A hybrid optimization decision-making method for ground fault in distribution network
By employing a multi-agent hierarchical scheme and multiple optimization schemes, the decision-making and analysis process at each stage of distribution network grounding faults is optimized, solving the problems of confusion and low efficiency in the diagnosis and handling of distribution network grounding faults, and achieving efficient fault diagnosis and power restoration.
Patent Information
- Application Number
- CN202111492569.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-08
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2041-12-08
AI Technical Summary
Frequent grounding faults in power distribution networks lead to accidents in the operation of power equipment and the power grid. Furthermore, the lack of a global coordination and processing system in existing equipment and systems results in low efficiency in diagnosis and processing, and easily leads to chaos and disorder.
A multi-agent hierarchical scheme and multiple optimization schemes are adopted to re-optimize and integrate the decision-making and analysis processes at each stage after a ground fault. This includes establishing a multi-agent hierarchical system for the distribution network, performing collaborative optimization of ground fault detection and fault handling, power restoration decision optimization, verification optimization, and operation optimization, to ensure collaborative work and decision consistency among the agents.
It has improved the efficiency of grounding fault diagnosis and subsequent handling in the distribution network, solved the problems of chaos and disorder, and improved the efficiency and effectiveness of the handling.
Smart Images

Figure CN114156847B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of distribution network grounding control, and in particular to a hybrid optimization decision-making processing method for distribution network grounding faults. Background Art
[0002] Grounding faults in distribution networks occur frequently, causing accidents in power equipment and power grid operation, as well as personal injury and death, with serious consequences, resulting in huge economic losses and adverse social impacts.
[0003] At present, there are many devices and systems for handling grounding faults in distribution networks, and different devices and systems belong to different departments of power companies. At present, a relatively sound overall coordination and processing system has not yet been established. After a grounding fault occurs in the distribution network, the "island" phenomenon of each device and system is more prominent, and the diagnosis and subsequent processing of grounding faults in the distribution network are prone to confusion and disorder. There are often prominent problems such as acting independently, low efficiency and poor problem handling results.
[0004] In addition, in the process of handling distribution network grounding faults and restoring power supply, there are many decision-making and analysis processes, and there is no unified processing plan that can be effectively implemented. It is also necessary to re-optimize and integrate the various decision-making and analysis processes to improve the efficiency of distribution network grounding fault diagnosis and subsequent processing. Summary of the invention
[0005] In order to solve the technical problems raised by the background technology, the present invention provides a hybrid optimization decision-making and processing method for grounding faults in a distribution network, which adopts a multi-agent grading scheme and a multiple optimization scheme to re-optimize and integrate the decision-making and analysis processes at each stage after the grounding fault, thereby improving the efficiency of the diagnosis and subsequent processing of grounding faults in the distribution network, solving the problem that the diagnosis and subsequent processing of grounding faults in the distribution network are prone to confusion and disorder, and often there are problems of each doing its own thing, low efficiency and poor problem handling results, thereby solving the problem of low efficiency in the diagnosis and subsequent processing of grounding faults in the distribution network.
[0006] In order to achieve the above object, the present invention adopts the following technical solutions:
[0007] A hybrid optimization decision-making method for distribution network grounding fault, the method comprising the following steps:
[0008] Step 1: Establish a multi-agent hierarchical system for distribution network to perform collaborative optimization of grounding detection and fault handling;
[0009] Step 2: Optimize power supply restoration decision;
[0010] Step 3: Distribution network restoration power supply verification optimization;
[0011] Step 4: Optimize power supply restoration operations;
[0012] Step 5: Optimization after the distribution network resumes normal operation.
[0013] Furthermore, in the step 1, the distribution network multi-agent hierarchical system includes the whole network intelligent agent, regional intelligent agent, line intelligent agent, segmented intelligent agent and terminal intelligent agent of the distribution network system. The whole network intelligent agent, regional intelligent agent, line intelligent agent, segmented intelligent agent and terminal intelligent agent are arranged hierarchically from superior to subordinate in the distribution network system. Each intelligent agent only communicates with the superior or subordinate intelligent agent within the adjacent range, and the adjacent range is the directly connected superior or subordinate.
[0014] Furthermore, in step 1, the objectives of collaborative optimization of the multi-agent hierarchical system include:
[0015] 1) There is no conflict in decision-making commands between agents;
[0016] 2) The ground fault processing time is shortest;
[0017] 3) The degree of cooperation between the agents is the best.
