Intelligent fusion terminal distributed collaborative decision-making method and system for distribution network self-healing

Abstract

The application discloses a smart fusion terminal distributed collaborative decision-making method and system for distribution network self-healing, relates to the technical field of distribution network collaborative decision-making, and comprises the following steps: multiple smart fusion terminals of each node of an interactive distribution network obtain monitoring and identification information and establish a self-healing control intention; a smart fusion terminal set participating in this time collaborative decision-making is determined; in a collaborative decision-making granularity ring, each smart fusion terminal performs dynamic permission allocation and collaborative relationship negotiation, judges a candidate self-healing control scheme, performs safety and feasibility verification, and sends the self-healing control scheme to the smart fusion terminals for collaborative execution after verification. The application solves the technical problems of the prior art, such as static rigidity of a collaborative decision-making range, inflexible permission and role allocation, and lack of a distributed scheme verification mechanism, and thus the reliability and adaptability of distribution network self-healing are insufficient, and the technical effects of dynamically adapting to the influence of abnormal events, optimizing decision-making efficiency, and improving the response speed, adaptability and reliability of distribution network self-healing are achieved.

Description

Technical Field

[0001] This invention relates to the technical field of distribution network collaborative decision-making, specifically to a distributed collaborative decision-making method and system for intelligent converged terminals used for distribution network self-healing. Background Technology

[0002] The distribution network is a crucial link connecting the main grid with massive distributed power sources and diverse loads. Traditional fault handling and control models, relying on centralized analysis from the main station and hierarchical command issuance, exhibit limitations such as slow response speed, strong communication dependence, and insufficient decision-making flexibility when dealing with localized sudden faults and random fluctuations in distributed power sources. This is particularly true in the event of abnormal events like line faults, voltage exceeding limits, and sudden load changes, where efficient and reliable distribution network self-healing is difficult to achieve. Intelligent fusion terminals with local information acquisition, processing, and control capabilities act as intelligent agents for physical nodes, capable of real-time monitoring of electrical quantities, switch status, and environmental information. However, the perspective of a single intelligent terminal is limited to local information, and its control decisions may lack global coordination, potentially even triggering cascading risks. Existing distributed intelligent feeder automation (FA) and multi-agent systems (MAS) based on peer-to-peer communication have improved the distribution and speed of distribution network self-healing to some extent, but they still face many problems. The granularity of collaborative decision-making is pre-defined statically, making it difficult to adapt to changes in network topology and the dynamic impact range of abnormal events of different natures. The collaborative mechanism between nodes is often based on fixed master-slave or polling rules, lacking the flexibility to dynamically assign roles and negotiate permissions based on the nature of the event, node status and capabilities, which may lead to low decision-making efficiency or unclear responsibility. The generated recovery scheme lacks an efficient and reliable collaborative verification mechanism in a distributed environment, and the security and execution coordination of the scheme are not adequately guaranteed.

[0003] Therefore, current technologies suffer from several technical problems, including a static and rigid scope of collaborative decision-making, inflexible allocation of permissions and roles, and a lack of distributed solution verification mechanisms, which result in insufficient reliability and adaptability of the distribution network's self-healing mechanism. Summary of the Invention

[0004] This application provides a distributed collaborative decision-making method and system for intelligent converged terminals for distribution network self-healing, which solves the technical problems in the prior art, such as static and rigid collaborative decision-making scope, inflexible allocation of permissions and roles, and lack of distributed scheme verification mechanism, resulting in insufficient reliability and adaptability of distribution network self-healing. It achieves the technical effects of dynamically adapting to the impact of abnormal events, optimizing decision-making efficiency, and improving the response speed, adaptability, and reliability of distribution network self-healing.

[0005] This application provides a distributed collaborative decision-making method for intelligent fusion terminals for distribution network self-healing. The method includes: interactively deploying multiple intelligent fusion terminals at various nodes of the distribution network to obtain monitoring and identification information of each intelligent fusion terminal; when any intelligent fusion terminal detects a local abnormal event, generating abnormal evidence with confidence and establishing a preliminary self-healing control intention; based on the topological relationship of the multiple intelligent fusion terminals, and combined with the abnormal event and the intelligent fusion terminal nodes affected by the preliminary self-healing control intention, determining the set of intelligent fusion terminals participating in this collaborative decision-making, forming a collaborative decision granularity circle; within the collaborative decision granularity circle, each intelligent fusion terminal, based on its node role, status information, and the abnormal evidence, dynamically allocates permissions and negotiates collaborative relationships through information interaction, jointly judges the preliminary self-healing control intention, and generates at least one candidate self-healing control scheme; verifying the security and feasibility of the candidate self-healing control scheme, and sending it to the intelligent fusion terminals with corresponding permissions for collaborative execution of the self-healing control scheme after passing the verification.

[0006] In a possible implementation, the set of intelligent fusion terminals participating in this collaborative decision-making is determined to form a collaborative decision-making granularity circle, including: based on the intelligent fusion terminal nodes where the local abnormal event occurred and the data flow relationship reflected by the abnormal evidence, correlation and positioning are performed in the topological relationship to determine the topologically associated terminal nodes; according to the correlation control relationship and control response relationship corresponding to the preliminary self-healing control intention, the relevant intelligent fusion terminal nodes affected by the self-healing control strategy are determined to form a control-affected terminal node set; and the collaborative decision-making granularity circle is constructed based on the topologically associated terminal nodes and the control-affected terminal node set.

[0007] In a possible implementation, constructing the collaborative decision-making granularity circle further includes: obtaining the self-healing control permissions and control granularity strategies corresponding to the topology-associated terminal nodes and the control-affected terminal node set; determining the collaborative association granularity between each terminal node based on the self-healing control permissions and control granularity strategies of each terminal node; and using the collaborative association granularity to label the collaborative decision-making granularity circle with self-healing granularity rules, which is used to limit the self-healing control participation scope and control level of each intelligent fusion terminal within the collaborative decision-making granularity circle.

