Intelligent distributed feeder automation power grid self-healing method

Abstract

The invention discloses an intelligent distributed feeder automation power grid self-healing method, and the method comprises the steps: building an intelligent distributed single-node model for a switch SW and switches at the upstream and downstream of the switch SW, and carrying out the analysis of fault information and multicast signals through employing a short-circuit fault positioning and isolation logic algorithm based on directional locking and multicast communication; the method is suitable for a multi-power-source complex network with a distributed power source, the complex process that a corresponding algorithm needs to be selected after network identification in the traditional intelligent distributed technology is omitted, and the function design of the intelligent terminal is simplified; the method does not need to consider the network attitude and the characteristic of random change of the output power of the self-adaptive distributed power supply, can achieve accurate fault positioning calculation whether the distributed power supply is in a full-power operation state, a weak-power operation state or a zero-power operation state, corrects error judgment through fault-tolerant verification, and avoids secondary tripping power loss.

Description

Technical Field

[0001] This invention relates to the field of distribution network automation technology, specifically to a smart distributed feeder automation power grid self-healing method. Background Technology

[0002] With the accelerated development of distribution network automation, intelligent distributed feeder automation (FA) technology has been widely applied in county-level distribution networks in recent years due to its ability to quickly detect, isolate, and recover from faults without relying on a master station and through peer-to-peer communication between distribution terminals. However, the traditional "closed-loop design, open-loop operation" network structure of county-level distribution networks has evolved from a single-source radial network to a complex network with multiple sources coexisting after the integration of a large number of distributed generation (DG) sources, represented by photovoltaics. The output power of distributed generation is highly random and intermittent due to multiple factors such as sunshine, weather, and environment, causing the grid operation state to frequently switch between single-source and multi-source modes. This results in a situation where multiple modes coexist in actual operation of the distribution network, including open-loop distribution networks with distributed generation, closed-loop distribution networks with distributed generation, open-loop distribution networks without distributed generation, and closed-loop distribution networks without distributed generation.

[0003] Against this backdrop, existing intelligent distributed power supply (FA) technologies have revealed significant shortcomings. On the one hand, existing solutions (such as patents CN 110994786 A and CN 105207182 A) only design corresponding fault location algorithms for traditional open-loop or closed-loop distribution networks without distributed power sources, failing to consider the complexity of the network structure and the dynamic changes in operating modes after the integration of distributed power sources. This leads to the failure of fault location algorithms in practical applications. Although some patents (such as CN108173249 A) ​​propose a fault area location method for distribution networks considering the integration of inverter-type distributed power sources, their premise is that the output power of the distributed power source is constant, neglecting the random fluctuations in the output power during actual operation. When the output power of the distributed power source is low, the short-circuit current may be lower than the setting value of the distribution terminal unit (FTU), leading to missed fault information and thus causing the fault location algorithm to fail. Therefore, existing fault location algorithms cannot adapt to the characteristics of random changes in the output power of distributed power sources and are insufficient to meet the fault location requirements under complex operating modes.

[0004] On the other hand, existing intelligent distributed power distribution (FA) technologies generally lack effective fault tolerance mechanisms. Once fault information is distorted due to communication interference or equipment malfunction, it can easily lead to incorrect fault location, resulting in prolonged power outages for the entire transmission line. Furthermore, intelligent distributed FA technology is highly dependent on fiber optic communication, while county-level power distribution networks generally suffer from weak fiber optic communication infrastructure, high construction costs, and significant maintenance difficulties, severely hindering its widespread application in county-level power distribution networks.

[0005] Furthermore, in the 10kV low-current grounding systems widely present in county-level distribution networks, the rapid and accurate handling of single-phase grounding faults has become a rigid requirement for power grid companies. Currently, both intelligent distributed feeder automation (FA) technology and the adaptive integrated feeder automation technology proposed by the State Grid Corporation of China rely on the transient low-current grounding fault location function embedded in the primary and secondary integrated equipment to achieve fault location. However, due to the complex and variable actual operating conditions, the practical application effect of this technology is not ideal. According to the relevant full-scale test research results published in the Chinese core journal "Power System Protection and Control" (No. 12, 2021) ("Research on Full-Scale Test Technology for Single-Phase Grounding Fault Handling of Integrated Primary and Secondary Circuit Breakers"), in the full-scale tests conducted by the China Electric Power Research Institute, the accuracy rate of handling 1kΩ single-phase grounding faults was only 45%, and the accuracy rate of handling 2kΩ single-phase grounding faults was only 28%. Therefore, it is evident that the accuracy of existing technologies in handling single-phase grounding faults in actual operation urgently needs further improvement.

[0006] In summary, existing intelligent distributed fault diagnosis (FA) technologies are insufficient to meet the urgent needs of county-level distribution networks for highly reliable and adaptable fault handling technologies, especially given the complex and ever-changing operation modes of distribution networks after large-scale integration of distributed power sources, the risk of fault information distortion, and the low accuracy of single-phase grounding fault handling in low-current grounding systems. How to achieve adaptive fault location algorithms to random variations in the output power of distributed power sources, improve fault information tolerance, reduce dependence on fiber optic communication, and effectively improve the accuracy of single-phase grounding fault handling in low-current grounding systems have become key technical issues that urgently need to be addressed in the current automation construction of county-level distribution networks. Summary of the Invention

[0007] Based on the problems raised in the background technology above, the purpose of this invention is to provide a self-healing method for intelligent distributed feeder automated power grids, which solves the problems.

[0008] This invention is achieved through the following technical solution: The first aspect of this invention provides a self-healing method for intelligent distributed feeder automated power grids, comprising the following steps: Step S1: Establish an intelligent distributed single-node model based on multicast communication; wherein the intelligent distributed single-node model includes a switch SW, and the switch SW has at least one side connected to other switches; Step S2: The switch SW performs a self-test to confirm whether there is any abnormality in the transmission and reception of multicast signals. If there is an abnormality, the intelligent distributed FA is locked. If there is no abnormality, step S3 is executed. Step S3: Collect the fault information of the switch SW and the multicast signals of the adjacent switches of the switch SW, and use the short-circuit fault location and isolation logic algorithm to analyze the fault information and the multicast signals, and determine the fault location based on the analysis results; Step S4: Use a slow-motion intelligent distributed strategy to isolate faults at the fault boundary switches based on the fault location. Step S5: Except for the fault boundary switches that have been isolated, the remaining switches on the feeder are restored to power using the VIT strategy.