[0018] Furthermore, in step 1, the collaborative optimization of grounding detection and fault handling of the multi-agent hierarchical system specifically includes the following:
[0019] 1) According to the hierarchical positioning of each intelligent agent from the upper level to the lower level, different decision-making modes and functions are assigned to each intelligent agent; the whole network intelligent agent is assigned a single mode and function of centralized decision-making; the regional intelligent agent, line intelligent agent and segment intelligent agent are assigned two modes and functions of collaborative decision-making and edge decision-making; the terminal intelligent agent is assigned two modes and functions of collaborative decision-making and edge decision-making; among them, coordinated decision-making refers to the centralized decision-making of the upper level, and edge decision-making refers to the coordinated decision-making of the lower level;
[0020] 2) When the communication between the whole network agent, regional agent, line agent and segment agent is interrupted step by step, the lower level agent automatically enters the edge decision mode;
[0021] 3) When a ground fault occurs in the distribution network, any of the line agent, segment agent and terminal agent detects the ground fault judgment start signal, i.e., the first detection signal. The agent immediately enters the continuous high-frequency measurement point verification mode and simultaneously initiates the collaborative confirmation mechanism of the same-level, upper-level and lower-level agents of the segment or line to which it belongs. If the first detection signal agent passes the continuous measurement point verification, if the joint confirmation results of the ground fault of the same-level, upper-level and lower-level agents are inconsistent, confirm whether the ground fault occurs by comparing the sensitivity accumulation;
[0022] 4) When the sensitivity of ground fault determination is high or the ground fault results of all intelligent bodies are confirmed to be consistent, the distribution network ground fault type and fault location analysis and diagnosis are started, and the distribution network ground fault diagnosis and location are realized based on the expert system and knowledge base.
[0023] 5) After the type and location of the grounding fault in the distribution network are determined, the expert system is used to determine the fault isolation plan based on the principle of minimum impact and loss, and relevant operation instructions are generated and issued to the terminal intelligent agent for operation and control; the grounding fault location includes section, line, phase, etc.
[0024] Furthermore, in step 2, the optimization of power restoration decision includes:
[0025] The decision on power restoration after the isolation of the distribution network grounding fault is initiated. Based on the network topology analysis, the load transfer and power restoration path is searched, and the load transfer optimization algorithm is used to calculate the decision-making plan. If the optimization algorithm falls into the local optimum or diverges, the expert system is used to decide the suboptimal solution. The power restoration after the isolation of the distribution network grounding fault is carried out within the scope of the smallest regional distribution network.
[0026] Furthermore, in step 3, the optimization of the power supply restoration verification of the distribution network includes:
[0027] A verification system is established for the power supply restoration plan after the isolation of the distribution network grounding fault. According to the distribution network decompression and closing loop analysis method and its influencing factors, constraints including voltage, power factor, protection setting, current carrying capacity, and power supply capacity, based on the expert system and distribution network flow calculation and analysis, an operational plan that meets the constraints is solved; when all operational plans cannot meet the constraints or the number is less than N; return to the previous step and use the smallest regional distribution network as a unit to search for paths in the same level power grid, perform optimal calculation or suboptimal solution for power restoration, and solve operational plans; if the same level power grid cannot obtain an operational plan that meets the constraints, then search for paths in the next level power grid, perform optimal calculation or suboptimal solution for power restoration, and solve operational plans; and so on. If there is no solution, terminate the search and push an alarm to remind the power grid control personnel.
[0028] Furthermore, in step 4, the power restoration operation optimization includes:
[0029] A multi-objective optimization model is constructed for the operational scheme of ground fault isolation and power restoration in the distribution network. The optimization objectives of the model are:
[0030] 1) The terminal agent has the highest action reliability;
[0031] 2) The total number of terminal agent operations is minimal;
[0032] 3) The grid is most economical to operate;
[0033] Determine the final plan for isolating the ground fault in the distribution network and restoring power supply, and output the terminal intelligent body's power supply restoration operation control instructions.
