An intelligent distribution hierarchical self-healing fault isolation and power supply restoration method

Abstract

The present invention discloses an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method, which is applicable to quickly isolating faults and restoring power supply to non-fault areas when faults occur in the distribution system. This method adopts a hierarchical control architecture. The intelligent switch devices at the lower layer are responsible for real-time collecting the line current and voltage states, and immediately completing local identification and cutting off the fault section after detecting the preset fault signal. Subsequently, the fault information is uploaded to the distribution automation master station at the upper layer. The master station combines the current network topology analysis to restore the power supply area and generate a network reconfiguration scheme, and switches the non-fault area to the standby power supply path through controlling the tie switch. The system supports retrospective analysis of historical data such as waveform distortion and partial discharge signals before the fault, identifies potential abnormal trends and optimizes the operation strategy to achieve the adaptive early warning function. This method has the advantages of rapid response, clear structure and strong autonomy, and can significantly improve the reliability and continuity of the distribution system.

Description

Technical Field

[0001] The present invention relates to the field of power technology, and particularly to an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method. Background Art

[0002] In the prior art, the distribution network usually relies on the distribution automation system to achieve fault detection, isolation and power supply restoration. Most of such systems adopt a centralized control structure. The master station analyzes the fault location through the data of the acquisition terminal, and issues operation instructions to the on-site switch equipment to complete fault isolation and network reconstruction. In addition, some systems integrate feeder automation or small-scale self-healing functions, and can achieve fault response and restoration control in a local area under specific conditions, improving power supply reliability.

[0003] However, the prior art has problems of slow response speed, insufficient information closed-loop and single restoration strategy when dealing with complex topological structures or multi-point faults. The centralized control method has a high dependence on communication. Limited by data transmission and processing delays, it is difficult to complete large-scale adaptive power supply restoration in a short time. At the same time, the lack of an effective operation status early warning mechanism results in the system being unable to identify potential risks in time before a fault, restricting the further improvement of the self-healing ability.

[0004] Therefore, there is an urgent need to propose an intelligent distribution fault isolation and power supply restoration method with faster response, more flexible structure and adaptive early warning ability. Summary of the Invention

[0005] The present application provides an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method to improve the reliability and continuity of the distribution system.

[0006] The present application provides an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method, including:

[0007] Establish a hierarchical control architecture in the distribution system, where the hierarchical control architecture includes intelligent switch devices at the lower layer and a distribution automation master station at the upper layer;

[0008] The intelligent switch devices collect the current and voltage states of the connected lines. When a signal that meets the preset fault criterion is detected, the fault section is identified locally and the corresponding line section is cut off to achieve fault isolation at the first time;

[0009] The intelligent switch devices upload the fault information and key electrical data to the distribution automation master station. The distribution automation master station combines the network topology information to judge the power supply recoverable area and dynamically generates a reconstruction plan;

[0010] The distribution automation master station controls a group of tie switches to perform operations according to the reconstruction plan, and switches the non-fault area to other power supply paths to achieve power supply restoration;

[0011] While performing power supply restoration, retrospectively analyze the current waveform, harmonic characteristics, or partial discharge signal before the fault occurred. If an abnormal trend is identified, record the abnormal event and update the operation strategy for early warning discrimination of future similar faults.

[0012] The beneficial effects of the technical solution provided by this application include:

[0013] (1) By establishing a hierarchical control architecture, realizing the coordinated cooperation between intelligent switch devices and the distribution automation master station, it can quickly complete fault identification and isolation locally, effectively shorten the fault response time, and improve the immediate processing ability of the distribution system. (2) Using the automation master station for network topology analysis and reconfiguration path generation, quickly restore power supply to the non-fault area without manual intervention, significantly reduce the power outage scope and duration, and ensure power supply continuity. (3) Introducing a retrospective analysis mechanism for electrical abnormal signals before the fault, enabling the system to have an adaptive early warning ability, being able to identify potential risks in advance and optimize the operation strategy, thereby improving the operation safety and fault resistance ability of the distribution network. (4) The entire method has highly automated characteristics, can achieve distributed cooperative control and flexible power supply path switching under various complex working conditions, and improve the intelligent level and operation efficiency of the system. Brief Description of the Drawings

[0014] Figure 1 is a flowchart of an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided by the first embodiment of this application. Detailed Embodiment

[0015] Many specific details are set forth in the following description in order to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar generalizations without departing from the connotation of this application. Therefore, this application is not limited by the specific embodiments disclosed below.

[0016] The first embodiment of this application provides an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method. Please refer to Figure 1 , which is a schematic diagram of the first embodiment of this application. The following combines Figure 1 to describe in detail an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided by the first embodiment of this application.

[0017] Step S101: Establish a hierarchical control architecture in the distribution system, where the hierarchical control architecture includes intelligent switch devices at the lower layer and a distribution automation master station at the upper layer.

[0018] In the intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided in this embodiment, step S101 is the basic preparation step for the entire method, aiming to construct a distribution system structure with hierarchical perception, rapid response, and coordinated control capabilities to support the efficient execution of subsequent fault handling and power supply restoration operations. The following provides a detailed description of step S101 as follows.

[0019] The "establishment of a hierarchical control architecture in the distribution system" described in step S101 means that in an existing or newly built medium-voltage or low-voltage distribution network, according to the differences in function division and information processing capabilities, the control architecture is divided into at least two logical levels: the lower layer is the on-site intelligent control layer, and the upper layer is the centralized coordinated control layer, so as to achieve the functional decoupling and distributed deployment of data acquisition, fault detection, decision-making analysis, and control execution.

[0020] The intelligent switch devices in the lower layer refer to the power switch devices installed at each key node (such as feeder switches, branch switches, ring main units, etc.). This device has the function of real-time acquisition of basic electrical parameters such as current and voltage, and is built with edge computing capabilities and fault recognition logic, and can independently complete fault feature recognition and opening operations. Preferably, this type of device supports the IEC 61850 communication protocol and is equipped with components such as a breaker control unit (BCU), current transformer (CT), voltage transformer (PT), and programmable logic controller (PLC). The sampling frequency should not be lower than 5 kHz to ensure the accurate capture of fast-changing events such as short circuits and overloads. This device should support at least one digital signal processor (DSP) or an equivalent embedded computing unit to implement the operation of local fault recognition algorithms, such as short-circuit judgment based on the current mutation rate ΔI / Δt or zero-sequence current analysis.

[0021] The distribution automation master station in the upper layer is usually located in the dispatching center or control room and is composed of a high-performance server or an industrial control computing platform. It runs the distribution automation master station software and has the capabilities of modeling the whole network topology structure, communication management, generating fault handling strategies, and controlling the reconstruction of the power supply path. The master station maintains a stable connection with each intelligent switch device through a communication network (such as fiber optic Ethernet, wireless public network, LoRa, NB-IoT, etc.). The master station system needs to integrate a network topology diagram, SCADA interface, real-time database, and expert control module, and support telemetry, tele-signaling, and remote control operations of external devices.

[0022] The specific process of establishing the hierarchical control architecture includes the following implementation steps. First, digital modeling of the distribution network structure is carried out, including the geographical information and electrical connection relationships of each feeder, node, substation, and switch location. Second, intelligent switch devices are installed at each key node, and on-site commissioning and communication network access are completed. Subsequently, the corresponding distribution network topology map and device address information are loaded into the master station system, and a fault response strategy template is configured. The configuration of message format, communication cycle, heartbeat mechanism, encryption protocol, etc. should be completed between the master station and each intelligent device to ensure data security and real-time performance.