[0008] In a possible implementation, obtaining the self-healing control permissions and control granularity strategies corresponding to the topology-associated terminal nodes and the control-affected terminal node set includes: determining the device risk parameters of each smart converged terminal based on the controllable distribution network device type, device operating status, and device risk level corresponding to each smart converged terminal; determining the permission level that each smart converged terminal is allowed to participate in self-healing control based on the device risk parameters and the impact tolerance level corresponding to the device abnormal events; and determining the control granularity strategy corresponding to each smart converged terminal based on the permission level and the impact tolerance level, which is used to limit the self-healing control participation depth and control action range of each smart converged terminal.

[0009] In possible implementations, determining the impact tolerance level corresponding to an abnormal equipment event includes: determining the direct risk level of the abnormal event based on the abnormal type, duration, and amplitude characteristics; obtaining the impact range risk of the abnormal event based on the number of distribution network devices affected, the load importance level, and the power supply impact range; and comprehensively determining the impact tolerance level of the abnormal event based on the direct risk level and the impact range risk, which characterizes the acceptability of the distribution network operation safety and power supply reliability without immediately implementing self-healing control operations.

[0010] In possible implementations, the self-healing control's authority levels include: a first authority level that allows the execution of abnormal isolation control; a second authority level that allows the execution of abnormal isolation and load transfer control; and a third authority level that allows the execution of abnormal isolation, load transfer, and tie switch closing control.

[0011] In a possible implementation, within the collaborative decision-making granularity circle, each intelligent fusion terminal, based on its node role, status information, and the aforementioned anomaly evidence, dynamically allocates permissions and negotiates collaborative relationships through information interaction. This allows for a joint judgment of the initial self-healing control intent, generating at least one candidate self-healing control scheme. The process includes: cross-verifying the anomaly evidence using monitoring network information from topologically related terminal nodes to determine the verification result; performing consistency judgment and correction on the initial self-healing control intent based on the verified anomaly evidence; and, under the constraint of the corrected self-healing control intent, dynamically allocating self-healing control permissions and negotiating collaborative relationships through information interaction, in conjunction with the corresponding permission levels and control granularity strategies of each intelligent fusion terminal within the collaborative decision-making granularity circle, generating at least one candidate self-healing control scheme.

[0012] In possible implementations, the candidate self-healing control scheme is subjected to security and feasibility verification, including: performing security prediction verification based on the current operating state parameters to assess the operational risks that may be caused by the execution of the candidate self-healing control scheme; performing feasibility verification based on the execution node state and control constraints to assess the execution reachability of the candidate self-healing control scheme; determining the execution timing of the candidate self-healing control scheme based on the expected self-healing effect of the candidate self-healing control scheme and the tolerance of the impact of abnormal events; and allowing the execution of the candidate self-healing control scheme when the verification results meet the preset execution conditions.

[0013] In a possible implementation, the intelligent fusion terminal distributed collaborative decision-making method for distribution network self-healing further includes: after executing the self-healing control scheme, performing state verification on the self-healing control execution result; when the verification result does not meet the preset operational safety conditions, performing self-healing control rollback processing, wherein the rollback processing includes at least one of the following: canceling the executed self-healing control operation to restore the distribution network operation state to the safe state before execution; prohibiting the continued execution of subsequent control actions in the self-healing control scheme; reducing the self-healing control permission level that can be granted in subsequent self-healing control processes or narrowing the collaborative decision-making granularity circle based on the rollback result; and sending abnormal evidence, execution results, and rollback information to the superior system or entering the manual intervention process.

[0014] This application also provides a distributed collaborative decision-making system for intelligent fusion terminals for distribution network self-healing. The system includes: an anomaly evidence generation module, used to interact with multiple intelligent fusion terminals deployed at various nodes of the distribution network to obtain monitoring and identification information of each intelligent fusion terminal. When any intelligent fusion terminal detects a local anomaly event, it generates anomaly evidence with confidence and establishes a preliminary self-healing control intention; a collaborative decision granularity circle formation module, used to determine the set of intelligent fusion terminals participating in this collaborative decision based on the topological relationship of the multiple intelligent fusion terminals and the intelligent fusion terminal nodes affected by the anomaly event and the preliminary self-healing control intention, forming a collaborative decision granularity circle; a self-healing control intention judgment module, used within the collaborative decision granularity circle, for each intelligent fusion terminal to dynamically allocate permissions and negotiate collaborative relationships through information interaction based on its node role, status information, and the anomaly evidence, to jointly judge the preliminary self-healing control intention and generate at least one candidate self-healing control scheme; and a control scheme verification module, used to verify the security and feasibility of the candidate self-healing control scheme, and after passing the verification, send it to the intelligent fusion terminals with corresponding permissions to collaboratively execute the self-healing control scheme.

[0015] This application proposes a distributed collaborative decision-making method and system for distribution network self-healing using intelligent fusion terminals. This method involves multiple intelligent fusion terminals interacting across nodes in the distribution network to obtain monitoring and identification information and establish self-healing control intentions. It determines the set of intelligent fusion terminals participating in the collaborative decision-making process. Within the collaborative decision-making granularity, each intelligent fusion terminal dynamically allocates permissions and negotiates collaborative relationships, generating at least one candidate self-healing control scheme. Security and feasibility verification are performed, and after successful verification, the scheme is sent to the intelligent fusion terminals for collaborative execution. This method solves the technical problems in existing technologies, such as static and rigid collaborative decision-making scope, inflexible permission and role allocation, and the lack of distributed scheme verification mechanisms, which lead to insufficient reliability and adaptability of distribution network self-healing. It achieves the technical effects of dynamically adapting to the impact of abnormal events, optimizing decision-making efficiency, and improving the response speed, adaptability, and reliability of distribution network self-healing. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 This is a schematic diagram of the distributed collaborative decision-making method for intelligent converged terminals used for distribution network self-healing, provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the structure of a distributed collaborative decision-making system for intelligent converged terminals used for self-healing of distribution networks, provided in an embodiment of this application.

[0019] Figure labeling: Abnormal evidence generation module 10, collaborative decision granularity circle formation module 20, self-healing control intent judgment module 30, control scheme verification module 40. Detailed Implementation

[0020] To further illustrate the technical means and effects adopted by the present invention in order to achieve the intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structures, features and effects of the present invention.