[0009] In the above technical solution, an intelligent distributed single-node model is established for the switch SW and its upstream and downstream switches. Since the model does not contain any network identification or topology discrimination unit, the same model can be seamlessly applied to four complex operating modes: "with / without distributed power supply" and "open loop / closed loop". The terminal does not need to identify the network structure, which eliminates the complex process of network identification and selection of the corresponding algorithm in traditional intelligent distributed technology, improves the adaptability of the fault location algorithm, and simplifies the functional design of the intelligent terminal.

[0010] The fault location and isolation logic algorithm is used to analyze the fault information and the multicast signal. It does not need to consider the network mentality and is adaptive to the random changes in the output power of the distributed power source. It can achieve accurate fault location calculation regardless of whether the distributed power source is operating at full power, low power or zero power.

[0011] In one optional embodiment, an intelligent distributed single-node model based on multicast communication is established, including: The preset power reference direction of switch SW and the preset multicast address for other switches connected to switch SW; In this context, the upstream side is defined as the direction opposite to the reference direction of the switch SW, and the downstream side is defined as the direction opposite to the reference direction of the switch SW. There are two preset power reference directions for the switch SW: 0 points to the busbar and 1 points to the line. The preset power reference direction can be any direction. The upstream side is the M side and the downstream side is the N side. The preset multicast addresses for other switches connected to the switch SW include: A multicast address A is preset for the upstream switch connected to the switch SW, and the corresponding address of the upstream switch is added to the multicast address A. The multicast address A is controlled by Gocb1. A multicast address B is preset for the downstream switch connected to the switch SW, and the corresponding address of the downstream switch is added to the multicast address B, which is controlled by Gocb2. A multicast address C is preset for the upstream and downstream switches connected to the switch SW, and the corresponding addresses of the upstream and downstream switches are added to the multicast address C, which is controlled by Gocb3.

[0012] In one optional embodiment, a short-circuit fault location and isolation logic algorithm is used to analyze the fault information and the multicast signal, including the following steps: Step S31: Based on the fault information, determine the fault of the fault itself and obtain the fault result. Step S32: The switch SW and the adjacent switch use peer-to-peer communication to send and receive Goose multicast signals, and extract the blocking signal and fault signal from the multicast information to obtain the blocking signal result and the fault signal result. Step S33: Perform a comprehensive analysis based on the self-fault result, the interlocking signal result, and the fault signal result to obtain the fault location and record the fault location result. After recording, proceed to step S4.

[0013] In one optional embodiment, determining a fault based on the fault information includes: Obtain the phase current setting value; Extract the phase current flowing through the switch SW from the fault information, compare the phase current with the phase current setting value, and if the phase current is greater than the phase current setting value, the self-fault result is that a self-fault exists; otherwise, the self-fault result is that a self-fault does not exist.

[0014] In one optional embodiment, the switch SW and adjacent switches communicate peer-to-peer to send and receive Goose multicast signals, including: If the self-fault result indicates that a self-fault exists, a "node fault" Goose multicast signal is sent to the multicast address C, and then broadcast to the adjacent upstream and downstream switches through the multicast address C.

[0015] Obtain a preset fault power reference direction, determine the fault power direction based on the fault information, and if the fault power direction is consistent with the preset fault power reference direction, send a "blocking trip" Goose multicast signal to the multicast address A, and then broadcast it to the adjacent upstream switch through the multicast address A; otherwise, send a "blocking trip" Goose multicast signal to the multicast address B, and then broadcast it to the adjacent downstream switch through the multicast address B.

[0016] In one optional embodiment, a comprehensive analysis is performed based on the self-fault result, the interlocking signal result, and the fault signal result, including: If the self-fault result indicates that a self-fault exists, and the interlocking signal result indicates that an interlocking signal has been received, then it is determined that the switch SW is not in the fault section and will not be activated. If the self-fault result indicates that a self-fault exists, and the interlocking signal result indicates that no interlocking signal has been received, then the switch SW is determined to be in the fault section, the fault location is determined and the fault location result is recorded, and then step S4 is executed. If the self-fault result is that there is no self-fault, but the fault signal result is that there is a fault on only one side of the adjacent upstream and downstream, and the interlocking signal result is that no interlocking signal is received, then it is determined that the switch SW is in the fault section, the fault location is determined and the fault location result is recorded, and then step S4 is executed. If the self-fault result is that there is no self-fault, but the fault signal result is that there is a fault on only one side of the adjacent upstream and downstream, and the interlocking signal result is that an interlocking signal is received, then it is determined that the switch SW is not in the fault section and will not be operated. If the self-fault result indicates that there is no self-fault, and the fault signal result indicates that there is no fault in the adjacent upstream and downstream, then it is determined that the switch SW is not in the fault section and will not be activated.

[0017] In one optional embodiment, a slow-motion intelligent distributed strategy is used to isolate the fault boundary switch based on the fault location, including: judging the voltage and current of the operating condition of the switch SW; if the switch SW has no voltage and no current, then tripping the fault location according to the fault location result recorded in step S33, and sending a "fault isolated" Goose multicast signal to the multicast address C, and then broadcasting it to the adjacent upstream and downstream switches through the multicast address C.

[0018] In an optional embodiment, the method further includes: using a VIT strategy for power restoration; The VIT strategy for power restoration includes: coordinating the reclosing of switches before the fault section with the substation's primary reclosing, and coordinating the closing of switches after the fault section with the tie switches to transfer power. The VIT strategy is used to sequentially close and supply power to the fault boundary switch.

[0019] In one optional embodiment, the VIT strategy is used for power restoration while VIT technology is used for fault tolerance verification, including: The switch SW is used to correct erroneous closing. If the switch SW is mistakenly closed in the fault location, the fault location and isolation are accelerated by using phase-to-phase overcurrent on the side of the switch SW connected to the fault location, and the fault location and isolation are performed by using residual voltage blocking on the side of the switch SW not connected to the fault location.