[0023] The hierarchical control architecture established in the above manner not only has the front-end distributed acquisition and local response capabilities, but also has the global coordination and control capabilities of the central end, and can realize a collaborative control mode in which information is concentrated upward, decisions are distributed downward, and execution is at the edge. The setting of this architecture provides a basic support for subsequent steps such as fault detection, isolation, reconstruction, and early warning, ensuring that the entire system has significant advantages in terms of response speed, processing accuracy, and operation reliability.

[0024] In summary, step S101 not only requires the establishment of a hierarchical structure at the hardware deployment level, but also emphasizes the hierarchical division of functional logic and control processes.

[0025] Furthermore, establishing a hierarchical control architecture in the distribution system includes:

[0026] In the initialization stage, the distribution automation master station generates a logical hierarchical structure of multi-level power supply areas based on the geographical information system GIS and historical operation topology data, and assigns corresponding switch device addresses, fault criterion templates, and communication strategies to each level of nodes, realizing the synchronous matching of topology structure and functional configuration;

[0027] Through the issued configuration information, the operating roles and action permissions of each intelligent switch device are synchronized, so that they have different levels of fault response priorities, coordination waiting times, and reporting window parameters according to the network levels they are in, so as to achieve adaptive collaborative response among multiple devices in a concurrent fault scenario;

[0028] After the hierarchical structure deployment of the distribution automation master station is completed, a chain mutual recognition operation is performed on each group of intelligent switch devices in the partition, so that adjacent switch devices complete identity authentication and communication handshake locally, establishing the self-organizing boundary recognition ability between devices, in order to support regional autonomous fault handling in the offline state of the master station.

[0029] In this embodiment, further limitations on how to establish a hierarchical control architecture are proposed. The key lies in a series of configurations and collaborations in the system initialization stage, so that the distribution system not only has hierarchical control capabilities, but also has structured, self-organizing, and adaptive operating characteristics.

[0030] Before the power distribution system is put into operation, it first enters the initialization stage, which is led by the main station of distribution automation. Using the built-in Geographic Information System (GIS) platform and the historical operation topology data accumulated over a long time, it conducts hierarchical modeling of the entire distribution network. GIS data usually includes spatial location information of substations, switch stations, feeders, branch lines, load points, communication nodes, etc., while topology data reflects electrical connection relationships, breaker states, switch logic, and historical load paths. Based on this information, the main station automatically divides multiple levels of power supply area logical levels, such as substation layer, main feeder layer, branch line layer, terminal layer, etc., with reference to area division, feeder attribution, transformer nodes, load density, etc.

[0031] After the definition of the hierarchical structure is completed, the main station will assign exclusive configuration parameters to the key nodes in each level. These parameters include the unique address code (such as the logical node LN identifier) bound to the intelligent switch device, the preset fault criterion template (such as short-circuit current threshold, duration judgment window, waveform distortion characteristics, etc.), and the communication mechanism (such as reporting period, receiving window size, congestion control strategy, etc.). The above configuration not only ensures the accurate logical positioning of each intelligent switch device in the structure but also forms a matching relationship with the overall system scheduling strategy in terms of function allocation.

[0032] The configuration information generated by the main station will be sent to each level of intelligent switch devices through the communication network (such as fiber optic Ethernet, wireless public network, industrial-grade 5G, etc.). After receiving it, the switch devices will automatically load the configuration content and update the local parameters, thus forming a multi-level division of labor and collaborative operation mode. Specifically, the intelligent switch devices at the upper level will obtain a higher fault response priority and can take the lead in judging and isolating the local fault section in case of multiple-point faults occurring simultaneously; while the lower-level devices need to collect the status information of neighboring devices during the coordination waiting time before making action decisions. In addition, the reporting windows of each device can also be set according to the level. For example, the main feeder nodes report more frequently to ensure the synchronization of main path information, while the branch line devices use the event-triggered method to reduce communication pressure.

[0033] After all the configurations are sent and confirmed to be effectively deployed, the main station further starts the chain mutual recognition mechanism to enhance the self-organization control ability within each partition. This operation is carried out by the main station sending a mutual recognition start instruction to trigger the adjacent intelligent switch devices in the same area to sequentially execute the identity authentication and communication handshake process. The identity authentication is based on device identification, key verification, or configuration version consistency verification, and the communication handshake establishes a local low-latency communication channel, such as using RS-485, CAN bus, or a dedicated wireless frequency hopping link, to achieve local status exchange without central scheduling.

[0034] Through the chained mutual recognition operation, each group of intelligent switch devices will form an adaptive small autonomous domain. In the case of the interruption of the main station communication link or the offline operation of the main station, these autonomous domains can, relying on the established collaborative relationships among devices, independently judge local faults and execute corresponding isolation and recovery strategies, thus significantly improving the system's anti-communication fault ability and autonomous response efficiency.

[0035] In summary, the steps from topology modeling to parameter configuration and then to autonomous mutual recognition form a highly adaptive, reasonably structured hierarchical control architecture with self-organization ability. This architecture not only breaks through the over-reliance on the main station in traditional distribution automation systems but also effectively supports the distributed response and coordinated control among lower-layer intelligent devices, significantly enhancing the self-healing ability, response speed, system elasticity, etc. of the present invention.

[0036] Furthermore, the chained mutual recognition operation includes:

[0037] After receiving the chained mutual recognition instruction issued by the distribution automation main station, the intelligent switch device conducts mutual recognition negotiation with other intelligent switch devices physically adjacent to it through a low-latency local communication protocol, and determines the fault response priority order of each intelligent switch device in the link based on device identification, geographical location information, and historical response performance indicators;

[0038] After the mutual recognition of each intelligent switch device is completed, it periodically broadcasts its own status summary information to adjacent intelligent switch devices. The status summary information includes the current load status, fault judgment summary, and received control instruction identification, and is used to establish a running state consistency judgment mechanism within the local area;

[0039] In the case of the interruption or inaccessibility of the distribution automation main station communication, the intelligent switch devices that have completed mutual recognition independently operate reconstruction strategies within the local area according to the pre-negotiated response order, including delayed closing control, cascaded fault isolation, and communication link integrity self-check, so as to achieve autonomous self-healing operations within a limited range.

[0040] In this embodiment, the specific technical process of the chained mutual recognition operation is further defined. This operation aims to enhance the local collaborative ability among intelligent switch devices so that they can still rely on local mechanisms to complete autonomous fault handling and power supply reconstruction in the case of the inability of the distribution automation main station to issue instructions in a timely manner or communication anomalies.

[0041] After the hierarchical control structure configuration is completed during the system initialization phase, the distribution automation master station sends a chained mutual recognition instruction to the intelligent switch devices in a group of predefined areas. This instruction includes content such as mutual recognition scope identification, communication handshake rules, response parameter templates, and mutual recognition algorithm configuration. After receiving this instruction, the intelligent switch devices perform two-way identification and negotiation with other physically adjacent intelligent switch devices through the local low-latency communication protocol. The communication protocol can be based on CAN bus, RS-485, low-power wireless (such as Zigbee), or industrial-grade wireless Ethernet, and it is necessary to ensure that the two-way handshake is completed within the range of 10 to 50 milliseconds.

[0042] During the mutual recognition process, the intelligent switch devices will actively exchange their respective identity identifiers (such as device addresses, unique serial numbers), geographical location information (provided by GIS data or GPS modules), and historical response performance indicators (such as the most recent action response time, false operation rate, recovery success rate, etc.). Based on this data, the devices will dynamically establish a priority order table in the local link through a negotiation mechanism. This table defines which device will perform isolation or reconstruction operations first in the event of a future fault, and which device needs to wait for other devices to complete the preliminary judgment before responding, thus preventing linkage conflicts or misjudgments caused by multiple devices acting simultaneously.