[0021] This application provides a distributed collaborative decision-making method for intelligent converged terminals for distribution network self-healing, such as... Figure 1 As shown, the method includes:

[0022] Step S100: Interact with the multi-intelligent fusion terminals deployed at each node of the distribution network to obtain the monitoring and identification information of each intelligent fusion terminal. When any intelligent fusion terminal detects a local abnormal event, it generates abnormal evidence with confidence and establishes a preliminary self-healing control intention.

[0023] Preferably, multiple intelligent fusion terminals deployed at distribution network nodes such as switching stations, ring main units, and distribution transformers are interconnected through communication networks such as fiber optics, power line carriers, and wireless private networks. These terminals collect and identify monitoring information in real time, including but not limited to electrical quantities such as voltage, current, power, and frequency; switch statuses such as circuit breakers, load switches, and disconnect switches; equipment statuses such as temperature, partial discharge, and protection signals; and environmental information such as weather conditions and security alarms. When any intelligent fusion terminal analyzes the monitoring information using overcurrent criteria, voltage limit comparison, and waveform distortion analysis, and identifies situations exceeding preset thresholds or conforming to specific fault modes, it determines it as a local abnormal event, such as phase-to-phase short circuits or grounding faults, bus voltage loss or severe limit exceedances, etc. If a key load trips abnormally, structured anomaly evidence with confidence is generated. This evidence includes at least the anomaly type, occurrence time, measured value, relevant measurement point identifiers, sampling rate, and a confidence level calculated through multiple rounds of sampling verification, redundant measurement point comparison, and historical data reference. This level characterizes the reliability of the anomaly evidence. Then, based on the anomaly evidence from the same intelligent fusion terminal, combined with locally stored network topology knowledge and preset rules, a preliminary response strategy is automatically generated, i.e., a preliminary self-healing control intention. This intention includes control objectives, such as isolating the faulty line segment, restoring power supply to the upstream healthy area, and transferring important downstream loads. The proposed operating equipment includes a list of switches that the terminal can control or recommends operating, along with their expected actions, and the expected effects, such as disconnecting the fault point from the power grid or transferring the undervoltage load to a backup power source.

[0024] Step S200: Based on the topological relationship of the multiple intelligent fusion terminals, and combined with the abnormal events and the intelligent fusion terminal nodes affected by the preliminary self-healing control intention, determine the set of intelligent fusion terminals participating in this collaborative decision-making, and form a collaborative decision-making granularity circle.

[0025] Step S200 further includes: based on the intelligent fusion terminal node where the local abnormal event occurred and the data flow relationship reflected by the abnormal evidence, performing correlation and positioning in the topology relationship to determine the topology-related terminal node; determining the relevant intelligent fusion terminal nodes affected by the self-healing control strategy according to the correlation control relationship and control response relationship corresponding to the preliminary self-healing control intention, forming a control-affected terminal node set; and constructing the collaborative decision granularity circle based on the topology-related terminal node and the control-affected terminal node set.

[0026] Preferably, the intelligent fusion terminal node where a local anomaly occurs is taken as the source node. The anomaly evidence generated by this node contains information reflecting data flow characteristics such as electrical quantity mutations and protection action signals. The dynamically learned distribution network topology describes the electrical connection relationships and communication paths between all terminal nodes. Specifically, in the topology graph, starting from the source node, and combining the flow characteristics in the anomaly evidence, such as the direction of fault current and the propagation path of voltage sag, the topology is tracked and located to identify other terminal nodes that are directly electrically related to this anomaly or have a strong data correlation with it. The topology-related terminal nodes are output, which can sense the abnormal electrical impact or provide key data for cross-validation, such as those related to the source node. The analysis focuses on adjacent terminals on the same power supply branch, terminals along the path of fault current flow, or terminals capable of monitoring voltage changes on the same busbar. It analyzes the initial self-healing control intent, identifying corresponding associated control relationships and control response relationships. Associated control relationships refer to terminal nodes that must directly operate their respective switches to execute the intent, i.e., the owners or controllers of the operated equipment. Control response relationships refer to terminal nodes whose operating status or power supply range will be affected after the intent is executed, such as load terminals that have their power restored or lost due to switch action, or line terminals that need to be monitored for overload due to power flow shift. This process determines the relevant intelligent fusion terminal nodes affected by the self-healing control strategy, forming a set of control-affected terminal nodes. Finally, the union of the topology-associated terminal nodes and the control-affected terminal node set is taken. All terminal nodes that can provide event-related data and are key parties or stakeholders in control actions are included in a unified collaborative group, forming a collaborative decision-making granularity circle for information exchange, permission negotiation, scheme generation and verification, ensuring high efficiency and high relevance in the decision-making process.

[0027] Furthermore, step S200 also includes obtaining the self-healing control permissions and control granularity strategies corresponding to the topology-associated terminal nodes and the control-affected terminal node set; determining the collaborative association granularity between each terminal node based on the self-healing control permissions and control granularity strategies of each terminal node; and using the collaborative association granularity to label the collaborative decision granularity circle with self-healing granularity rules, which is used to limit the self-healing control participation scope and control level of each intelligent fusion terminal within the collaborative decision granularity circle.

[0028] Preferably, all terminal nodes in the topology-associated terminal nodes and control-affected terminal node sets are identified. The predefined self-healing control permissions of each node are queried. These permissions are typically categorized based on factors such as the criticality and historical reliability of the controlled equipment, for example, allowing only reporting, allowing isolation operations, allowing transfer operations, or allowing full-function operations. Simultaneously, the control granularity strategy bound to each terminal node and its controlled equipment is obtained. This strategy details the set of actions allowed for the terminal when participating in self-healing control, the range of lines that can be intervened, the decision priority, and the voltage / current adjustment limits. Then, based on the self-healing control permissions and control granularity strategies of each terminal node, the degree of association and cooperation mode between any two nodes in this collaboration is analyzed and quantified through information interaction and comparison between nodes. The granularity of the collaboration is determined, including various node-to-node cooperation rules. For example, based on the level of permissions, one party is identified as the leading decision-making node and the other as a cooperating execution node, or both are equal negotiation nodes. The data level to be exchanged between the two parties is determined, and whether their control actions need to be strictly synchronized, sequentially executed, or independently executed. By utilizing collaborative correlation granularity to annotate the collaborative decision-making granularity circle with self-healing granularity rules, the collaborative behavior boundaries of each intelligent fusion terminal within the collaborative decision-making granularity circle are defined. This includes the scope of participation in self-healing control, which clarifies the behavior of each terminal in collaborative control. For example, node A can only provide monitoring data, node B can propose solutions but cannot make final decisions, and node C has the right to perform switching operations. It also defines the control level, which clarifies the role level and responsibility boundaries of each terminal in decision execution. For example, node D is responsible for security verification, node E is responsible for effect evaluation, and node F has the final execution triggering right of the solution.