[0020] Compared with the prior art, the present invention has the following advantages and beneficial effects: 1. This invention adapts to complex multi-power supply networks with distributed power sources and eliminates the need for network topology identification technology. This omits the complex process of network identification followed by algorithm selection in traditional technologies, thus simplifying the design of smart terminal functions. 2. This invention adopts a fault-tolerant verification mechanism. Even if the terminal erroneously sends a "blocking" signal, this technology can still accurately locate and isolate the fault during the power restoration process, and will not cause secondary power loss of the feeder, thus improving reliability. 3. The present invention features adaptive distributed power supply output power that varies randomly, enabling accurate fault location and isolation regardless of whether the distributed power supply is operating at full power, low power, or zero power. 4. This invention avoids the problem of traditional intelligent distributed technologies relying on transient methods embedded in primary and secondary fusion equipment for line and segment selection during single-phase grounding fault handling. This technical solution can seamlessly integrate with substation low-current grounding fault selection, utilizing the substation's low-current grounding fault selection device for "lateral identification" of grounding faults to achieve accurate line selection; and utilizing the mature voltage-type distribution network single-phase grounding fault handling method built into the intelligent terminal for "vertical identification" of grounding faults to achieve accurate segment selection. By adopting a collaborative operation mode of "lateral identification" and "vertical identification" both inside and outside the station, it overcomes the shortcomings of traditional intelligent distributed FA and adaptive integrated feeder automation systems that lack sufficient individual combat capability and have difficulty guaranteeing handling effectiveness when using transient methods in primary and secondary fusion equipment for handling low-current grounding faults. This improves the accuracy of grounding fault handling and simplifies the functional design of the intelligent terminal. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 This is a schematic diagram of a process for automated short-circuit fault location and isolation of an intelligent distributed feeder provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the structure of the intelligent distributed single-node model provided in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of a phase-to-phase short-circuit fault in an open-loop distribution network with distributed energy resources provided in Embodiment 2 of the present invention, under the full-power operation mode of the DG. Figure 4This is a schematic diagram of a phase-to-phase short-circuit fault in an open-loop distribution network with distributed energy resources provided in Embodiment 2 of the present invention, under the low-power operation mode of DG. Figure 5 This is a schematic diagram of a phase-to-phase short-circuit fault in an open-loop distribution network with distributed energy resources provided in Embodiment 2 of the present invention, under the zero-power operation mode of DG. Figure 6 A schematic diagram of a phase-to-phase short-circuit fault in a closed-loop distribution network with distributed energy provided in Embodiment 2 of the present invention, under full-power DG operation mode; Figure 7 This is a schematic diagram of a phase-to-phase short-circuit fault in a closed-loop distribution network with distributed energy resources provided in Embodiment 2 of the present invention, under the low-power operation mode of DG. Figure 8 A schematic diagram of a phase-to-phase short-circuit fault in a closed-loop distribution network with distributed energy provided in Embodiment 2 of the present invention under the zero-power operation mode of DG; Figure 9 This is a flowchart illustrating the logic algorithm for the entire process of phase-to-phase short-circuit fault location, isolation, and power grid self-healing provided in Embodiment 3 of the present invention. Figure 10 This is a schematic diagram of a single-phase grounding fault in an open-loop distribution network with distributed energy provided in Embodiment 4 of the present invention. Figure 11 This is a flowchart illustrating the single-phase grounding fault handling logic algorithm of the intelligent distributed feeder automation power grid self-healing technology provided in Embodiment 4 of the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0023] Example 1 Embodiment 1 of the present invention provides a self-healing method for intelligent distributed feeder automation power grids, such as... Figure 1 As shown, it includes the following steps: Step S1: Establish an intelligent distributed single-node model based on multicast communication; wherein the intelligent distributed single-node model includes a switch SW, and the switch SW has at least one side connected to other switches; Step S2: The switch SW performs a self-test to confirm whether there is any abnormality in the transmission and reception of multicast signals. If there is an abnormality, the intelligent distributed FA is locked. If there is no abnormality, step S3 is executed. Step S3: Collect the fault information of the switch SW and the multicast signals of the adjacent switches of the switch SW, and use the short-circuit fault location and isolation logic algorithm to analyze the fault information and the multicast signals, and determine the fault location based on the analysis results; Step S4: Use a slow-motion intelligent distributed strategy to isolate faults at the fault boundary switches based on the fault location. Step S5: Except for the fault boundary switches that have been isolated, the remaining switches on the feeder are restored to power using the VIT strategy.

[0024] It should be noted that by establishing an intelligent distributed single-node model for the switch SW and its upstream and downstream switches, since the model does not contain any network identification or topology discrimination unit, the same model can be seamlessly applied to four complex operating modes: "with / without distributed power supply" and "open-loop / closed-loop". The terminal does not need to identify the network structure, which omits the complex process of network identification and selection of the corresponding algorithm in traditional intelligent distributed technology, thus simplifying the functional design of intelligent terminals.

[0025] The fault location and isolation logic algorithm is used to analyze the fault information and the multicast signal. It does not need to consider the network mentality and is adaptive to the random changes in the output power of the distributed power source. It can achieve accurate fault location calculation regardless of whether the distributed power source is operating at full power, low power or zero power.

[0026] The distributed power source can identify the unique and correct fault segment whether it is generating at full capacity, low capacity, or zero capacity. Even if a terminal mistakenly sends a "blocking" signal, the location logic can be corrected in the next fault-tolerant verification, thus avoiding misjudgment and preventing secondary tripping and power loss.

[0027] This technical solution avoids the problem of traditional intelligent distributed technologies relying on transient methods embedded in primary and secondary fusion equipment for fault selection in single-phase grounding fault handling. It seamlessly integrates with substation low-current grounding fault location, utilizing the substation's low-current grounding fault location device for "lateral identification" of grounding faults to achieve accurate fault selection; and employing mature voltage-type distribution network single-phase grounding fault handling methods built into the intelligent terminal for "vertical identification" of grounding faults to achieve accurate segment selection. By adopting a collaborative operation mode of "lateral identification" and "vertical identification" both inside and outside the station, it overcomes the shortcomings of traditional intelligent distributed FA and adaptive integrated feeder automation systems that lack sufficient individual operational capability and have difficulty guaranteeing the handling effect when using transient methods in primary and secondary fusion equipment for handling low-current grounding faults. This improves the accuracy of grounding fault handling and simplifies the functional design of the intelligent terminal.

[0028] In one optional embodiment, an intelligent distributed single-node model based on multicast communication is established, including: The preset power reference direction of switch SW and the preset multicast address for other switches connected to switch SW; The upstream side is defined as the direction opposite to the reference direction of the switch SW, and the downstream side is defined as the direction opposite to the reference direction of the switch SW. There are two preset power reference directions for the switch SW: 0 points to the busbar and 1 points to the line. The preset power reference direction can be any direction. The preset multicast addresses for other switches connected to the switch SW include: A multicast address A is preset for the upstream switch connected to the switch SW, and the corresponding address of the upstream switch is added to the multicast address A. The multicast address A is controlled by Gocb1. A multicast address B is preset for the downstream switch connected to the switch SW, and the corresponding address of the downstream switch is added to the multicast address B, which is controlled by Gocb2. A multicast address C is preset for the upstream and downstream switches connected to the switch SW, and the corresponding addresses of the upstream and downstream switches are added to the multicast address C, which is controlled by Gocb3.