[0043] After the mutual recognition negotiation is completed, each intelligent switch device will enter the information synchronization state. The device will broadcast a set of status summary information to the adjacent intelligent switch devices in its link at a set period (for example, every 500 milliseconds to 2 seconds). This status summary information is generated locally by the device and includes the current load status (such as current magnitude, load rate), real-time fault judgment summary (such as fault type, confidence level), and the control instruction identification code of the most recently received or issued. This mechanism enables the device to quickly compare by judging the status of neighbor devices when the master station does not issue a unified instruction, so as to judge whether there is a local anomaly or whether it should enter the self-healing mode.

[0044] The broadcast data should adopt a redundant compression coding method to ensure both communication throughput and avoid transmission conflicts. It is recommended to use a polling broadcast or a short-frame communication mechanism based on frequency hopping. When any device in the link detects that the distribution automation master station is unreachable (for example, the heartbeat packet has not been received for multiple consecutive communication cycles or the PING of the master station address fails), it will enter the autonomous control mode. In this case, the intelligent switch devices that have completed mutual recognition can independently initiate the local fault isolation and power supply restoration process according to the response order table generated by the negotiation.

[0045] In autonomous mode, the first operations to be performed include a delayed closing mechanism. That is, after detecting that the standby path meets the closing conditions, it does not execute immediately. Instead, it waits for a preset delay time and confirms that the neighbor device status is consistent to avoid closed-loop conflicts. Secondly, the device will execute a cascading fault isolation strategy. For example, when the upstream device does not operate, the downstream device infers that there may be an unresolved fault upstream and will actively expand the isolation range to ensure power supply safety. In addition, the intelligent switch device will also perform a communication link self-check operation, verifying whether it is still connected to other devices on its mutual recognition link by broadcasting a response packet to determine whether it has complete autonomous control conditions.

[0046] The entire autonomous process is completed locally and maintains a data recording synchronization interface with the master station to ensure that after the master station resumes communication, the device status and control history can be seamlessly integrated.

[0047] In summary, the chained mutual recognition operation not only constructs a set of dynamically updatable local response priority mechanisms, but also introduces device status sharing and a fault emergency autonomous process, significantly improving the robustness and recovery ability of the system in an abnormal communication environment.

[0048] Based on the chained mutual recognition mechanism proposed in this embodiment, the intelligent switch devices within the link can not only achieve autonomous fault response in the traditional sense, but can also be further extended to more complex and higher-level scenarios, such as emergency dispatch linkage control, distributed power access management, load dynamic reconfiguration negotiation, etc., to construct a decentralized and scenario-adaptive distribution intelligent collaboration system.

[0049] In the emergency dispatch scenario, the chained mutual recognition intelligent switch devices are not limited to responding to local fault information, but dynamically evaluate the power supply safety margin based on the state coordination of the entire link. When the distribution automation master station issues a warning instruction in advance about possible extreme weather, short-term load shocks, or risks of upstream power interruption, each intelligent switch device within the link can actively enter the "pre-coordination state". In this state, in addition to the regular state information, the devices also share parameters such as risk pre-estimation values calculated based on a preset model, adjustable load ratios, and standby path feasibility scores. The devices can establish a distributed event voting mechanism on the link. If the majority of nodes in the link predict that the short-term power supply shortage reaches the warning threshold, they will proactively execute distributed open-loop isolation, load sinking, standby connection pre-closing and other actions in advance to form a regional-level defensive dispatch response, significantly improving the overall response ability to sudden risks. This process is different from the existing master station centralized control and is driven by link coordination, with stronger local rapidity and prediction accuracy.

[0050] In the distributed power access scenario, the chain structure can be expanded to a power-aware mutual recognition chain. Once the intelligent switching device at the access point detects that the distributed power source has the conditions for grid connection, such as stable inverter output, voltage frequency matching, and sufficient grid harmonic tolerance, it can automatically send a grid connection request to other devices in the chain. The request will be accompanied by key parameters such as injectable power, maximum response delay, and planned operating time. Other devices in the link will dynamically feedback the acceptability based on data such as the current load status, topological node voltage sensitivity, and fault response priority. Ultimately, the on-chain consensus algorithm (such as the minimum response delay path calculation based on device priority weighting) determines whether to accept the power access. If a local power source is connected to the grid, the link will automatically generate a local energy flow reconstruction plan and adjust the fault strategy within the link, such as setting the power node isolation priority to ensure that the power side can be quickly separated in the event of a fault, while not affecting the self-healing logic of the main power supply path. This method breaks through the traditional fixed power access point or master station approval mechanism, allowing devices to locally collaborate to complete access negotiation and power supply mode switching, and improve the adaptability of the distribution system to the dynamic access of distributed power sources.

[0051] In addition, the chain coordination mechanism can also be extended to multi-path load distribution optimization. When multiple backup paths are available at the same time, the link device can actively initiate path sharing negotiation based on the real-time load status. Each device constructs a local power supply path score based on the local load margin, the current bus voltage level, and the historical operation reliability. Through chain propagation and convergence, the score result will automatically form a power supply optimization recommendation. For example, for the same load point, if the two contact paths come from different feeders, the link will recommend a load splitting mode, and the load will be shared by the two paths in proportion, and the distribution ratio will be automatically adjusted according to real-time feedback, thereby achieving dynamic power supply balancing at the link level. This type of self-balancing logic is significantly different from the traditional master station fault-based reconstruction strategy, and has stronger scenario adaptability and inter-device negotiation capabilities.

[0052] Through the above mechanism, this embodiment not only expands the real-time collaboration capability between devices on the basis of chain mutual recognition, but also realizes the structural transition of the distribution system from "control command chain" to "autonomous intelligent chain". The key is that the equipment has dynamic mutual recognition identity, can propagate context status, can establish operation consensus on its own in the absence of a master station, and can adaptively reconstruct local control strategies according to different scenario goals (such as safety, efficiency, and flexibility) in various types of power events. This architecture is fundamentally different from traditional hierarchical centralized master-slave control. It does not rely on central computing or static configuration, but is an embedded, linked, and evolving chain intelligent collaboration mechanism.

[0053] In terms of implementation, the above functions can be achieved by deploying a controller with chain - type state management, dynamic consensus negotiation, and adaptive policy execution modules in the intelligent switch device. In combination with a communication protocol stack that supports link - state broadcasting and collaborative decision - making, this mechanism can be directly embedded in the existing distribution automation architecture, with good scalability and practical deployment prospects.

[0054] Step S102: The intelligent switch device collects the current and voltage states of the connected line. When a signal that meets the preset fault criterion is detected, the fault section is identified locally and the corresponding line section is cut off to achieve fault isolation in the first time.

[0055] In this embodiment, step S102 is the core link to achieve rapid fault isolation and reduce the scope and duration of power supply interruption. The basic goal of this step is to rely on the intelligent switch devices deployed at the lower layer to achieve local identification and excision of faults when a fault occurs in the distribution system.

[0056] First of all, the intelligent switch device should have the ability to monitor the current and voltage states of the connected distribution lines in real - time. To ensure detection accuracy and response timeliness, the device should integrate current transformers (CTs) and voltage transformers (PTs) or high - precision sensor modules, and be equipped with a microprocessor or digital signal processor (DSP) with edge - computing capabilities. The device should acquire three - phase current and voltage data at a sampling frequency not lower than 5 kHz to ensure a high - fidelity record of the initial fault state, and at the same time generate characteristic parameters such as effective values, phase angles, and zero - sequence components in real - time.