[0029] Furthermore, step S200 also includes: determining the equipment risk parameters of each intelligent converged terminal based on the controllable distribution network equipment type, equipment operating status, and equipment risk level corresponding to each intelligent converged terminal; determining the permission level that each intelligent converged terminal is allowed to participate in self-healing control according to the equipment risk parameters and the impact tolerance level corresponding to the abnormal equipment events; and determining the control granularity strategy corresponding to each intelligent converged terminal according to the permission level and the impact tolerance level, which is used to limit the self-healing control participation depth and control action range of each intelligent converged terminal.

[0030] Preferably, the controllable distribution network equipment type corresponding to each intelligent converged terminal is determined, that is, the category of physical equipment controlled by the terminal, such as tie switches, sectionalizing switches, pole-mounted circuit breakers, load switches, and distributed power grid connection point controllers. Different types of equipment have different functions and importance in the power grid. The equipment operating status is determined, that is, the current real-time operating condition, such as normal service, operation with defects, maintenance shutdown, overload, and excessive temperature rise. The equipment risk level is determined, that is, the static risk level estimated based on the equipment's historical failure rate, health assessment, and location in the topology. By performing a weighted scoring calculation on the controllable distribution network equipment type, equipment operating status, and equipment risk level, quantitative equipment risk parameters are output. The higher the value, the greater the potential risk that the equipment controlled by the terminal may bring during operation.

[0031] Preferably, based on the direct risk level and scope of impact of the event, the impact tolerance level corresponding to the equipment abnormality event is determined, representing the tolerance of the current power grid condition for delayed response. Using the equipment risk parameters and impact tolerance level as inputs, the decision rules determine the level of authority that the terminal is allowed to participate in self-healing control under the current event. Specifically, for high-risk equipment with high tolerance, lower authority is granted to avoid accidental operation of high-risk equipment under high tolerance conditions; for low-risk equipment with low tolerance, higher authority is granted to enable rapid use of low-risk equipment for response in emergency situations; for high-risk equipment with low tolerance, conditional high authority may be granted or full authority may be granted in extremely urgent situations.

[0032] Preferably, based on the permission level and the impact tolerance level, a control granularity strategy corresponding to each intelligent fusion terminal is determined. This control granularity strategy is a refinement and operationalization of the permission level, used to limit the self-healing control participation depth of each intelligent fusion terminal. That is, it clarifies whether the terminal provides data, participates in negotiation, or has veto power in collaborative decision-making, and the scope of control actions, including at least allowing action types such as "closing", allowing voltage regulation range, current cut-off threshold, operation delay time and other action parameter boundaries, as well as intervention range restrictions. This achieves refined permission management, and achieves a dynamic optimal balance between ensuring grid security and pursuing self-healing speed, so as to improve the overall security, adaptability and efficiency of the distributed self-healing system.

[0033] Furthermore, step S200 also includes determining the direct risk level of the abnormal event based on the abnormal type, duration, and amplitude characteristics corresponding to the abnormal event; obtaining the impact range risk of the abnormal event based on the number of distribution network devices affected by the abnormal event, the load importance level, and the power supply impact range; and comprehensively determining the impact tolerance level of the abnormal event based on the direct risk level and the impact range risk, which characterizes the acceptable degree of distribution network operation safety and power supply reliability without immediately executing self-healing control operations.

[0034] Preferably, the system identifies the anomaly type corresponding to the equipment malfunction event. Different types have different initial hazard levels, such as three-phase short circuit > two-phase short circuit > single-phase grounding; voltage dips > voltage swells; equipment overheating > communication interruption; anomaly duration, i.e., the length of time the event lasts from its occurrence to the current moment. The longer the duration, the greater the cumulative risk of physical damage to the equipment and the greater the continuous damage to power quality; and anomaly amplitude characteristics, i.e., the severity of the event deviating from normal values, such as fault current multiples, voltage deviation percentages, and temperature exceedance degrees. The larger the amplitude, the stronger the instantaneous destructive force. By weighting and scoring the anomaly type, anomaly duration, and anomaly amplitude characteristics, the direct risk level of the anomaly event is calculated and determined, characterizing the severity of the event's origin.

[0035] Preferably, the number of distribution network devices affected by the abnormal event refers to the number of switches, lines, transformers, and other equipment directly affected by the abnormal event, reflecting the physical spread of the fault; the load importance level is the priority of the affected loads, which are pre-classified according to the nature of the users, with the consequences of power loss for important loads being more severe; the power supply impact range refers to the number of users, area, or total load capacity affected by power outages or substandard power quality, reflecting the scale of social and economic impact; by weighting and scoring the number of distribution network devices affected by the abnormal event, the load importance level, and the power supply impact range, the risk of the impact range of the abnormal event is determined.

[0036] Preferably, a weighted comprehensive analysis of the direct risk level and the scope of impact risk is performed to determine the impact tolerance level of an abnormal event. This level represents the time window and safety margin under the current abnormal event, allowing for "waiting" or "tolerance" without immediate mandatory self-healing control operations. It also reflects the acceptability of distribution network operation safety and power supply reliability. A low impact tolerance level corresponds to high urgency, indicating that the abnormal event itself is very serious and / or has a wide impact, and safety or power supply reliability is rapidly deteriorating, requiring immediate or rapid activation of self-healing control. A high tolerance level corresponds to low urgency, indicating that the abnormal event is relatively isolated, minor, or has a controllable impact, allowing relatively ample time for more thorough analysis, consultation, or waiting for human intervention, without the need to immediately trigger automatic control. This enables differentiated and optimized handling of events with different levels of urgency.