[0029] Intelligent distributed single-node model such as Figure 2 As shown, the upstream side connected to switch SW is the M side, and the downstream side connected to switch SW is the N side. The upstream side switch has a preset multicast address A, the downstream side switch has a preset multicast address B, and both the upstream and downstream side switches have preset multicast addresses C.

[0030] Specifically, for all upstream switches connected to switch SW, a multicast address A is preset, and the corresponding addresses of all upstream switches are added to this multicast group, controlled by goose control block 1 (Gocb1); for all downstream switches connected to switch SW, a multicast address B is preset, and the corresponding addresses of all downstream switches are added to this multicast group, controlled by goose control block 2 (Gocb2); for all upstream and downstream switches connected to switch SW, a multicast address C is preset, and the corresponding addresses of both upstream and downstream switches are added to this multicast group, controlled by goose control block 3 (Gocb3).

[0031] In one optional embodiment, a short-circuit fault location and isolation logic algorithm is used to analyze the fault information and the multicast signal, including the following steps: Step S31: Based on the fault information, determine the fault of the fault itself and obtain the fault result. Step S32: The switch SW and the adjacent switch use peer-to-peer communication to send and receive Goose multicast signals, and extract the blocking signal and fault signal from the multicast information to obtain the blocking signal result and the fault signal result. Step S33: Perform a comprehensive analysis based on the self-fault result, the interlocking signal result, and the fault signal result to obtain the fault location and record the fault location result. After recording, proceed to step S4.

[0032] See the specific judgment logic. Figure 1 The short-circuit fault location and isolation logic algorithm first needs to determine whether the intelligent distributed FA has completed charging. If not, the FA is blocked. If it is, the fault is determined and the fault result is obtained. Then, the fault power direction is determined, the blocking signal is determined, and the upstream and downstream fault signals are determined.

[0033] The charging conditions for FA are as follows: switch SW is in the closed position; switch SW is fault-free and its adjacent sides are also fault-free; at least one side of switch SW has voltage; and there are no discharge conditions. After all the above conditions are met, the system will switch to charging mode after a 60-second delay.

[0034] The FA discharge conditions are: switch SW is in the open position; distributed FA function is off; there is no voltage delay for 60 seconds on both sides of switch SW; no discharge conditions. When any of the above conditions are met, the fault location is instantaneously discharged.

[0035] In one optional embodiment, determining a fault based on the fault information includes: Obtain the phase current setting value; Extract the phase current flowing through the switch SW from the fault information, compare the phase current with the phase current setting value, and if the phase current is greater than the phase current setting value, the self-fault result is that a self-fault exists; otherwise, the self-fault result is that a self-fault does not exist.

[0036] When a short-circuit fault occurs in the system, the phase current flowing through switch SW is greater than the set value. The fault of switch SW node is pre-judged. Gocb3 controls the instantaneous triggering of the "node fault" goose signal to the preset multicast address C (the corresponding addresses of all adjacent upstream and downstream switches have been added). Then, the "node fault" goose signal is broadcast to all adjacent upstream and downstream switches through multicast address C.

[0037] In one optional embodiment, the switch SW and adjacent switches communicate peer-to-peer to send and receive Goose multicast signals, including: If the self-fault result indicates that a self-fault exists, a "node fault" Goose multicast signal is sent to the multicast address C, and then broadcast to the adjacent upstream and downstream switches through the multicast address C.

[0038] Obtain a preset fault power reference direction, determine the fault power direction based on the fault information, and if the fault power direction is consistent with the preset fault power reference direction, send a "blocking trip" Goose multicast signal to the multicast address A; otherwise, send a "blocking trip" Goose multicast signal to the multicast address B.

[0039] Specifically, the SW direction element determines the current fault current direction and compares it with a preset power reference direction. If the two directions are the same, Gocb1 controls the instantaneous triggering of a "blocking" goose signal to the preset multicast address A (the corresponding addresses of all adjacent upstream switches have been added), and then broadcasts the "blocking" goose signal to all adjacent upstream switches through multicast address A. If the two directions are opposite, Gocb2 controls the instantaneous triggering of a "blocking" goose signal to the preset multicast address B (the corresponding addresses of all adjacent downstream switches have been added), and then broadcasts the "blocking" goose signal to all adjacent downstream switches through multicast address B.

[0040] In one optional embodiment, the blocking signal result includes: receiving a blocking signal and not receiving a blocking signal; the fault signal result includes: no fault in both adjacent upstream and downstream, fault in only one of the adjacent upstream and downstream, and fault in both adjacent upstream and downstream.

[0041] In one optional embodiment, a comprehensive analysis is performed based on the self-fault result, the interlocking signal result, and the fault signal result, including: If the self-fault result indicates that a self-fault exists, and the interlocking signal result indicates that an interlocking signal has been received, then it is determined that the switch SW is not in the fault section and will not be activated. If the self-fault result indicates that a self-fault exists, and the interlocking signal result indicates that no interlocking signal has been received, then the switch SW is determined to be in the fault section, the fault location is determined and the fault location result is recorded, and then step S4 is executed. If the self-fault result is that there is no self-fault, but the fault signal result is that there is a fault on only one side of the adjacent upstream and downstream, and the interlocking signal result is that no interlocking signal is received, then it is determined that the switch SW is in the fault section, the fault location is determined and the fault location result is recorded, and then step S4 is executed. If the self-fault result is that there is no self-fault, but the fault signal result is that there is a fault on only one side of the adjacent upstream and downstream, and the interlocking signal result is that an interlocking signal is received, then it is determined that the switch SW is not in the fault section and will not be operated. If the self-fault result indicates that there is no self-fault, and the fault signal result indicates that there is no fault in the adjacent upstream and downstream, then it is determined that the switch SW is not in the fault section and will not be activated.

[0042] In one optional embodiment, a slow-motion intelligent distributed strategy is used to isolate the fault based on the fault location, including: judging the voltage and current of the operating condition of the switch SW; if the switch SW has no voltage and no current, then tripping the fault location according to the fault location result recorded in step S33, and sending a "fault isolated" Goose multicast signal to the multicast address C, and then broadcasting it to the adjacent upstream and downstream switches through the multicast address C.

[0043] In an optional embodiment, the method further includes: using a VIT strategy for power restoration; The VIT strategy for power restoration includes: coordinating the reclosing of switches before the fault section with the substation's primary reclosing, and coordinating the closing of switches after the fault section with the tie switches to transfer power. The VIT strategy is used to sequentially close and supply power to the fault boundary switch.