[0057] Based on the real - time collected data, the intelligent switch device needs to run a set of preset fault - identification criteria. This criterion can make a comprehensive judgment based on parameters such as sudden change in fault - current amplitude, duration, change in current direction, zero - sequence current amplitude, and degree of phase - voltage drop. For example, if the system current amplitude exceeds three times the rated current threshold within a very short time and lasts for more than two sampling periods, and is accompanied by a corresponding voltage dip or an increase in the unbalance factor, it can be determined as a typical short - circuit fault. To improve the recognition robustness, multi - condition fusion criteria can be introduced and priorities can be set to distinguish between instantaneous disturbances and permanent faults.

[0058] After determining the existence of a fault, the device needs to perform the positioning and judgment of the fault section. The intelligent switch device can use methods such as current - direction method, longitudinal - differential method, negative - sequence current method, or artificial - intelligence - assisted algorithms to analyze the relationship between the current directions and current amplitudes of upstream and downstream switches, so as to infer whether the fault is located in the current line section. If the current at the outlet of the current switch rises significantly while there is no significant current input at the inlet, it can be preliminarily determined that the fault is located downstream of this section; if there is fault current flowing in both upstream and downstream, it is necessary to compare and judge by combining the data of adjacent switches.

[0059] Once it is determined that this section of the line is the fault source, the device will issue a control command to perform a physical opening operation through the internal breaker drive mechanism. The breaker can be in the form of a vacuum breaker, SF6 breaker, solid-state switch, etc. The opening time is required to be controlled within the millisecond range to ensure that the fault current is cut off before the zero crossing point, preventing the spread of the arc or damage to the equipment. While performing the opening, the device should also record information such as the current event timestamp, the current and voltage curves before and after the fault, and the switch operation status, forming a complete local fault event record for subsequent upload to the master station system for further analysis.

[0060] It should be noted that step S102 emphasizes that both the fault identification and isolation operations are completed locally by the intelligent switch device without relying on external instructions from the distribution master station. Therefore, the system has a significant advantage in response speed and can still complete the first-time local self-healing process even when there is communication delay or the master station is unreachable.

[0061] In summary, step S102 realizes the rapid identification and isolation of the fault section without manual intervention after the fault occurs by deploying intelligent switch devices with real-time acquisition and local fault identification capabilities, laying a key foundation for subsequent power supply restoration and system reconstruction, and significantly improving the operation stability and reliability of the distribution system.

[0062] Furthermore, the intelligent switch device collects the current and voltage states of the connected line. When a signal that meets the preset fault criterion is detected, the fault section is identified locally and the corresponding line section is cut off to achieve first-time fault isolation, including:

[0063] The intelligent switch device adopts a dynamic criterion adjustment mechanism. According to the topological position of the device itself, the fluctuation characteristics of the historical load curve, and the cooperation parameters of other intelligent switch devices in the link, the threshold range of local fault identification is adjusted in real time during operation, making the fault discrimination more in line with the scenario characteristics and avoiding misoperation or refusal to operate;

[0064] When a specified intelligent switch device detects that the electrical state of the connected line meets the preliminary fault characteristics, it does not immediately operate to cut off, but synchronizes the preliminary fault judgment information to other adjacent intelligent switch devices in the link through a chain communication mechanism. Multiple devices jointly complete the multi-point synchronous verification judgment to confirm it as a consistent fault, and then select the device with the highest priority to execute the isolation action, and other devices enter the waiting response state to prevent repeated cutting;

[0065] After the cutting action is completed, a post-fault data caching mechanism is executed. The intelligent switch device that executes the isolation action records the complete electrical waveforms and criterion parameters before and after the fault, and saves the recorded data in a structured manner in the chain communication network for subsequent master station backtracking analysis or other device linkage strategy adjustment, realizing event-level cascade closed-loop management.

[0066] To achieve fast, accurate, and false - operation - free fault response, this embodiment provides a highly intelligent dynamic criterion adjustment and multi - device collaborative confirmation mechanism. Combined with the structured caching and sharing of event data, an adaptive local fault handling logic that can adapt to the complex operating environment of the distribution network is constructed.

[0067] First, in the fault recognition strategy of intelligent switch devices, different from the method of using static fixed current and voltage thresholds as judgment criteria in existing distribution automation systems, this embodiment introduces a dynamic criterion adjustment mechanism. Specifically, the intelligent switch device will, based on its current network topology location, such as a main feeder node, branch node, or end - node, etc., refer to the historical load fluctuation characteristics during its long - term operation, such as morning and evening load change patterns, periodic spikes, etc., and receive in real - time the operating state parameters from other devices within the link, such as load rate changes, voltage sag frequencies, response delay conditions, etc., to dynamically calculate the criterion window applicable to the current scenario. This window includes but is not limited to short - circuit current judgment thresholds, voltage sag criterion time windows, zero - sequence current amplitudes, and durations, etc. This dynamic adjustment method enables the device to have stronger false - operation suppression capabilities during high - load cycles or high - disturbance sections, while in low - interference and sensitive areas, the response sensitivity can be appropriately increased, effectively avoiding false operations and missed operations.

[0068] Second, after detecting a signal that may meet the preliminary fault characteristics, the intelligent switch device does not immediately perform a circuit - breaking operation. Instead, it synchronously sends this preliminary judgment result to other adjacent intelligent switch devices within the link through a low - latency chain - type communication mechanism. This preliminary judgment information includes the detection time, current and voltage mutation parameters, criterion trigger details, and the current state label of this device, etc. After receiving this information, adjacent devices perform a quick comparison in combination with their own monitoring data to confirm whether they detect a similar fault trend or determine whether the fault may be located in this section of the line. Through this collaborative verification mechanism, multi - point synchronous judgment can be achieved. Once the response consistency threshold set in the link is exceeded (for example, more than two devices simultaneously report matching fault trends), the device with the highest negotiated priority initiates the isolation action. This priority can be generated based on preset indicators such as device response delay, topology location weight, action stability record, etc., while other devices enter a waiting state and make subsequent responses when the isolation is confirmed to be successful or the link is abnormal. This chain - type consistency confirmation method effectively prevents multi - point repeated excision or false triggering caused by signal reflection, transient disturbances, etc., which is beneficial to maintaining the stability of the system and the integrity of the restoration path.

[0069] After the fault isolation action is completed, the intelligent switch device that executes the action will also automatically trigger the local data caching and reporting mechanism. The device not only records the action time and the change of switch state, but also packages the full-wave current and voltage data, dynamic criterion values, communication interaction records, etc. within a certain number of seconds before and after the fault, and writes them into the local data module in a structured format. At the same time, it uploads them to the link shared buffer through the chained communication interface. Other intelligent switch devices and the upper-layer master station system can call this data for retrospective analysis, event verification, policy optimization, etc. during the subsequent fault recovery or system reconstruction process. The recording method of structured data can adopt JSON or compressed markup formats (such as CBOR), and embed event tags and link node numbers to achieve cross-device and cross-link time-series comparison and behavior consistency auditing.

[0070] Through the above mechanism, this embodiment has achieved a substantial breakthrough in the existing technology in three dimensions: the intelligence of fault identification, the coordination of execution actions, and the closed-loop nature of data processing. The intelligent switch device is no longer a single-point static response unit, but an autonomous node with environmental perception, adaptive collaborative judgment, and data traceability capabilities, forming a network of intelligent self-healing units with evolutionary capabilities. Just by embedding a dynamic threshold adjustment module, a lightweight chained communication protocol stack, a fault collaborative determination algorithm, and a structured data recording module into the control logic of conventional intelligent switch devices, the functions described in this step can be realized.