[0034] Furthermore, the method further comprises:
[0035] Terminal intelligent body fault isolation operation control and power supply restoration operation control, establish regional or line intelligent body operation control nested "application-permission" and nested "lock-unlock" mechanism, strictly follow the regulation requirements to perform sequential operations step by step; the original state of each terminal intelligent body is "locked", after receiving the operation command, the terminal intelligent body needs to "apply" level by level, and after obtaining the "permission" of the highest-level intelligent body in the operable plan, the "locked" state is changed to the "unlocked" state, and the command operation is performed, and other terminal intelligent bodies remain "locked"; after the current terminal intelligent body operation is successful, it enters the "locked" state; the next command terminal intelligent body "application"-"permission"-"unlock"-operation-"lock", and so on; after a terminal intelligent body fails to perform an operation, it is allowed to repeat the operation, the number of times shall not exceed 3 times or the setting of manual intervention. If the operation fails continuously, the operation command shall be rolled back in turn, the original state shall be restored and an alarm shall be pushed to remind the power grid control personnel; the rollback operation shall also be verified according to the distribution network decompression and closure analysis method and its influencing factors and constraints. If rollback is impossible during the period, the rollback shall be terminated and an alarm shall be pushed to remind the power grid control personnel.
[0036] Furthermore, the step 5 comprises:
[0037] After the distribution network grounding fault diagnosis, determination, location, isolation and power supply restoration are completed, the event is recorded and pushed to the control personnel for reading and analysis. During the distribution network grounding fault diagnosis, determination, location, isolation and power supply restoration, if the distribution network grounding fault disappears and it is judged to be restored to normal through continuous measurement points and joint diagnosis, the operation instructions are rolled back in sequence to restore the original state and record the event; the rollback operation is verified according to the distribution network decompression and closing loop analysis method and its influencing factors and constraints. If rollback is impossible during the period, the rollback is terminated and an alarm is pushed to remind the power grid control personnel.
[0038] Compared with the prior art, the present invention has the following beneficial effects:
[0039] The method of the present invention adopts a multi-agent classification scheme and a multiple optimization scheme to re-optimize and integrate the decision-making and analysis processes at each stage after the grounding fault, improve the efficiency of the distribution network grounding fault diagnosis and subsequent processing, solve the problem that the distribution network grounding fault diagnosis and subsequent processing are prone to confusion and disorder, often there are problems of each doing their own thing, low efficiency and poor problem handling results, and solve the problem of low efficiency in the distribution network grounding fault diagnosis and subsequent processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 A flow chart of a hybrid optimization decision-making method for grounding faults in a distribution network according to the present invention;
[0041] Figure 2 This is a structural diagram of the distribution network multi-agent hierarchical system of the present invention. DETAILED DESCRIPTION
[0042] The specific implementation modes provided by the present invention are described in detail below with reference to the accompanying drawings.
[0043] like Figure 1 As shown, a hybrid optimization decision-making method for distribution network grounding fault is provided, wherein the method comprises the following steps:
[0044] Step 1: Establish a multi-agent hierarchical system for distribution network to perform collaborative optimization of grounding detection and fault handling;
[0045] Step 2: Optimize power supply restoration decision;
[0046] Step 3: Distribution network restoration power supply verification optimization;
[0047] Step 4: Optimize power supply restoration operations;
[0048] Step 5: Optimization after the distribution network resumes normal operation.
[0049] The following is a description of the specific implementation process
[0050] 1. Step 1
[0051] like Figure 2 As shown, in the step 1, the distribution network multi-agent hierarchical system includes the whole network intelligent agent, regional intelligent agent, line intelligent agent, segmented intelligent agent and terminal intelligent agent of the distribution network system. The whole network intelligent agent, regional intelligent agent, line intelligent agent, segmented intelligent agent and terminal intelligent agent are hierarchically arranged from superior to inferior in the distribution network system. Each intelligent agent only communicates with the superior or inferior intelligent agent within the adjacent range, and the adjacent range is the directly connected superior or inferior.
[0052] In step 1, the optimization objectives of the collaborative optimization of the multi-agent hierarchical system include:
[0053] 1) There is no conflict in decision-making commands between agents;
[0054] 2) The ground fault processing time is shortest;
[0055] 3) The degree of cooperation between the agents is the best.