[0044] In one optional embodiment, the VIT strategy is used for power restoration while VIT technology is used for fault tolerance verification, including: The switch SW is used to correct erroneous closing. If the switch SW is mistakenly closed in the fault location, the fault location and isolation are accelerated by using phase-to-phase overcurrent on the side of the switch SW connected to the fault location, and the fault location and isolation are performed by using residual voltage blocking on the side of the switch SW not connected to the fault location.

[0045] To address FA fault location errors caused by signal distortion, this embodiment employs fault-tolerant technology, using VIT as the verification fault-tolerant control logic. It should be noted that VIT and distributed FA are performed simultaneously. If signal distortion causes FA fault location error, when the switch SW is mistakenly closed to the fault section, the switch on the same side connected to the fault location uses the "phase-to-phase overcurrent acceleration" function in VIT control, and the switch on the opposite side not connected to the fault location uses the "residual voltage blocking" function in VIT control to locate and isolate the fault, thus achieving fault tolerance verification.

[0046] Example 2 Taking a typical county-level distribution network with distributed energy (photovoltaics) as an example, we will implement the intelligent distributed feeder automation grid self-healing method provided in Example 1.

[0047] S1: Each smart switch establishes a single-node model, presets the power reference direction, and defines adjacent switches that are opposite to the reference direction as upstream M-side switches and those that are in the reference direction as downstream N-side switches; S2: Each smart switch SW checks whether the multicast signal based on goose is normal. If the communication is abnormal, the FA is blocked; if the communication is normal, proceed to S3.

[0048] S3: When a short circuit fault occurs at line f1, each smart switch SW determines whether it is in the fault section based on its own fault information and the A, B, and C multicast signals received from adjacent switches, and follows specific steps according to different operating conditions.

[0049] When in Figure 3 Under normal operating conditions, the process for determining whether switch SW is in a fault zone is as follows: With the interconnection switch K7 disconnected, the main grid operates in open-loop mode, and all photovoltaic power stations operate at full power. The grid structure is a multi-source distribution network with the main grid open-loop and distributed energy.

[0050] When a short-circuit fault occurs at point f1, CB1 provides short-circuit fault current to K1 and K2; photovoltaic power station DG1 provides short-circuit fault current to K3; photovoltaic power station DG2 provides short-circuit fault current to K4 and K5; the direction of the fault current is shown by the dashed arrow. The table below shows the fault determination process and results for each node.

[0051] For node K2, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is the same as the preset power direction, it sends a "blocking" signal to multicast address A (adjacent upstream switch K1); according to the algorithm of embodiment 1, since K2 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0052] For node K3, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is opposite to the preset power direction, it sends a "blocking" signal to multicast address B (adjacent downstream switch); according to the algorithm of embodiment 1, since K2 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0053] For node K4, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is opposite to the preset power direction, it sends a "blocking" signal to multicast address B (adjacent downstream switches K5 and K6); according to the algorithm of embodiment 1, since K2 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0054] For node K1, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section and does not take any action because it receives the "blocking" multicast signal from K2.

[0055] For node K5, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section because it receives the "blocking" multicast signal from K4, and therefore does not take any action.

[0056] For node K6, according to the algorithm of Example 1, it does not sense any fault current and there is a fault on only one side of the adjacent upstream / downstream side. It also receives a "blocking" multicast signal from K4. Therefore, it is determined to be a non-faulty section and no action is taken.

[0057] The above steps complete the fault location and isolation, and the faulty section is determined to be the power supply area between switches K2, K3, and K4.

[0058] When in Figure 4 Under normal operating conditions, the process for determining whether switch SW is in a fault zone is as follows: With the interconnection switch K7 disconnected, the main grid operates in open-loop mode, and each photovoltaic power station operates at low power. The grid structure is a multi-source distribution network with the main grid open-loop and distributed energy (DG).

[0059] When a short-circuit fault occurs at f1, CB1 provides short-circuit fault current to K1 and K2; the photovoltaic power station DG1, operating at low power, provides a short-circuit fault current to K3 that is less than the FTU setting value; the photovoltaic power station DG2, operating at low power, provides a short-circuit fault current to K4 and K5 that is less than the FTU setting value; the direction of the fault current is shown by the dashed arrow. The table below shows the fault determination process and results for each node.

[0060] For node K2, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is the same as the preset power direction, it sends a "blocking" signal to multicast address A (adjacent upstream switch K1); according to the algorithm of embodiment 1, since K2 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0061] For node K3, since the photovoltaic power station DG1 is operating at low power, the short-circuit fault current supplied to K3 is less than the FTU setting value. Therefore, K3 itself does not feel the fault current. According to the algorithm of Example 1, K3 itself does not feel the fault current, there is a fault on only one side of the adjacent upstream / downstream side, and no "blocking" multicast signal is received. Therefore, it is determined to be a fault section and the fault location result is recorded, and the process proceeds to step S4.

[0062] For node K4, since the photovoltaic power station DG2 is operating at low power, the short-circuit fault current supplied to K4 and K5 is less than the FTU setting value. Therefore, K4 itself does not feel the fault current. According to the algorithm of Example 1, K4 itself does not feel the fault current and there is a fault on only one side of the adjacent upstream / downstream side, and no "blocking" multicast signal is received. Therefore, it is determined to be a fault section and the fault location result is recorded, and the process proceeds to step S4.

[0063] For node K1, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section and does not take any action because it receives the "blocking" multicast signal from K2.

[0064] For nodes K5 and K6, according to the algorithm in Example 1, they do not sense any fault current and there are no faults on either the adjacent upstream or downstream sides, so they are determined to be non-faulty sections and no action is taken.

[0065] The faulty section is identified as the power supply area between switches K2, K3, and K4.

[0066] When in Figure 5 Under normal operating conditions, the process for determining whether switch SW is in a fault zone is as follows: With the interconnection switch K7 disconnected, the main grid operates in open loop, each photovoltaic power station outputs zero power, and the grid structure becomes an open-loop distribution network without distributed energy (DG).

[0067] When a short-circuit fault occurs at point f1, only CB1 provides short-circuit fault current to K1 and K2; photovoltaic power station DG1 and DG2 do not provide short-circuit current. The table below shows the fault determination process and results for each node.

[0068] The fault location steps and algorithm results are the same as in operating condition two. The fault section is determined to be the power supply area between switches K2, K3, and K4.