[0071] Furthermore, the dynamic criterion adjustment mechanism of the intelligent switch device includes calculating the local short-circuit current action threshold based on the following formula 1 :

[0072] ;

[0073] Where is the maximum load current within the last 24 hours; is the standard deviation of the load current within the last 1 hour, used to characterize the current load volatility; is the average phase voltage within the current 15 minutes; is the target voltage value issued by the distribution automation master station; represents the topological level parameter of the intelligent switch device. The value of the main feeder device is 1.0, the value of the branch line device is 0.5, and the value of the end node device is 0; is the empirical coefficient.

[0074] In the intelligent power distribution hierarchical self-healing fault isolation and power supply restoration method provided in this embodiment, in order to achieve a more sensitive and stable fault identification function, the intelligent switch device no longer uses a fixed short-circuit current threshold for judgment, but introduces a dynamic judgment adjustment mechanism. This mechanism combines the historical load operation status, voltage stability and topological structure characteristics of the equipment to calculate the current short-circuit current action threshold in real time, so as to adaptively adjust the judgment standard according to the actual situation on site, and avoid false operation or refusal to operate due to different equipment locations or changes in operating status.

[0075] This embodiment uses formula 1 to dynamically calculate the short-circuit current action threshold , This formula is constructed based on physical meaning and actually available data, and has good engineering feasibility. Now we explain each parameter in the formula item by item:

[0076] in, It is the maximum load current in the past 24 hours, in amperes (A), which is used to characterize the maximum power load of the current device access point in the past complete operation cycle. This value is recorded by the built-in sampling module of the intelligent switch device for historical current sampling data, and the maximum value is generated in the hourly update or sliding time window as the operating scale benchmark of the current device.

[0077] parameter It is the standard deviation of the load current in the past hour (unit: A), which is used to reflect the current fluctuation degree of the device access point in the short term, that is, whether the load is stable. By collecting current data once a minute, the standard deviation can be calculated in the sliding window. The larger the standard deviation, the more severe the load fluctuation, so the action threshold of the device should be increased accordingly to avoid false action caused by short-term spikes.

[0078] Indicates the average phase voltage within the current 15 minutes, in volts (V), which is calculated by continuous sampling of the voltage measurement module inside the device. This value reflects whether the current power supply voltage is stable and is suitable for judging whether the system is in an undervoltage, light load or local fluctuation state.

[0079] It is the target voltage value sent by the master station. The unit is also volts (V). It is generally 220V, 230V or the rated phase voltage set by the system. It is used as a reference for the device to judge voltage offset. The device obtains and periodically refreshes this parameter from the master station through the communication interface.

[0080] It is a topological level factor, which is used to reflect the structural position of the intelligent switch device in the entire power distribution network. Considering that the main feeder equipment has a greater impact on the global power supply, its action needs to be more cautious. Set to 1.0; the branch device is set to 0.5, and the end-user access node device is set to 0. This parameter is generated by the system in combination with GIS and the topology diagram during network initialization and written into the configuration tables of each device.

[0081] The empirical coefficients k1, k2, and k3 are weighting factors that can be set according to the system debugging results. The recommended values are respectively:

[0082] : Represents the sensitivity to load fluctuations. The recommended value is adjustable between 0.5 and 1.0. The action threshold is higher when the fluctuations are more severe;

[0083] : Represents the sensitivity to voltage deviation. The recommended value is adjustable between 1.0 and 1.5, reflecting that the device needs to increase the tolerance threshold under undervoltage / overvoltage;

[0084] : Represents the importance weight of the topological position. The recommended value is adjustable between 0.3 and 0.7. The protection threshold of the main trunk device is slightly increased to avoid misoperation.

[0085] Through the operation result of the above formula, the short-circuit action current threshold at the current moment will be correspondingly one level higher than the historical maximum load current, and its value dynamically adjusts according to the real-time load fluctuation, voltage stability, and topological position. For example, when a device is at the end of a branch, with stable load and normal voltage, the threshold is close to itself, which is convenient for quickly isolating faults; while when the device is in the middle section of the main feeder, with large load fluctuations and obvious voltage fluctuations, the threshold will be appropriately increased to avoid misoperation. This action criterion based on scenario weight weighted correction is completely different from the static setting method commonly used in the existing technology.

[0086] In summary, this dynamic criterion adjustment mechanism has significant engineering implementation value. The parameters required by the above formula can be directly implemented through standard modules such as the sampling function, standard deviation operation, voltage average calculation, topology identification, and master station communication interface in the intelligent switch device, and the threshold update is executed in a timed task manner in the device embedded logic or edge controller. This mechanism is particularly suitable for the adaptive protection judgment of multi-level and heterogeneous nodes in large-scale distribution networks, effectively improving the robustness and misoperation suppression ability of the system.

[0087] Step S103: The intelligent switch device uploads the fault information and key electrical data to the distribution automation master station, and the distribution automation master station combines the network topology information to judge the recoverable power supply area and dynamically generates a reconstruction plan.

[0088] In this embodiment, step S103 mainly involves information uploading, global situation awareness, topology structure analysis, and the generation of power supply reconstruction plans. It is a key bridge link connecting local fault identification and network-wide restoration scheduling. The core objective of this step is to enable the distribution automation master station to accurately obtain fault-related data in the first place and, based on the current network operating status and device topology, intelligently analyze feasible power supply restoration paths. The following provides a detailed description of its implementation method.

[0089] After the intelligent switch device completes the identification and removal of the fault section locally, it will upload the fault-related information to the distribution automation master station through the distribution automation communication network. This information at least includes the time stamp of the fault occurrence, the identified fault type (such as single-phase grounding, phase-to-phase short circuit, etc.), the current and voltage waveform data before and after the operation, the unique identification code of the device where it is located, the current state of the switch, and the electrical state parameters of the adjacent lines. The communication method can adopt standard protocols such as IEC 60870-5-104, DNP3, or IEC 61850, and realize the stable transmission of information through networks such as optical fiber, Ethernet, 4G / 5G, or LoRa. In high-reliability application scenarios, redundant channels and time synchronization mechanisms can also be set up to ensure the timely and accurate reception of fault information by the master station.

[0090] After the master station system receives the uploaded data, it will first update the received device status in its distribution network information management module to complete the real-time refresh of the distribution network topology. The master station system should have a complete network topology diagram and a node connection relationship database. By comparing the original structure with the current state changes, it can automatically judge the topological location of the fault section, the isolation range, and its impact on the remaining loads.

[0091] Next, the master station system will call the topology analysis and path reconstruction algorithm to analyze the current power supply layout, line operation conditions, and the accessibility of the backup power supply path, and judge which power supply areas are not directly affected by the fault but lose power due to the interruption of the network structure, and then mark out the "power supply recoverable areas". In this analysis process, factors such as feeder load capacity, switch operation constraints, bus section status, availability of tie switches, backup power supply capacity, and safety margin need to be comprehensively considered to avoid overloading, reverse power transmission, or system instability caused by blind closing.

[0092] Subsequently, the main station system dynamically generates a power supply reconstruction plan based on the calculation results. This plan includes operation instructions such as which tie switches to close, which original feeders no longer undertake power supply tasks, how to redistribute the load, and whether to adjust the voltage control points. To achieve rapid deployment, this plan should be issued in the form of an instruction queue or scripted operation, and a safety logic verification mechanism is provided to avoid further deterioration of the system due to misoperation. When necessary, the main station can also perform multi-objective optimization on the reconstruction plan. For example, select the plan with the fewest operation steps, the smallest voltage fluctuation, or the best load balance among multiple feasible paths to enhance the network's adaptability and operating efficiency.

