reserve capacity margins and better voltage stability have higher mutual assistance priority and are given priority in providing power support to partitions with power shortages.

8. The method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, The prediction time window and control cycle of the model predictive control are preset according to the system response speed requirements. By predicting the system operating status in the future, the power of the SOP port and the output of the distributed power supply are planned in advance to achieve rapid fault recovery.

9. A method for autonomous and mutually supportive operation of AC / DC hybrid distribution networks based on flexible multi-state switches according to claim 1, characterized in that, Corresponding to step S6, the communication delay is the bidirectional data transmission delay between the global coordination unit and each partition control unit, and the preset communication threshold is set according to the real-time requirements of system control, usually 50ms~200ms; When the system detects that the communication delay exceeds the preset communication threshold or that the communication link is interrupted, it will immediately and automatically switch to fully autonomous mode, regardless of whether the system is currently in partition mutual aid mode or fault recovery mode. Each partition will perform independent control based solely on locally collected operating data, and stop the exchange of status information and power mutual aid between partitions. After the communication delay recovers to below the threshold and the communication link has been running stably for more than 30 seconds, the system will automatically switch back to the normal partition autonomous and mutual aid operation mode.

Images