[0069] When in Figure 6 Under normal operating conditions, the process for determining whether switch SW is in a fault zone is as follows: When the interconnection switch K7 is closed, the main grid operates in closed loop, and all photovoltaic power stations operate at full power. The grid structure is a multi-source distribution network with main grid closed loop and distributed energy (DG).

[0070] When a short-circuit fault occurs at f1, CB1 provides short-circuit fault current to K1 and K2; CB2 provides short-circuit fault current to K4, K6, K7, K8, and K9; photovoltaic DG1 provides short-circuit fault current to K3; and photovoltaic DG2 provides short-circuit fault current to K4 and K5. The table below shows the fault determination process and results for each node.

[0071] For node K2, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is the same as the preset power direction, it sends a "blocking" signal to multicast address A (adjacent upstream switch K1); according to the algorithm of embodiment 1, since K2 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0072] For node K3, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is opposite to the preset power direction, it sends a "blocking" signal to multicast address B (adjacent downstream switch); according to the algorithm of embodiment 1, since K3 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0073] For node K4, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is opposite to the preset power direction, it sends a "blocking" signal to multicast address B (adjacent downstream switches K5 and K6); according to the algorithm of embodiment 1, since K4 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and the process proceeds to step S4.

[0074] For node K1, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section and does not take any action because it receives the "blocking" multicast signal from K2.

[0075] For node K5, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section because it receives the "blocking" multicast signal from K4, and therefore does not take any action.

[0076] For node K6, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section because it receives the "blocking" multicast signal from K4, and therefore does not take any action.

[0077] For node K7, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section because it receives the "blocking" multicast signal from K6, and therefore does not take any action.

[0078] For node K8, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section because it receives the "blocking" multicast signal from K7, and therefore does not take any action.

[0079] For node K9, according to the algorithm of Example 1, although it senses the fault current, it is determined to be a non-faulty section because it receives the "blocking" multicast signal from K8, and therefore does not take any action.

[0080] The above steps complete the fault location, and the faulty section is determined to be the power supply area between switches K2, K3, and K4.

[0081] When in Figure 7 Under normal operating conditions, the process for determining whether switch SW is in a fault zone is as follows: When the interconnection switch K7 is closed, the main grid operates in a closed loop, and each photovoltaic power station operates at low power. The grid structure is a multi-source distribution network with a main grid closed loop and distributed energy (DG).

[0082] When a short-circuit fault occurs at f1, CB1 provides short-circuit fault current to K1 and K2; CB2 provides short-circuit fault current to K4, K6, K7, K8, and K9; the short-circuit fault current provided by the photovoltaic power station's DG1 operating at low power to K3 is less than the FTU setting value; the short-circuit fault current provided by the photovoltaic power station's DG2 operating at low power to K4 and K5 is less than the FTU setting value; the direction of the fault current is shown by the dashed arrow. The table below shows the fault determination process and results for each node.

[0083] For node K2, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is the same as the preset power direction, it sends a "blocking" signal to multicast address A (adjacent upstream switch K1); according to the algorithm of embodiment 1, since K2 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and proceed to step S4.

[0084] For node K3, since the photovoltaic power station DG1 is operating at low power, the short-circuit fault current supplied to K3 is less than the FTU setting value. Therefore, K3 itself does not feel the fault current. According to the algorithm of Example 1, K3 itself does not feel the fault current and there is a fault on only one side of the adjacent upstream / downstream side, and no "blocking" multicast signal is received. Therefore, it is determined to be a fault section and the fault location result is recorded, and the process proceeds to step S4.

[0085] For node K4, it senses the fault current and sends a "node fault" signal to multicast address C; since the direction of the fault current is opposite to the preset power direction, it sends a "blocking" signal to multicast address B (adjacent downstream switches K5 and K6); according to the algorithm of embodiment 1, since K4 senses the fault current and does not receive the "blocking" signal, it is determined to be a fault section and the fault location result is recorded, and the process proceeds to step S4.

[0086] The faulty section is identified as the power supply area between switches K2, K3, and K4.

[0087] When in Figure 8 Under normal operating conditions, the process for determining whether switch SW is in a fault zone is as follows: When the interconnection switch K7 is closed, the main grid operates in a closed loop. At night, the solar irradiance drops to zero, and each photovoltaic power station outputs zero power. The grid structure becomes a closed-loop distribution network without distributed energy (DG).

[0088] When a short-circuit fault occurs at point f1, CB1 provides short-circuit fault current to K1 and K2; CB2 provides short-circuit fault current to K4, K6, K7, K8, and K9. The table below shows the fault determination process and results for each node.

[0089] The fault location steps and algorithm results are the same as in operating condition five. The fault section is determined to be the power supply area between switches K2, K3, and K4.

[0090] When a fault occurs at point f1, the substation outlet switch CB1 and the photovoltaic power station grid connection point switch trip. After the switch nodes are de-energized and de-currented, according to the fault location determination result recorded in step S33, switches K2, K3, and K4 are tripped, and "fault isolated" goose signals are sent to their respective preset multicast addresses C, thereby completing the intelligent distributed FA fault location and isolation.

[0091] Example 3 This embodiment uses a common distribution network with distributed energy (photovoltaic) and... Figure 3 Taking the distribution network structure as an example, the intelligent distributed feeder automation power grid self-healing method provided in Example 1 is implemented. See [link to example]. Figure 9 The steps for handling phase-to-phase short-circuit faults are ordered according to the time of action and include the following steps: S1: A phase-to-phase short circuit occurred at feeder f1. S2: The substation outlet circuit breaker CB1 relay protection and each intelligent switch on the feeder are started simultaneously by the intelligent distributed FA.

[0092] S3: After the substation outlet circuit breaker CB1 time-limited instantaneous protection is activated, it trips after a 0.3-second delay, resulting in a complete power outage on the entire line.

[0093] S4: After each intelligent switch FA on the feeder in step S2 is activated, the easing-type intelligent distributed FA logic is processed (detailed steps for fault location and isolation are shown in Example 1, and the algorithm logic flowchart is shown in...). Figure 2 Within the fault confirmation time limit (3.5S), the fault boundary switches K2, K3, and K4 are tripped and the circuit breaker is locked, and a "fault isolated" GOOSE signal is sent. S5: After step S3, all intelligent switches on the line (except for fault boundary switches K2, K3, and K4, which have already been blocked by FA tripping) execute the VIT control logic. That is, after the fault confirmation time Z (set to 3.5 seconds), all switches on the feeder line execute the "power failure tripping" logic function to trip. At this point, all intelligent switches on the line are in the tripped state, and the three fault boundary switches are blocked and closed, putting them in a fault isolation state, preparing for the line to be restored to power.