[0093] It should be noted that the execution of the entire process must be based on real-time performance. Generally, it is required to complete data upload, topology identification, and reconstruction plan generation within seconds to ten seconds after a fault occurs. Therefore, the main station system should have multi-thread processing capabilities and a high-performance data processing architecture to support asynchronous event response and dynamic visualization.

[0094] In summary, the key to implementing step S103 lies in: on the one hand, relying on the intelligent switch device to upload complete fault information, and on the other hand, relying on the main station system's global perception and fast calculation capabilities of the distribution network. Through this step, it is possible to quickly prepare for the reconstruction of the power supply capacity in the non-fault area, create conditions for the automatic operation of subsequent tie switches and power supply restoration, and thus ensure that the system has strong self-healing capabilities and operating resilience in most fault situations.

[0095] Furthermore, the intelligent switch device uploads the fault information and key electrical data to the distribution automation main station. The distribution automation main station combines the network topology information to judge the recoverable power supply area and dynamically generates a reconstruction plan, including:

[0096] Before uploading the fault information, the intelligent switch device performs time series segmentation and feature extraction processing on the current and voltage waveform data within a specified time before and after the fault. Through the principal component analysis and standard deviation filter compression algorithm, it extracts a low-dimensional feature vector including the core content such as the fault trend, waveform disturbance amplitude, and spectrum distortion characteristics, and packs it together with the switch state and device identifier to form structured compressed data and upload it to the distribution automation main station;

[0097] After receiving the structured compressed data, the distribution automation main station constructs a weighted graph model based on the topological position of the fault node, the historical power supply path record, and the current tie switch state. The edge weights comprehensively consider path resistance, remaining load capacity, switch state, power supply priority, and safety margin, and perform a minimum cost path search on this graph to determine a set of alternative power supply path sets;

[0098] After generating multiple feasible power supply reconstruction schemes, the master station calculates the comprehensive scores of each scheme in terms of power supply coverage, load balance, and control efficiency through a multi-objective optimization algorithm based on the local operation evaluation data returned by each intelligent switch device in the link, including the estimated switching load, switch response delay, and voltage regulation ability. The scheme with the highest score is selected as the final power supply reconstruction strategy and written into the fault recovery operation log before execution to achieve traceable management of the decision-making process.

[0099] First, after completing the fault identification and isolation operations, the intelligent switch device does not directly upload the full volume of original sampling data. Instead, it preprocesses the current and voltage waveforms within a specified time period before and after the fault. This time period usually includes the sampling data from 1 second before the fault to 3 seconds after the fault, and the sampling frequency is recommended to be no less than 2 kHz to cover common fault transient processes. During the data processing, the device first divides the waveform data into multiple equal-length segments along the time axis and performs principal component analysis (PCA) operations on each segment to identify the most representative waveform change features. On this basis, it further identifies the local disturbance sections, that is, the waveform distortion mutation regions, through a standard deviation filter, and extracts statistical features such as the corresponding disturbance amplitude and spectral energy change. Finally, a set of low-dimensional feature vectors containing fault trends, disturbance features, frequency domain information, etc. are formed and further compressed into a structured data block. At the same time, the device packs and merges metadata such as the current switch state (such as whether it has tripped, action delay, etc.), device unique identifier, and timestamp to form a complete structured compressed data unit, which is uploaded to the distribution automation master station through the communication link.

[0100] After receiving the compressed data, the master station quickly locates the fault node from the network topology model it maintains based on the location of the faulty device identified in the received data. The master station combines the historical power supply path records (including the operation log and success rate of each tie switch) and the real-time state of the current available tie switches (opening and closing positions, telemetry signal validity, etc.) to construct a regional weighted graph with the fault node as the root. The nodes of the graph represent each power supply unit or switch position, and the edge weights comprehensively consider the following multiple dimensions: line resistance value, remaining load capacity of the devices in the path (inferred from the real-time load rate of the devices), controllability of the switch state (whether automatic switching is allowed), historical power supply priority of this path (such as increasing the score if it is an important load channel), and topological safety margin (such as whether there is a risk of branch linkage in the branch). Based on this graph structure, the master station uses a minimum-cost path search algorithm (such as Dijkstra or an improved algorithm) to traverse the graph structure to generate a set of alternative power supply reconstruction paths with relatively small path costs and feasible structures.

[0101] After the generation of alternative paths is completed, the master station further enters the path evaluation stage. During this process, the master station will query the local operation evaluation data reported by all relevant intelligent switch devices in the link, including but not limited to the expected load capacity that can be carried (calculated based on the difference between the current operating current and the rated capacity of the device), the expected switching response time (which can be dynamically predicted by combining the historical average response of the switch device and the current action queue), and whether it has voltage regulation capabilities (such as being equipped with on-load voltage regulation control or reactive power compensation devices), and use these data as path adaptability indicators for scoring.

[0102] After the reconstruction plan is determined, the master station will also write information such as the generation time of the plan, the scoring model, the selected path number, and the status snapshot before execution into the fault recovery operation log, and form a traceable record for future system maintenance, strategy correction, and operation tracking. This log can not only be used for post-event review by operation and maintenance personnel, but also serve as a data source for the system's self-learning mechanism to promote the evolution and optimization of future plan selection models.

[0103] In summary, this embodiment combines the intelligent compression of edge devices and the global analysis of the central system to construct an intelligent power supply restoration framework with a clear structure, flexible response, path evaluability, and execution traceability. Its actual deployment only depends on conventional sampling hardware, embedded controllers, and the master station control platform, and can be evolved and implemented based on the current distribution automation technology.

[0104] Furthermore, the comprehensive scoring function used to evaluate the quality of power supply restoration paths in the multi-objective optimization algorithm

[0105] ;

[0106] The goal of this scoring function is to convert the performance of multiple paths on key operation indicators into a unified scoring value, where the smaller it is, the better the path is and the more suitable it is as a power supply restoration path.

[0107] represents the number of loads that the current candidate path can cover, that is, the number of users or nodes that can be re-powered once this path is put into operation. The unit can be the number (number of nodes) or the total kilowatt load. This value can be calculated by the master station based on simulating the power supply result on the network topology diagram.

[0108] Correspondingly represents the total load quantity in all current fault cut-off areas; measured in the same unit. The ratio of this item measures the ability of the path in terms of power supply restoration coverage rate. The closer the value is to 1, the stronger the restoration ability. Therefore, its complementary value is used to represent the "unrestored ratio", and the smaller this ratio is, the better the path is.

[0109] Indicates the remaining capacity of the device where the node with the largest load is located in the candidate path, in kilovolt - amperes (kVA), which can be estimated by the difference between the current operating current and the rated capacity of the device. For example, if a device has a rated capacity of 100 kVA and a current operating load of 70 kVA, then its is 30 kVA. Is the theoretical maximum power supply capacity of this path, which can be determined comprehensively by the device capacity and line transmission capacity in all power supply paths.

[0110] Indicates the average response delay returned by all intelligent switch devices participating in the connection operation in this path, in milliseconds (ms). This value can be estimated by the master station sending an inquiry command to the device or based on historical operation records, including communication transmission delay and the time required for the device to execute control commands internally.

[0111] Is the reference maximum allowable response time defined in the system, and it is recommended to be set in the range of 500 ms to 1000 ms.