[0037] Furthermore, step S200 also includes the following: the self-healing control permission levels include a first permission level that allows the execution of abnormal isolation control; a second permission level that allows the execution of abnormal isolation and load transfer control; and a third permission level that allows the execution of abnormal isolation, load transfer and tie switch closing control.

[0038] Preferably, the self-healing control authority levels include a first authority level, a second authority level, and a third authority level. The first authority level allows for abnormal isolation control, which electrically disconnects the abnormal equipment or line segment from the power grid. For example, it disconnects circuit breakers, load switches, or sectionalizing switches directly controlled by the terminal and located upstream or downstream of the abnormal point, achieving electrical isolation of the fault or abnormal area. The second authority level allows for abnormal isolation and load transfer control, which simultaneously operates the sectionalizing switch or load switch controllable by the terminal on the non-fault path, transferring the load of the non-fault-affected section downstream of the fault point to other healthy feeders or power sources on the same busbar or in the same substation, thereby partially restoring power supply. The third authority level allows for abnormal isolation, load transfer, and tie switch closing control, which simultaneously operates the tie switch controllable by the terminal, connecting the fault-affected area to an adjacent, normal feeder, achieving a wider range of load transfer and power restoration. The tie switch is a switch connecting feeders of different substations or main transformers, and is normally in the open state.

[0039] In step S300, within the collaborative decision-making granularity circle, each intelligent fusion terminal, based on its node role, status information, and the abnormal evidence, dynamically allocates permissions and negotiates collaborative relationships through information interaction, jointly judges the preliminary self-healing control intention, and generates at least one candidate self-healing control scheme.

[0040] Step S300 further includes: verifying the abnormal evidence using the monitoring distribution network information of the topology-associated terminal nodes to determine the verification result of the abnormal evidence; based on the abnormal evidence that has passed the verification result, making a consistency judgment and correction on the preliminary self-healing control intention; under the constraint of the corrected self-healing control intention, and combining the permission level and control granularity strategy corresponding to each intelligent fusion terminal within the collaborative decision granularity circle, dynamically allocating self-healing control permissions and negotiating collaborative relationships through information interaction to generate at least one candidate self-healing control scheme.

[0041] Preferably, within the collaborative decision-making granularity circle, each intelligent fusion terminal performs data cross-verification of anomaly evidence through the monitoring distribution network information of the topology-related terminal nodes, including comparing relevant data from nodes at different spatial locations. For example, when verifying overcurrent at point M, it simultaneously checks whether the current protection of its upstream node N is activated, whether the current of its downstream node O drops sharply, and whether the voltage of the adjacent line node P dips, outputting the verification result of the anomaly evidence. It may pass, meaning that multiple independent data sources support the anomaly evidence, increasing the confidence level; some data may not support it, possibly indicating that the evidence type or location is incorrect, such as a ground fault being falsely reported as a phase-to-phase fault, weakening and correcting the confidence level; it may fail, meaning that most key data do not support it, possibly due to terminal false alarms or communication interference. Using the abnormal evidence from the verification results, the initial self-healing control intention is consistent and corrected. Specifically, it is analyzed whether the goals and actions in the initial self-healing control intention correspond reasonably with the abnormal evidence after verification. If they are inconsistent, the intention is adjusted based on more reliable evidence. For example, if the fault is found to be downstream of point I after mutual verification, the intention is corrected to "isolate the downstream switch of point I". Or if the fault type is found to be transient, the intention may be corrected from "permanent isolation" to "trial power supply".

[0042] Preferably, under the constraints of the modified self-healing control intent, the dynamic allocation of self-healing control permissions and negotiation of collaborative relationships are carried out through information interaction. That is, by combining the permission level and control granularity strategy of each intelligent fusion terminal within the collaborative decision granularity circle, the final role and responsibility relationship in the specific task are dynamically confirmed through information interaction. For example, a terminal with the third permission level may be nominated as the "leading decision node", and a terminal with the second permission and controlling the key transfer switch is identified as the "key execution node". Then, under the guidance of the clear role relationship and the modified intent, each intelligent fusion terminal constructs local suggestions based on the status of the controlled equipment, local load information, etc. The leading node or through the negotiation mechanism integrates and generates multiple candidate self-healing control schemes. The content of these schemes clearly defines the actions and timing of each intelligent fusion terminal operating the specific equipment to achieve the modified self-healing intent, thereby ensuring reliable and flexible collaborative decision-making.

[0043] Step S400: The candidate self-healing control scheme is verified for security and feasibility. After passing the verification, it is sent to an intelligent fusion terminal with corresponding permissions to collaboratively execute the self-healing control scheme.

[0044] Step S400 further includes: performing a safety prediction and verification of the operational risks that may be caused by the execution of the candidate self-healing control scheme based on the current operating state parameters; performing a feasibility verification of the execution reachability of the candidate self-healing control scheme based on the execution node state and control constraints; determining the execution timing of the candidate self-healing control scheme based on the expected self-healing effect of the candidate self-healing control scheme and the tolerance of the impact of abnormal events; and allowing the execution of the candidate self-healing control scheme when the verification result meets the preset execution conditions.

[0045] Preferably, based on the current power grid operating status parameters, including real-time power flow distribution, voltage level, equipment load rate, protection settings, etc., the safety of candidate self-healing control schemes is verified. That is, the network state after the candidate self-healing control scheme is executed is rapidly calculated or simulated to predict whether new safety risks will be generated after execution, such as whether it will cause any line / transformer overload, whether it will cause voltage overrun at any node, whether it may cause protection maloperation or failure to operate, and then output the safety assessment result, such as safe, risky, or unsafe.

[0046] Preferably, based on the real-time readiness status of the terminal and controlled equipment specified in the scheme for performing the operation, such as whether the communication is normal, whether the control loop is intact, whether the energy storage mechanism has stored energy, whether there is a local interlocking signal, and the physical and logical conditions that each operation instruction in the scheme must meet, such as confirming that the voltage difference and phase angle difference on both sides are within the allowable range before the closing operation, and confirming that its load current is lower than the disconnection capacity before the sectionalizing switch operation, the feasibility of the candidate self-healing control scheme is verified. This includes checking each operation step of the candidate self-healing control scheme, confirming whether its execution terminal has the current execution capability and meets all the preconditions, and then determining whether the candidate self-healing control scheme is executed correctly, avoiding issuing commands that cannot be executed or will result in an error upon execution, and then outputting the feasibility assessment result, such as fully feasible, some operations are temporarily not feasible or not feasible.