[0094] S6: The CB1 switch at the substation outlet recloses and supplies power after 5 seconds. S7: Each terminal after the CB1 switch on the line executes the VIT logic, and each switch executes the "energize and close" control logic in sequence until the "fault boundary switch K2" is energized.

[0095] S8: After step S4, the tie switch K7 is fully charged and receives the "fault isolated" GOOSE signal. After a set delay, the tie switch is closed and powered on.

[0096] S9: Terminals K6 and K4 before the tie switch K7 execute VIT logic, and switch K6 executes "energize and close" to send power to fault boundary switch K4.

[0097] At this point, the integrated feeder automation (FA) process is complete.

[0098] Specifically, during step S4, if signal distortion in the slow-motion intelligent distributed FA leads to errors in fault location and isolation during fault handling, for example, if K3, due to signal distortion, incorrectly sends a "blocking" signal to multicast address A (adjacent upstream switches K2 and K4), causing fault boundary switches K2 and K4 to incorrectly identify themselves as non-faulty sections and fail to take action, resulting in the K2 to K4 area (the real faulty section) not being effectively isolated, then during steps S7 and S9, each terminal executes VIT logic (which also serves as a verification and fault-tolerance strategy for the intelligent distributed FA), and each switch sequentially "energizes and closes" until power is supplied to the real faulty section. At this time, the intelligent distribution terminal of fault boundary switch K2 activates the "phase-to-phase overcurrent acceleration protection," trips and blocks the closing, and sends a "fault isolated" GOOSE signal; simultaneously, fault boundary switch K4 detects residual voltage and blocks the closing, sending a "fault isolated" GOOSE signal. Thus, the real faulty section is effectively isolated by the verification and fault-tolerance strategy.

[0099] Furthermore, to ensure that the intelligent distribution terminal can operate independently when executing the verification and fault-tolerance strategy (VIT logic) and avoid confusion with the slow-acting intelligent distributed FA logic, and to ensure that they do not interfere with each other in terms of algorithm logic, a fault clearing and isolation charging and discharging condition for the intelligent distributed FA is designed. This condition shields the intelligent distributed FA when the VIT logic verification and fault-tolerance scheme is activated, allowing the VIT logic to operate independently. Specifically, when step S5 is executed, all switches activate the "power failure tripping" logic function and trip. Because the switches are in the open position, the discharge condition is met, thereby instantaneously discharging the fault clearing and isolation function of the intelligent distributed FA, temporarily shielding the slow-acting intelligent distributed FA, and ensuring that the intelligent distribution terminal can operate independently when executing the VIT logic (verification and fault-tolerance strategy).

[0100] This invention employs a fault-tolerant verification mechanism. Even if the terminal erroneously sends a "blocking" signal, the fault can still be accurately located and isolated during power restoration without causing secondary power loss to the feeder, thus improving reliability. It is accommodating to the information interaction time of the distribution terminal, the inherent action time of the distribution terminal, and the inherent opening time of the circuit breaker. As long as the sum of these three times is less than the fault confirmation Z time (set to 3.5 seconds), this technology is satisfied. This solves the dependence of intelligent distributed FA on low-latency communication, i.e., optical fiber, enabling the application of higher-latency 4G and IoT communication in peer-to-peer communication of intelligent distributed FA. This makes it possible to fully implement intelligent distributed FA in county-level distribution networks.

[0101] Example 4 This embodiment takes a common distribution network with a 10KV low-current grounding system for distributed energy (photovoltaics) as an example and implements a smart distributed feeder automation grid self-healing method provided in Embodiment 1.

[0102] See the route diagram below. Figure 10 It should be noted that in this embodiment, K7 disconnection is an open-loop multi-source network, and its process is as follows: Figure 11 As shown, it includes the following steps: S1: A single-phase ground fault occurred at feeder f1, and zero-sequence overvoltages were detected at all nodes K1, K2, K3, K4, K5, and K6 along the entire line.

[0103] S2: Because the transient method for selecting and segmenting grounding lines and grounding is not enabled at each distribution terminal, the intelligent distributed FA at each node is not started.

[0104] S3: The substation low-current grounding fault location system is working. After the fault is located, the substation outlet CB1 switch is tripped. Subsequently, the photovoltaic (DG) anti-islanding protection trips the grid-connected switch, and the zero-sequence overvoltage of the entire network disappears.

[0105] S4: After the entire faulty line loses power, switches K1, K2, K3, K4, K5, and K6 on the feeder line activate the "power failure tripping" logic function. After the fault confirmation Z time (set to 3.5 seconds), the switches will trip. At this point, all intelligent switches on the line are in the tripped state, and the intelligent distributed FA fault location and isolation function will discharge instantaneously.

[0106] S5: The CB1 switch at the substation outlet will reclose once after 5 seconds.

[0107] S6: The first smart switch K1 on the line is energized on one side, executes the VIT logic, and closes after a delay of X time (set to 7S), and supplies power to the next smart switch K2. K2 is energized on one side, closes after a delay of X time (set to 7S), and if the fault at f1 is permanent, it closes and supplies power to the ground fault section. At this time, K2 terminal detects zero-sequence overvoltage, and the "zero-sequence overvoltage acceleration protection" action trips K2 switch and blocks the closing. K2 switch acts as fault boundary switch 1 and sends a "fault isolated" GOOSE signal. At the same time, the adjacent next smart switches K3 and K4 are blocked from closing due to the detection of residual voltage, acting as fault boundary switches 2. Fault boundary switch 3 sends a "fault isolated" GOOSE signal.

[0108] S7: When the tie switch K7 has finished charging and received the "Fault Isolated" GOOSE signal, the tie switch closes and supplies power after a set delay.

[0109] S8: Switches K5 and K6 before the tie switch K7 on the line execute VIT logic, and each switch is "energized and closed" in sequence until power is supplied to the fault boundary switch K4.

[0110] At this point, the automated execution of the intelligent distributed feeder is complete.

[0111] In this embodiment, each power distribution terminal does not need to enable the transient method small current grounding line selection and segment selection function.