[0056] In step 1, the collaborative optimization of grounding detection and fault handling of the multi-agent hierarchical system specifically includes the following:
[0057] 1) According to the hierarchical positioning of each intelligent agent from the upper level to the lower level, different decision-making modes and functions are assigned to each intelligent agent; the whole network intelligent agent is assigned a single mode and function of centralized decision-making; the regional intelligent agent, line intelligent agent and segment intelligent agent are assigned two modes and functions of collaborative decision-making and edge decision-making; the terminal intelligent agent is assigned two modes and functions of collaborative decision-making and edge decision-making; among them, coordinated decision-making refers to the centralized decision-making of the upper level, and edge decision-making refers to the coordinated decision-making of the lower level;
[0058] 2) When the communication between the whole network agent, regional agent, line agent and segment agent is interrupted step by step, the lower level agent automatically enters the edge decision mode;
[0059] 3) When a ground fault occurs in the distribution network, any of the line agent, segment agent and terminal agent detects the ground fault judgment start signal, i.e., the first detection signal. The agent immediately enters the continuous high-frequency measurement point verification mode and simultaneously initiates the collaborative confirmation mechanism of the same-level, upper-level and lower-level agents of the segment or line to which it belongs. If the first detection signal agent passes the continuous measurement point verification, if the joint confirmation results of the ground fault of the same-level, upper-level and lower-level agents are inconsistent, confirm whether the ground fault occurs by comparing the sensitivity accumulation;
[0060] 4) When the sensitivity of ground fault determination is high or the ground fault results of all intelligent bodies are confirmed to be consistent, the distribution network ground fault type and fault location analysis and diagnosis are started, and the distribution network ground fault diagnosis and location are realized based on the expert system and knowledge base.
[0061] 5) After the type and location of the grounding fault in the distribution network are determined, the expert system is used to determine the fault isolation plan based on the principle of minimum impact and loss, and relevant operation instructions are generated and issued to the terminal intelligent agent for operation and control; the grounding fault location includes section, line, phase, etc.
[0062] Step 2
[0063] In step 2, the optimization of power restoration decision includes:
[0064] The decision on power restoration after the isolation of the distribution network grounding fault is initiated. Based on the network topology analysis, the load transfer and power restoration path is searched, and the load transfer optimization algorithm is used to calculate the decision-making plan. If the optimization algorithm falls into the local optimum or diverges, the expert system is used to decide the suboptimal solution. The power restoration after the isolation of the distribution network grounding fault is carried out within the scope of the smallest regional distribution network.
[0065] Step 3
[0066] In step 3, the optimization of the power supply restoration verification of the distribution network includes:
[0067] A verification system is established for the power supply restoration plan after the isolation of the distribution network grounding fault. According to the distribution network decompression and closing loop analysis method and its influencing factors, constraints including voltage, power factor, protection setting, current carrying capacity, and power supply capacity, based on the expert system and distribution network power flow calculation and analysis, an operational plan that meets the constraints is solved; when all operational plans cannot meet the constraints or the number is less than N; N is 3 or set by manual intervention, return to the previous step and use the smallest regional distribution network as a unit to search for paths in the same level power grid, perform optimal calculation or suboptimal solution for power restoration, and solve operational plans; if the same level power grid cannot obtain an operational plan that meets the constraints, then the path search is performed on the next level power grid, the optimal calculation or suboptimal solution for power restoration is performed, and the operational plan is solved; and so on. If there is no solution, the search is terminated and an alarm is pushed to remind the power grid control personnel.
[0068] Step 4
[0069] In step 4, the power supply restoration optimization operation includes:
[0070] A multi-objective optimization model is constructed for the operational scheme of ground fault isolation and power restoration in the distribution network. The optimization objectives of the model are:
[0071] 1) The terminal agent has the highest action reliability;
[0072] 2) The total number of terminal agent operations is minimal;
[0073] 3) The grid is most economical to operate;
[0074] Determine the final plan for isolating the ground fault in the distribution network and restoring power supply, and output the terminal intelligent body's power supply restoration operation control instructions.
[0075] 5. During the fault isolation operation in step 1 and the power restoration operation in step 4, the terminal agent also performs the following control process:
[0076] Terminal intelligent body fault isolation operation control and power supply restoration operation control, establish regional or line intelligent body operation control nested "application-permission" and nested "lock-unlock" mechanism, strictly follow the regulation requirements to perform sequential operations step by step; the original state of each terminal intelligent body is "locked", after receiving the operation command, the terminal intelligent body needs to "apply" level by level, and after obtaining the "permission" of the highest-level intelligent body in the operable plan, the "locked" state is changed to the "unlocked" state, and the command operation is performed, and other terminal intelligent bodies remain "locked"; after the current terminal intelligent body operation is successful, it enters the "locked" state; the next command terminal intelligent body "application"-"permission"-"unlock"-operation-"lock", and so on; after a terminal intelligent body fails to perform an operation, it is allowed to repeat the operation, the number of times shall not exceed 3 times or the setting of manual intervention. If the operation fails continuously, the operation command shall be rolled back in turn, the original state shall be restored and an alarm shall be pushed to remind the power grid control personnel; the rollback operation shall also be verified according to the distribution network decompression and closure analysis method and its influencing factors and constraints. If rollback is impossible during the period, the rollback shall be terminated and an alarm shall be pushed to remind the power grid control personnel.