[0047] Preferably, the execution timing of candidate self-healing control schemes is determined based on their expected self-healing effect and the tolerance of abnormal events. The expected self-healing effect refers to the expected power supply range, load, and time required to restore the grid state to a safe level after the scheme is executed. Specifically, the expected self-healing effect of candidate self-healing control schemes is matched with the tolerance of abnormal events. If the tolerance of abnormal events is low, indicating a very urgent situation, a candidate self-healing control scheme with fast execution speed and the ability to quickly contain risks is prioritized. If the tolerance of abnormal events is high, indicating a less urgent situation, a candidate self-healing control scheme with more complex execution steps but better recovery results or waiting for better conditions is selected, and a timing suggestion is output, such as immediate execution, delaying execution until a certain condition is met, or waiting for manual confirmation. When all verification results meet the preset execution conditions, such as "safe," "fully feasible," and "immediate execution," the candidate self-healing control scheme is allowed to be sent to an intelligent fusion terminal with appropriate permissions for collaborative execution, thereby greatly reducing the risk of erroneous operation and improving the reliability and robustness of the entire grid self-healing process.

[0048] Furthermore, step S400 also includes, after executing the self-healing control scheme, performing state verification on the self-healing control execution result. When the verification result does not meet the preset operational safety conditions, performing self-healing control rollback processing, wherein the rollback processing includes at least one of the following: canceling the executed self-healing control operation to restore the distribution network operation state to the safe state before execution; prohibiting the continued execution of subsequent control actions in the self-healing control scheme; reducing the self-healing control permission level that can be granted in subsequent self-healing control processes or narrowing the collaborative decision-making granularity circle based on the rollback result; and sending abnormal evidence, execution results, and rollback information to the superior system or entering the manual intervention process.

[0049] Preferably, after executing the self-healing control scheme, the status verification of the self-healing control execution result is performed through monitoring data from the intelligent fusion terminal, and the verification result is compared with the preset operational safety conditions, including but not limited to whether critical lines / equipment are overloaded, whether node voltages are within the allowable range, whether fault or abnormal signals have disappeared, and whether the expected power supply restoration has been achieved. When the verification result does not meet the preset operational safety conditions, it is determined that "the execution effect does not meet safety expectations" or "it has caused secondary problems," and self-healing control rollback processing is performed. The rollback processing includes at least one of the following: canceling the executed self-healing control operation, i.e., automatically issuing a sequence command opposite to the executed operation to restore the distribution network operation status to the safe state before execution; immediately freezing or canceling the issuance of instructions for all subsequent control actions that have not yet been executed in the self-healing control scheme, and prohibiting... The system will stop executing subsequent control actions in the self-healing control scheme; based on the rollback results, the level of self-healing control permissions that can be granted in subsequent self-healing control processes will be reduced or the granularity of collaborative decision-making will be narrowed. For example, for intelligent fusion terminals participating in the failed execution or in similar scenarios, a more conservative permission granting strategy will be adopted. In future collaborative decision-making, a smaller and more core decision-making circle will be formed to reduce collaborative complexity and improve reliability; abnormal evidence, execution results, and rollback information will be sent to the superior system or enter the manual intervention process. For example, they will be sent to the distribution network master station or regional monitoring center to trigger experts or operators of the superior system to conduct in-depth analysis, decision-making, and remote intervention, or directly switch to the manual processing process and be taken over by the dispatcher. This will greatly improve the resilience and reliability of the entire self-healing system and ensure its reliable operation in complex real-world environments.

[0050] In the above text, refer to Figure 1 This paper describes in detail a distributed collaborative decision-making method for intelligent converged terminals for distribution network self-healing according to embodiments of the present invention. Next, reference will be made to... Figure 2 This invention describes a distributed collaborative decision-making system for distribution network self-healing using intelligent converged terminals, according to embodiments of the present invention.

[0051] The intelligent converged terminal distributed collaborative decision-making system for distribution network self-healing according to embodiments of the present invention addresses the technical problems in existing technologies, such as static and rigid collaborative decision-making scope, inflexible allocation of permissions and roles, and lack of distributed solution verification mechanisms, leading to insufficient reliability and adaptability of distribution network self-healing. It achieves the technical effects of dynamically adapting to the impact of abnormal events, optimizing decision-making efficiency, and improving the response speed, adaptability, and reliability of distribution network self-healing. Figure 2 As shown, the intelligent fusion terminal distributed collaborative decision-making system for distribution network self-healing includes: an abnormal evidence generation module 10, a collaborative decision granularity circle formation module 20, a self-healing control intent judgment module 30, and a control scheme verification module 40.

[0052] An anomaly evidence generation module 10 is used to interact with multiple intelligent fusion terminals deployed at various nodes of the distribution network to obtain monitoring and identification information of each intelligent fusion terminal. When any intelligent fusion terminal detects a local anomaly event, it generates anomaly evidence with confidence and establishes a preliminary self-healing control intention. A collaborative decision granularity circle formation module 20 is used to determine the set of intelligent fusion terminals participating in this collaborative decision based on the topological relationship of the multiple intelligent fusion terminals and the intelligent fusion terminal nodes affected by the anomaly event and the preliminary self-healing control intention, forming a collaborative decision granularity circle. A self-healing control intention judgment module 30 is used to, within the collaborative decision granularity circle, each intelligent fusion terminal, based on its node role, status information, and the anomaly evidence, dynamically allocates permissions and negotiates collaborative relationships through information interaction to jointly judge the preliminary self-healing control intention and generate at least one candidate self-healing control scheme. A control scheme verification module 40 is used to verify the security and feasibility of the candidate self-healing control scheme, and after passing the verification, sends it to the intelligent fusion terminal with the corresponding permissions to collaboratively execute the self-healing control scheme.