[0112] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A self-healing method for intelligent distributed feeder automated power grids, characterized in that, Includes the following steps: Step S1: Establish an intelligent distributed single-node model based on multicast communication; wherein the intelligent distributed single-node model includes a switch SW, and the switch SW has at least one side connected to other switches; Step S2: The switch SW performs a self-test to confirm whether there is any abnormality in the transmission and reception of multicast signals. If there is an abnormality, the intelligent distributed FA is locked. If there is no abnormality, step S3 is executed. Step S3: Collect the fault information of the switch SW and the multicast signals of the adjacent switches of the switch SW, and use the short-circuit fault location and isolation logic algorithm to analyze the fault information and the multicast signals, and determine the fault location based on the analysis results; Step S4: Use a slow-motion intelligent distributed strategy to isolate faults at the fault boundary switches based on the fault location. Step S5: Except for the fault boundary switches that have been isolated, the remaining switches on the feeder are restored to power using the VIT strategy.

2. The intelligent distributed feeder automation power grid self-healing method according to claim 1, characterized in that, Establish an intelligent distributed single-node model based on multicast communication, including: The preset power reference direction of switch SW and the preset multicast address for other switches connected to switch SW; In this context, the upstream side is defined as the direction opposite to the reference direction of the switch SW, and the downstream side is defined as the direction opposite to the reference direction of the switch SW. There are two preset power reference directions for the switch SW: 0 points to the busbar and 1 points to the line. The preset power reference direction can be any direction. The upstream side is the M side and the downstream side is the N side. The preset multicast addresses for other switches connected to the switch SW include: A multicast address A is preset for the upstream switch connected to the switch SW, and the corresponding address of the upstream switch is added to the multicast address A. The multicast address A is controlled by Gocb1. A multicast address B is preset for the downstream switch connected to the switch SW, and the corresponding address of the downstream switch is added to the multicast address B, which is controlled by Gocb2. A multicast address C is preset for the upstream and downstream switches connected to the switch SW, and the corresponding addresses of the upstream and downstream switches are added to the multicast address C, which is controlled by Gocb3.

3. The intelligent distributed feeder automation power grid self-healing method according to claim 2, characterized in that, The fault information and the multicast signal are analyzed using a short-circuit fault location and isolation logic algorithm, including the following steps: Step S31: Based on the fault information, determine the fault of the fault itself and obtain the fault result. Step S32: The switch SW and the adjacent switch use peer-to-peer communication to send and receive Goose multicast signals, and extract the blocking signal and fault signal from the multicast information to obtain the blocking signal result and the fault signal result. Step S33: Perform a comprehensive analysis based on the self-fault result, the interlocking signal result, and the fault signal result to obtain the fault location and record the fault location result. After recording, proceed to step S4.

4. The intelligent distributed feeder automation power grid self-healing method according to claim 3, characterized in that, Based on the fault information, perform self-fault diagnosis, including: Obtain the phase current setting value; Extract the phase current flowing through the switch SW from the fault information, compare the phase current with the phase current setting value, and if the phase current is greater than the phase current setting value, the self-fault result is that a self-fault exists; otherwise, the self-fault result is that a self-fault does not exist.

5. The intelligent distributed feeder automation power grid self-healing method according to claim 3, characterized in that, The switch SW communicates with adjacent switches via peer-to-peer communication to send and receive Goose multicast signals, including: If the self-fault result indicates that a self-fault exists, then a "node fault" Goose multicast signal is sent to the multicast address C, and then broadcast to the adjacent upstream and downstream switches through the multicast address C; Obtain a preset fault power reference direction, determine the fault power direction based on the fault information, and if the fault power direction is consistent with the preset fault power reference direction, send a "blocking trip" Goose multicast signal to the multicast address A, and then broadcast it to the adjacent upstream switch through the multicast address A; otherwise, send a "blocking trip" Goose multicast signal to the multicast address B, and then broadcast it to the adjacent downstream switch through the multicast address B.

6. The intelligent distributed feeder automation power grid self-healing method according to claim 3, characterized in that, The interlocking signal results include: interlocking signal received and interlocking signal not received; the fault signal results include: no fault in adjacent upstream and downstream, fault in only one side of adjacent upstream and downstream, and fault in both adjacent upstream and downstream.

7. The intelligent distributed feeder automated power grid self-healing method according to claim 3, characterized in that, A comprehensive analysis is performed based on the self-fault results, the interlocking signal results, and the fault signal results, including: If the self-fault result indicates that a self-fault exists, and the interlocking signal result indicates that an interlocking signal has been received, then it is determined that the switch SW is not in the fault section and will not be activated. If the self-fault result indicates that a self-fault exists, and the interlocking signal result indicates that no interlocking signal has been received, then the switch SW is determined to be in the fault section, the fault location is determined and the fault location result is recorded, and then step S4 is executed. If the self-fault result is that there is no self-fault, but the fault signal result is that there is a fault on only one side of the adjacent upstream and downstream, and the interlocking signal result is that no interlocking signal is received, then it is determined that the switch SW is in the fault section, the fault location is determined and the fault location result is recorded, and then step S4 is executed. If the self-fault result is that there is no self-fault, but the fault signal result is that there is a fault on only one side of the adjacent upstream and downstream, and the interlocking signal result is that an interlocking signal is received, then it is determined that the switch SW is not in the fault section and will not be operated. If the self-fault result indicates that there is no self-fault, and the fault signal result indicates that there is no fault in the adjacent upstream and downstream, then it is determined that the switch SW is not in the fault section and will not be activated.

8. The intelligent distributed feeder automation power grid self-healing method according to claim 3, characterized in that, The fault isolation of the fault boundary switch is performed by adopting a slow-motion intelligent distributed strategy based on the fault location. This includes: judging the voltage and current of the operating condition of the switch SW; if the switch SW has no voltage and no current, tripping is performed on the fault location based on the fault location result recorded in step S33, and a "fault isolated" Goose multicast signal is sent to the multicast address C, which is then broadcast to the adjacent upstream and downstream switches through the multicast address C.

9. The intelligent distributed feeder automation power grid self-healing method according to claim 1, characterized in that, This also includes: using VIT strategies for power restoration; The VIT strategy for power restoration includes: coordinating the reclosing of switches before the fault section with the substation's primary reclosing, and coordinating the closing of switches after the fault section with the tie switches to transfer power. The VIT strategy is used to sequentially close and supply power to the fault boundary switch.

10. A self-healing method for intelligent distributed feeder automated power grids according to claim 9, characterized in that, The VIT strategy is used for power restoration, while VIT technology is used for fault tolerance verification, including: The switch SW is used to correct erroneous closing. If the switch SW is mistakenly closed in the fault location, the fault location and isolation are accelerated by using phase-to-phase overcurrent on the side of the switch SW connected to the fault location, and the fault location and isolation are performed by using residual voltage blocking on the side of the switch SW not connected to the fault location.

Images