[0077] 6. Step 5:
[0078] After the grounding fault of the distribution network is handled, calculation and verification are performed based on the distribution network decompression and closing loop analysis method and its influencing factors and constraints, and the power supply restoration operation control instruction is obtained, and the distribution network resumes normal operation.
[0079] After the distribution network resumes normal operation, the following operations are also performed:
[0080] After the distribution network grounding fault diagnosis, determination, location, isolation and power supply restoration are completed, the event is recorded and pushed to the control personnel for reading and analysis. During the distribution network grounding fault diagnosis, determination, location, isolation and power supply restoration, if the distribution network grounding fault disappears and it is judged to be restored to normal through continuous measurement points and joint diagnosis, the operation instructions are rolled back in sequence to restore the original state and record the event; the rollback operation is verified according to the distribution network decompression and closing loop analysis method and its influencing factors and constraints. If rollback is impossible during the period, the rollback is terminated and an alarm is pushed to remind the power grid control personnel.
[0081] The method of the present invention adopts a multi-agent classification scheme and a multiple optimization scheme to solve the problem that the diagnosis and subsequent processing of grounding faults in the distribution network are prone to confusion and disorder, and often there are problems of each doing its own thing, low efficiency and poor problem handling results.
[0082] The above embodiments are implemented based on the technical solution of the present invention, and detailed implementation methods and specific operation processes are given, but the protection scope of the present invention is not limited to the above embodiments. The methods used in the above embodiments are conventional methods unless otherwise specified.
Claims
1. A hybrid optimization decision-making method for ground fault in distribution network. It is characterized in that The method comprises the following steps: Step 1: Establish a multi-agent hierarchical system for distribution network to perform collaborative optimization of grounding detection and fault handling; Step 2: Optimize power restoration decision; Step 3: Distribution network restoration power supply verification optimization; Step 4: Optimize power supply restoration operations; Step 5: Optimization after the distribution network resumes normal operation; In step 1, the optimization objectives of the collaborative optimization of the multi-agent hierarchical system include: 1) There is no conflict in decision-making commands between agents; 2) The ground fault processing time is shortest; 3) The cooperation between the agents is the best; In step 1, the collaborative optimization of grounding detection and fault handling of the multi-agent hierarchical system specifically includes the following: 1) According to the hierarchical positioning of each intelligent agent from the upper level to the lower level, different decision-making modes and functions are assigned to each intelligent agent; the whole network intelligent agent is assigned a single mode and function of centralized decision-making; the regional intelligent agent, line intelligent agent and segment intelligent agent are assigned two modes and functions of collaborative decision-making and edge decision-making; the terminal intelligent agent is assigned two modes and functions of collaborative decision-making and edge decision-making; among them, coordinated decision-making refers to the centralized decision-making of the upper level, and edge decision-making refers to the coordinated decision-making of the lower level; 2) When the communication between the whole network agent, regional agent, line agent and segment agent is interrupted step by step, the lower level agent automatically enters the edge decision mode; 3) When a ground fault occurs in the distribution network, any of the line agent, segment agent and terminal agent detects the ground fault judgment start signal, i.e., the first detection signal. The agent immediately enters the continuous high-frequency measurement point verification mode and simultaneously initiates the collaborative confirmation mechanism of the same-level, upper-level and lower-level agents of the segment or line to which it belongs. If the first detection signal agent passes the continuous measurement point verification, if the joint confirmation results of the ground fault of the same-level, upper-level and lower-level agents are inconsistent, confirm whether the ground fault occurs by comparing the sensitivity accumulation; 4) When the sensitivity of ground fault determination is high or the ground fault results of all intelligent bodies are confirmed to be consistent, the type of distribution network ground fault and fault location analysis and diagnosis are started, and the distribution network ground fault diagnosis and location are realized based on the expert system and knowledge base; 5) After the type and location of the grounding fault in the distribution network are determined, the expert system is used to determine the fault isolation plan based on the principle of minimum impact and loss, and relevant operation instructions are generated and issued to the terminal intelligent agent for operation and control; the grounding fault location includes section, line, and phase; In step 2, the optimization of power restoration decision includes: Start the decision to restore power supply after the distribution network grounding fault is isolated. Based on the network topology analysis, search for the load transfer and power restoration path, and use the load