[0053] The specific configuration of the collaborative decision-making granularity circle formation module 20 will be described in detail below. The collaborative decision-making granularity circle formation module 20 further includes: based on the intelligent fusion terminal nodes where the local abnormal event occurred and the data flow relationship reflected by the abnormal evidence, performing correlation and location within the topological relationship to determine the topologically associated terminal nodes; determining the relevant intelligent fusion terminal nodes affected by the self-healing control strategy according to the correlation control relationship and control response relationship corresponding to the preliminary self-healing control intention, forming a control-affected terminal node set; and constructing the collaborative decision-making granularity circle based on the topologically associated terminal nodes and the control-affected terminal node set.

[0054] The specific configuration of the collaborative decision-making granularity circle formation module 20 will be described in detail below. The collaborative decision-making granularity circle formation module 20 further includes: acquiring the self-healing control permissions and control granularity strategies corresponding to the topology-associated terminal nodes and the control-affected terminal node set; determining the collaborative association granularity between each terminal node based on its self-healing control permissions and control granularity strategies; and using the collaborative association granularity to label the collaborative decision-making granularity circle with self-healing granularity rules, thereby limiting the self-healing control participation scope and control level of each intelligent fusion terminal within the collaborative decision-making granularity circle.

[0055] The following will describe in detail the specific configuration of the collaborative decision-making granularity circle formation module 20. The collaborative decision-making granularity circle formation module 20 further includes: determining the equipment risk parameters of each intelligent fusion terminal based on the controllable distribution network equipment type, equipment operating status, and equipment risk level corresponding to each intelligent fusion terminal; determining the permission level that each intelligent fusion terminal is allowed to participate in self-healing control based on the equipment risk parameters and the impact tolerance level corresponding to the equipment abnormal events; and determining the control granularity strategy corresponding to each intelligent fusion terminal based on the permission level and the impact tolerance level, used to limit the self-healing control participation depth and control action range of each intelligent fusion terminal.

[0056] The following will describe in detail the specific configuration of the collaborative decision-making granularity circle formation module 20. The collaborative decision-making granularity circle formation module 20 further includes: determining the direct risk level of an abnormal event based on the abnormal type, duration, and amplitude characteristics corresponding to the equipment abnormal event; obtaining the impact range risk of the abnormal event based on the number of distribution network devices affected, the load importance level, and the power supply impact range; and comprehensively determining the impact tolerance level of the abnormal event based on the direct risk level and the impact range risk, characterizing the acceptable level of distribution network operation safety and power supply reliability without immediately executing self-healing control operations.

[0057] The specific configuration of the collaborative decision-making granularity circle forming module 20 will be described in detail below. The collaborative decision-making granularity circle forming module 20 further includes: the self-healing control permission levels include a first permission level that allows the execution of abnormal isolation control; a second permission level that allows the execution of abnormal isolation and load transfer control; and a third permission level that allows the execution of abnormal isolation, load transfer, and tie switch closing control.

[0058] The specific configuration of the self-healing control intent judgment module 30 will be described in detail below. The self-healing control intent judgment module 30 further includes: performing data cross-verification of the abnormal evidence using monitoring distribution network information from topology-associated terminal nodes to determine the verification result of the abnormal evidence; based on the abnormal evidence that passes verification, performing consistency judgment and correction on the preliminary self-healing control intent; under the constraints of the corrected self-healing control intent, and combining the permission levels and control granularity strategies corresponding to each intelligent fusion terminal within the collaborative decision granularity circle, dynamically allocating self-healing control permissions and negotiating collaborative relationships through information interaction to generate at least one candidate self-healing control scheme.

[0059] The specific configuration of the control scheme verification module 40 will be described in detail below. The control scheme verification module 40 further includes: performing safety prediction verification based on current operating state parameters to assess the operational risks that may arise after the execution of a candidate self-healing control scheme; performing feasibility verification based on the execution node status and control constraints to assess the execution reachability of the candidate self-healing control scheme; determining the execution timing of the candidate self-healing control scheme based on its expected self-healing effect and the tolerance of abnormal events; and allowing the execution of the candidate self-healing control scheme when the verification results meet preset execution conditions.

[0060] The specific configuration of the control scheme verification module 40 will be described in detail below. The control scheme verification module 40 further includes: after executing the self-healing control scheme, performing state verification on the self-healing control execution result; when the verification result does not meet the preset operational safety conditions, performing self-healing control rollback processing, wherein the rollback processing includes at least one of the following: canceling the executed self-healing control operation to restore the distribution network operation state to the safe state before execution; prohibiting the continued execution of subsequent control actions in the self-healing control scheme; reducing the self-healing control permission level that can be granted in subsequent self-healing control processes or narrowing the collaborative decision-making granularity circle based on the rollback result; and sending abnormal evidence, execution results, and rollback information to the superior system or initiating a manual intervention process.

[0061] The intelligent converged terminal distributed collaborative decision-making system for distribution network self-healing provided in this embodiment of the invention can execute the intelligent converged terminal distributed collaborative decision-making method for distribution network self-healing provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0062] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for intelligent fusion terminal distributed collaborative decision-making for network self-healing, characterized in that, The method comprises the steps of: interacting with the multi-intelligent fusion terminals deployed at each node of the power distribution network to obtain monitoring and identification information of each intelligent fusion terminal, generating abnormal evidence with a confidence level when any intelligent fusion terminal detects a local abnormal event, and establishing a preliminary self-healing control intention; determining a set of intelligent fusion terminals participating in the current collaborative decision based on the topological relationship of the multi-intelligent fusion terminals, and combining the abnormal event and the intelligent fusion terminal nodes affected by the preliminary self-healing control intention to form a collaborative decision granularity circle; in the collaborative decision granularity circle, each intelligent fusion terminal performs dynamic allocation of authority and negotiation of collaborative relationship based on its node role, state information and the abnormal evidence through information interaction, jointly judges the preliminary self-healing control intention, and generates at least one candidate self-healing control scheme; verifying the safety and feasibility of the candidate self-healing control scheme, and sending the self-healing control scheme to the intelligent fusion terminals with corresponding authority for collaborative execution after verification; determining a set of intelligent fusion terminals participating in the current collaborative decision to form a collaborative decision granularity circle, comprising: based on the intelligent fusion terminal node where the local abnormal event occurs and the data flow relationship reflected by the abnormal evidence, the topological relationship is associated and positioned to determine the topological associated terminal node; determining the related intelligent fusion terminal nodes affected by the self-healing control strategy according to the associated control relationship and control response relationship corresponding to the preliminary self-healing control intention to form a control affected terminal node set; constructing the collaborative decision granularity circle according to the topological associated terminal node and the control affected terminal node set; constructing the collaborative decision granularity circle further comprises: obtaining the self-healing control authority and control granularity strategy corresponding to the topological associated terminal node and the control affected terminal node set; determining the collaborative association granularity between each terminal node based on the self-healing control authority and control granularity strategy of each terminal node; using the collaborative association granularity to mark the self-healing granularity rule of the collaborative decision granularity circle for limiting the self-healing control participation range and control level of each intelligent fusion terminal in the collaborative decision granularity circle.