[0112] Is the comprehensive reliability score calculated by each device in the path according to its own local voltage regulation ability, historical stability score, and device action accuracy rate, with a value range of 0 to 1; for example, it can be set as:

[0113] ;

[0114] Among them Is the action correct rate in the last 10 operations; Is the stability score of the device within three months (generated by the master station operation and maintenance module according to the fault reporting rate); Indicates the ability score of the device to support voltage regulation or reactive power compensation (1 for support, 0 for non - support). The higher this score, the more reliable the devices in the path.

[0115] Are weighting factors, and their recommended values are 0.4, 0.3, 0.2, and 0.1 respectively.

[0116] The master station uses this scoring function to calculate the scoring value for each alternative path respectively , and finally selects the path with the lowest score as the implementation plan for this power supply restoration. The master station will also record this scoring result and the parameters involved in the calculation, and write them into the operation log for future traceability analysis or model optimization.

[0117] Step S104: The distribution automation master station controls a group of connection switches to perform operations according to the reconstruction plan, and switches the non - fault area to other power supply paths to achieve power supply restoration.

[0118] In the intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided in this embodiment, step S104 is the key link to achieve power supply reconstruction. Its core lies in the execution control of the reconstruction plan by the distribution automation master station, enabling the areas not affected by the fault to resume power supply as soon as possible, thereby reducing the scope of power outage impact and enhancing the system resilience.

[0119] In step S103, the distribution automation master station determines the power supply recoverable areas based on the fault information and network topology, and generates a complete power supply reconstruction plan. This plan clarifies the topological structure of the backup power path, the control actions of each switch device, the load distribution strategy, etc. In step S104, the master station will issue instructions for this reconstruction plan and complete the automatic switching of the power path by precisely controlling the closing action of the tie switches.

[0120] The distribution automation master station first needs to establish a stable connection with each on-site switch device through the communication system. The communication network can adopt forms such as optical fiber, dedicated wireless, LTE public network, 5G or industrial Ethernet, depending on the geographical distribution of the distribution network and communication requirements. The communication protocol preferably uses standard protocols such as IEC 60870-5-104, IEC 61850 or DNP3 to ensure the efficient transmission and execution confirmation of instructions.

[0121] On the basis of a stable communication link, the master station will sequentially send control instructions to the target tie switch devices according to the reconstruction plan. These tie switches are usually installed between adjacent feeders or at the junction positions in the distribution ring network structure, and have the capabilities of remote control and status feedback. Each instruction usually includes the device identification address, the expected action type (such as closing or maintaining the lock), the execution timing control parameters, etc. To prevent misoperation or electrical shock, the master station also needs to perform multiple verifications before sending, including the confirmation of the device online status, the verification of the current electrical parameters, the judgment of the operation interlock conditions, etc. For example, in the scenario where multiple tie switches need to be operated simultaneously, the master station can set the time sequence according to the path logical relationship to avoid parallel closing causing the power sources to operate in parallel.

[0122] Before performing the closing operation, the master station can also judge whether the load-carrying capacity of the backup power source is sufficient according to real-time voltage, current and other data, to prevent the power source voltage from dropping or the system protection action from occurring due to sudden load increase. When it is confirmed that the power supply path conditions are met, the master station issues a closing instruction, and after receiving the command, the tie switch will perform the switch action to complete the closing of the power supply path.

[0123] After each switching operation, the master station also needs to read back the telecontrol and telemetry data to confirm whether the operation has been successfully executed. For example, after the tie switch is closed, the master station can read the switch status "closed position" and cross-verify it in combination with the downstream current change. Once it is confirmed that the path has been successfully established and the load has returned to normal, the master station will update the network topology diagram and device status information to make the system in a new stable operating state.

[0124] In scenarios where multiple tie points need to be operated simultaneously or in stages, the master station can also use operation queues and sequence control logic to ensure compliance with electrical safety specifications during the power supply restoration process, such as avoiding improper operations like closing without voltage and closing in a closed loop. In addition, during the path switching process, the master station can cooperate with voltage regulating devices (such as reactive power compensation equipment and tap-changing transformers) to control the voltage quality and ensure the power supply stability after restoration.

[0125] Throughout the entire process, the operation does not require manual intervention and can be automatically completed within seconds after a fault occurs, significantly improving the power supply reliability and self-healing ability of the distribution system.

[0126] In summary, step S104 realizes the dynamic reconstruction of the power supply path through the precise control of the tie switch by the automated master station. This process not only relies on stable communication and control mechanisms, but also on the rationality of the reconstruction plan and the rigor of the operation logic, so as to achieve fast, safe and efficient power supply restoration in the non-fault area.

[0127] Furthermore, the distribution automation master station controls a group of tie switches to perform operations according to the reconstruction plan, and switches the non-fault area to other power supply paths to achieve power supply restoration, including:

[0128] Before issuing the power supply reconstruction command, the distribution automation master station pre-calculates the operation risk factors based on the current load status, voltage stability and action success history of all the devices where the tie switches involved in the optimal path determined by the comprehensive scoring function are located, and marks the nodes with operation risk greater than the preset threshold as nodes to be confirmed. For the nodes to be confirmed, the execution command is delayed and a confirmation request is broadcast.

[0129] After receiving the confirmation request, the intelligent switch device marked as a node to be confirmed negotiates the status with the adjacent intelligent switch devices in the link, obtains the voltage fluctuation at both ends, the load prediction after switching, the device temperature rise or the response saturation status information through chained mutual recognition communication, and makes a local operability judgment; if it is confirmed that the switching condition is met, it returns the allowable switching status flag, otherwise it returns a refusal to execute and attaches the negotiation reason.

[0130] After receiving the feedback from all key nodes, the distribution automation master station dynamically corrects the original path plan, and issues control instructions to all groups of tie switches according to the updated path sequence, controls the power supply reconstruction process in a step-by-step delayed closing manner, and real-time confirms the status of the tie nodes after each action. The unmet part enters the retry mechanism and is written into the operation log.

[0131] To ensure safe, stable, and controllable path switching during the power supply reconstruction process after fault isolation, this embodiment also provides a control process for the distribution automation master station to the tie switches. This process not only considers the score priority of path planning but also introduces an operation risk assessment, local feasibility negotiation, and status feedback mechanism for tie nodes, thereby realizing the whole-process management of sequential closing, sectional confirmation, and dynamic adjustment of tie switches, significantly improving the anti-risk ability and intelligent level of the system during the fault recovery process.

[0132] First, before the master station generates the power supply reconstruction path and is ready to issue the switching control instruction, it will first perform a status assessment on all tie switch devices involved in the selected path. This assessment is based on the current real-time load data of each intelligent switch device, the stability trend of the voltage at the device, and the previous action execution records of the device, including the action success rate, misoperation rate, and timeout frequency, etc. Through this multi-dimensional information, the master station calculates an operation risk factor for each device, which represents the safety and reliability of the device to perform the switching operation under the current working conditions. If the operation risk factor of a certain device exceeds the preset threshold set by the system, the master station marks it as a "node to be confirmed" and does not immediately send it a closing command, but first broadcasts a confirmation request.

[0133] After receiving the confirmation request, the intelligent switch device marked as a node to be confirmed does not make an independent judgment, but exchanges status information with other adjacent intelligent switch devices through a chain mutual recognition mechanism. Specifically, the device will actively request the real-time voltage fluctuation amplitude at both ends, whether there is transient waveform distortion, the expected load change trend after switching, whether there is equipment overheating or action queue congestion, etc. from the upstream and downstream switches. These collaborative data are integrated to judge whether the device has the on-site conditions to perform the closing operation. If the judgment result is positive, the device will return a feedback flag of "allow switching"; if it is found that there are abnormalities in the electrical environment or device status, it will return "refuse to execute" and attach negotiation reasons, such as "voltage fluctuation exceeds the limit" or "expected load overrun" and other information for the master station to make a decision.