transfer optimization algorithm to calculate the decision-making plan; if the optimization algorithm falls into the local optimum or diverges, use the expert system to decide the suboptimal solution; after the distribution network grounding fault is isolated, the power supply restoration is carried out within the scope of the smallest regional distribution network; In step 3, the optimization of the power supply restoration verification of the distribution network includes: A verification system is established for the power supply restoration plan after the isolation of the distribution network grounding fault. According to the distribution network decompression and closing loop analysis method and its influencing factors, constraints including voltage, power factor, protection setting, current carrying capacity, and power supply capacity, based on the expert system and distribution network flow calculation and analysis, an operational plan that meets the constraints is solved; when all operational plans cannot meet the constraints or the number is less than N; return to the previous step and use the smallest regional distribution network as a unit to search for paths in the same-level power grid, perform optimal calculations for power restoration or solve suboptimal solutions, and solve operational plans; if the same-level power grid cannot obtain an operational plan that meets the constraints, then search for paths in the next-level power grid, perform optimal calculations for power restoration or solve suboptimal solutions, and solve operational plans; and so on. If there is no solution, terminate the search and push an alarm to remind the power grid control personnel; In step 4, the power supply restoration optimization operation includes: A multi-objective optimization model is constructed for the operational scheme of ground fault isolation and power restoration in the distribution network. The optimization objectives of the model are: 1) The terminal agent has the highest action reliability; 2) The total number of terminal agent operations is minimal; 3) The grid is most economical to operate; Determine the final plan for isolating the ground fault of the distribution network and restoring power supply, and output the terminal intelligent body's power supply restoration operation control instructions; In step 5, the optimization after the distribution network resumes normal operation includes: After the distribution network grounding fault diagnosis, determination, location, isolation and power supply restoration are completed, the event is recorded and pushed to the control personnel for reading and analysis. During the distribution network grounding fault diagnosis, determination, location, isolation and power supply restoration, if the distribution network grounding fault disappears and it is judged to be restored to normal through continuous measurement points and joint diagnosis, the operation instructions are rolled back in sequence to restore the original state and record the event; the rollback operation is verified according to the distribution network decompression and closing loop analysis method and its influencing factors and constraints. If rollback is impossible during the period, the rollback is terminated and an alarm is pushed to remind the power grid control personnel.
2. A hybrid optimization decision-making method for ground fault in distribution network according to claim 1, It is characterized in that In the step 1, the distribution network multi-agent hierarchical system includes the whole network intelligent agent, regional intelligent agent, line intelligent agent, segmented intelligent agent and terminal intelligent agent of the distribution network system. The whole network intelligent agent, regional intelligent agent, line intelligent agent, segmented intelligent agent and terminal intelligent agent are arranged hierarchically from superior to subordinate in the distribution network system. Each intelligent agent only communicates with the superior or subordinate intelligent agent within the adjacent range, and the adjacent range is the directly connected superior or subordinate.
3. A hybrid optimization decision-making method for ground fault in distribution network according to claim 1, It is characterized in that The method further comprises: Terminal intelligent body fault isolation operation control and power supply restoration operation control, establish regional or line intelligent body operation control nested "application-permission" and nested "lock-unlock" mechanism, strictly follow the regulation requirements to perform sequential operations step by step; the original state of each terminal intelligent body is "locked", after receiving the operation command, the terminal intelligent body needs to "apply" level by level, and after obtaining the "permission" of the highest-level intelligent body in the operable plan, the "locked" state is changed to "unlocked" state, and the command operation is performed, and other terminal intelligent bodies remain "locked"; after the current terminal intelligent body operation is successful, it enters the "locked" state; the next command terminal intelligent body "applies"-"permission"-"unlock"-operates-"locked", and so on; after a terminal intelligent body fails to perform an operation, it is allowed to repeat the operation, the number of times shall not exceed 3 times or the setting of manual intervention. If the operation fails continuously, the operation command shall be rolled back in turn, the original state shall be restored and an alarm shall be pushed to remind the power grid control personnel; the rollback operation shall also be verified according to the distribution network decompression and closing loop analysis method and its influencing factors and constraints. If rollback is impossible during the period, the rollback shall be terminated and an alarm shall be pushed to remind the power grid control personnel.