2. The intelligent fusion terminal distributed collaborative decision method for network configuration healing according to claim 1, characterized in that, obtaining the self-healing control authority and control granularity strategy corresponding to the topological associated terminal node and the control affected terminal node set comprises: determining the device risk parameters of each intelligent fusion terminal based on the controllable power distribution equipment type, equipment operating state and equipment risk level corresponding to each intelligent fusion terminal; determining the permission level of each intelligent fusion terminal allowed to participate in self-healing control according to the device risk parameters and the influence tolerance level corresponding to the device abnormal event; determining the control granularity strategy corresponding to each intelligent fusion terminal according to the permission level and the influence tolerance level, for limiting the self-healing control participation depth and control action range of each intelligent fusion terminal.

3. The intelligent fusion terminal distributed collaborative decision-making method for network configuration self-healing according to claim 2, characterized in that, determining the influence tolerance level corresponding to the device abnormal event comprises: determining the direct risk degree of the abnormal event based on the abnormal type, abnormal duration and abnormal amplitude characteristics corresponding to the device abnormal event; According to the number of distribution network devices involved in the abnormal event, the load importance level, and the power supply influence range, an influence range risk of the abnormal event is obtained; According to the direct risk degree and the influence range risk, an influence tolerance level of the abnormal event is comprehensively determined, which represents an acceptable degree of power distribution network operation safety and power supply reliability without immediately performing a self-healing control operation.

4. The intelligent fusion terminal distributed collaborative decision method for network configuration healing according to claim 2, characterized in that, The self-healing control permission level includes: a first permission level allowing to perform abnormal isolation control; a second permission level allowing to perform abnormal isolation and load transfer control; and a third permission level allowing to perform abnormal isolation, load transfer, and tie switch closing control.

5. The intelligent fusion terminal distributed collaborative decision method for network configuration healing according to claim 3, characterized in that, Within the collaborative decision granularity circle, each intelligent fusion terminal performs dynamic permission allocation and collaborative relationship negotiation based on its node role, state information, and the abnormal evidence, jointly judges the preliminary self-healing control intention, and generates at least one candidate self-healing control scheme, including: The abnormal evidence is data-interrogated by monitoring distribution network information of topologically associated terminal nodes to determine a verification result of the abnormal evidence; The preliminary self-healing control intention is consistency judged and corrected based on the passed abnormal evidence of the verification result; Under the constraint of the corrected self-healing control intention, in combination with the permission level and control granularity strategy of each intelligent fusion terminal within the collaborative decision granularity circle, dynamic allocation of self-healing control permission and negotiation of collaborative relationship are performed through information interaction to generate at least one candidate self-healing control scheme.

6. The intelligent fusion terminal distributed collaborative decision method for network configuration healing according to claim 1, characterized in that, The candidate self-healing control scheme is verified for safety and feasibility, including: Safety prediction verification is performed on a possible operation risk caused by the candidate self-healing control scheme after execution based on current operation state parameters; Feasibility verification is performed on execution reachability of the candidate self-healing control scheme based on execution node state and control constraint; Execution timing of the candidate self-healing control scheme is judged based on expected self-healing effect of the candidate self-healing control scheme and influence tolerance of the abnormal event; When the verification result meets a preset execution condition, the candidate self-healing control scheme is allowed to be executed.

7. The intelligent fusion terminal distributed collaborative decision method for network configuration healing according to claim 1, characterized in that, Further including: After the self-healing control scheme is executed, state verification is performed on a self-healing control execution result, and when the verification result does not meet a preset operation safety condition, a self-healing control rollback processing is performed, wherein the rollback processing includes at least one of: canceling the executed self-healing control operation to restore the power distribution network operation state to a safe state before execution; Inhibiting subsequent control actions in the self-healing control scheme from being continuously executed; reducing a self-healing control permission level or narrowing a collaborative decision granularity circle in a subsequent self-healing control process based on a rollback result; uploading the abnormal evidence, the execution result, and the rollback information to a superior system or entering an artificial intervention process.

8. A smart fusion terminal distributed collaborative decision system for network self-healing, characterized in that, The system is used to implement the intelligent fusion terminal distributed collaborative decision method for distribution network self-healing according to any one of claims 1 to 7, and the system includes: An abnormality evidence generation module is configured to be interactively deployed in the multi-intelligent fusion terminals of each node of the power distribution network, to obtain monitoring and identification information of each intelligent fusion terminal, to generate abnormality evidence with a confidence level when any intelligent fusion terminal detects a local abnormal event, and to establish a preliminary self-healing control intention; A collaborative decision granularity circle formation module is configured to determine a set of intelligent fusion terminals participating in the current collaborative decision based on the topological relationship of the multi-intelligent fusion terminals, in combination with the abnormal event and the intelligent fusion terminal nodes affected by the preliminary self-healing control intention, and to form a collaborative decision granularity circle; A self-healing control intention judgment module is configured to, within the collaborative decision granularity circle, perform dynamic allocation of authority and negotiation of collaborative relationship through information interaction based on the node role, state information and the abnormality evidence of each intelligent fusion terminal, to jointly judge the preliminary self-healing control intention, and to generate at least one candidate self-healing control scheme; A control scheme verification module is configured to perform safety and feasibility verification on the candidate self-healing control scheme, and to send the self-healing control scheme to the intelligent fusion terminals with corresponding authority for collaborative execution after passing the verification.

Images