[0134] After the master station has collected the feedback from all the nodes to be confirmed, it dynamically adjusts the original power supply reconstruction path plan based on the returned results. If some nodes refuse to switch or time out without response, the master station will first try to perform local detours, replace nodes or adopt a path segment power supply strategy on the original path; if the power supply target cannot be met, it will automatically switch to the sub-optimal power supply path ranked by comprehensive scoring and restart the control preparation process. After confirming the final path, the master station sequentially issues control commands to all the tie switches participating in the switchover and executes them in a step-by-step delayed closing manner, that is, the action of each switch is initiated individually after a set safety delay, and after the action is completed, the master station immediately confirms its status feedback, including whether the closing is successful, whether the current has risen to the load level, and whether the voltage is stable within the allowable range, etc.

[0135] If a certain node fails to complete the closing or returns an abnormal state within the specified time, the master station will suspend the subsequent control commands to avoid power supply path interruption or parallel impact. Such abnormal nodes will be automatically included in the retry queue and the closing operation will be retried within the maximum number of retries set by the system. Each abnormal response or retry operation will be detailedly recorded in the master station operation log, including the device ID, action timestamp, abnormal type, retry round and final result, which is convenient for subsequent operation and maintenance analysis and algorithm optimization.

[0136] In summary, in this embodiment, by introducing a risk-based control sequence planning, a link-level status negotiation feedback mechanism, and an in-sequence delay control and failure retry logic, a highly reliable and self-adaptive adjustment capable power supply reconstruction control scheme is constructed. Without relying on manual intervention, this scheme realizes the intelligent collaborative closing control of multi-path tie switches, significantly improving the system recovery speed and power supply security.

[0137] Furthermore, when the distribution automation master station controls the tie switch to perform an operation, it generates a control sequence and a control delay according to the following rules:

[0138] The control sequence of the tie switch is comprehensively sorted according to the topological level where the device is located, the expected load increment and the operation risk level, and the devices with a lower level, small load change and lower risk are preferentially controlled;

[0139] An independent delay time is set for the control command of each tie switch, and the delay time is dynamically adjusted according to the basic safety interval in combination with the load mutation amplitude and operation risk level of the device;

[0140] The distribution automation master station performs status confirmation on each closing operation. If the response is abnormal or times out, it suspends the subsequent commands and includes the abnormal device in the retry queue. The retry operation has a maximum number of attempts and is automatically recorded in the operation log.

[0141] This embodiment also provides a closing sequence and time control strategy adopted when the distribution automation master station controls the tie switch to perform a power supply path switching operation. This strategy aims to ensure that the actions of the tie switch are completed one by one in the optimal order and at a reasonable pace on the premise of ensuring safety and power supply continuity, so as to achieve precise power supply restoration in the non-fault area and avoid phenomena such as overvoltage, current impact, or system oscillation.

[0142] Before the master station determines the final power supply reconstruction path and is ready to issue a tie switch closing control command, it will sort all the tie switch devices to be operated. The sorting basis is not the traditional physical position order, but three key parameters are considered comprehensively. First is the hierarchical position of the device in the topological structure. Generally, devices near the end or low-voltage branches are considered to be at a lower level. Since their actions have less impact on the overall system, closing these devices first can reduce the risk of system disturbance. Second is the expected load increment, that is, the amount of load change that each device may undertake after closing. The system will give priority to closing those nodes with a relatively small expected load jump to implement the distributed power supply strategy of "gradually loading and gradually restoring". Finally, it is the operation risk level of each device, which can be jointly determined by the device's action history, the evaluation result of the device's health status, and the stability of the electrical environment. The master station weights these three parameters to obtain the sorting priority and generates a complete control instruction issuance sequence according to the priority.

[0143] The control instructions for each tie switch are not issued simultaneously, but with independent delay times. This delay time is set based on the pre-defined basic safety interval time of the system and is dynamically adjusted in combination with the load change amplitude and operation risk level of the device. For example, for a certain device, if it needs to undertake a large load jump after closing, or there is a record of misoperation in its own history, then the master station will allocate a longer delay time for it to ensure that the system has enough time to complete the self-adjustment of voltage stability and load distribution after the previous node is closed. This dynamic delay mechanism can not only effectively prevent system fluctuations caused by concurrent device actions, but also make the power supply reconstruction process more flexible and controllable.

[0144] After each tie switch receives the closing command and executes the operation, the master station will immediately monitor its state changes, usually through indicators such as telecontrol signals, switch position indications, and load current rising trends for closed-loop confirmation. If the device fails to return to the expected state within the specified time, or there is abnormal feedback, such as closing failure, load non-response, or severe voltage fluctuation, the master station will immediately suspend the issuance of subsequent control instructions to avoid the continuous expansion of the fault path or causing system interlock errors.

[0145] At this time, the abnormal node will be automatically added to the retry queue. The master station will arrange for it to attempt to execute the closing action again at a later time according to the current system load condition and device stability. Usually, each device has a maximum allowable number of retry attempts, and it is generally recommended not to exceed three times. The results of each attempt, including success or failure, feedback signal status, execution time, abnormal type, etc., will be completely recorded in the system operation log. These logs can be used for operation and maintenance auditing, device health tracking, or iterative learning of subsequent power supply strategy models, building a highly reliable control system of "status perception - action execution - feedback recording - risk closed-loop" for the entire system.

[0146] Through the above control sequence generation and delay strategy setting, this embodiment not only realizes a refined power supply restoration execution mechanism, but also significantly improves the operation stability, response flexibility and abnormal response ability of the system. Different from the traditional centralized control method of unified command and non-differentiated actions, this embodiment uses a hierarchical topology perception and device behavior prediction model to realize a self-healing control logic that executes on demand, in sequence, and according to feedback in the full-path connection, which has great practical value.

[0147] Step S105: While executing power supply restoration, perform retrospective analysis on the current waveform, harmonic characteristics or partial discharge signals before the fault occurs. If an abnormal trend is identified, record the abnormal event and update the operation strategy for early warning discrimination of future similar faults.

[0148] In the intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided in this embodiment, step S105 not only undertakes the function of system retrospective analysis after the fault, but also constructs an adaptive optimization foundation for the future. Its role is to endow the system with the ability of continuous learning and evolution through abnormal trend identification and strategy update, so as to improve the self-healing and prediction level of the entire distribution network in the face of a complex operation environment.

[0149] After the power supply restoration operation starts, the distribution automation master station immediately starts a set of historical data retrospective mechanism to analyze the evolution process of key electrical parameters during a period of time before the fault occurs. These parameters mainly include current waveform, voltage waveform, harmonic distribution, zero-sequence component, negative-sequence current, partial discharge characteristics, power quality index, etc. The analysis time window can be flexibly set according to the fault type and system response characteristics. Usually, it is recommended to retrospectively analyze the data for 5 to 30 seconds to cover the omen change stage before the fault.

[0150] The sources of the retrospective data include the data locally cached by the intelligent switchgear and the historical sampling data stored in the master station database. To achieve high-resolution analysis, the system should adopt a sampling frequency of not less than 5 kHz and support transient waveform extraction when necessary. For these data, the master station uses a variety of algorithms for signal analysis and pattern recognition. For example, the fast Fourier transform (FFT) is used to analyze the change trend of harmonic distribution to identify whether the surge of high-order harmonics is an unstable precursor before a fault; the mean and standard deviation of the zero-sequence current in a sliding window are used to extract abnormal disturbances; for partial discharge signals, the high-frequency noise signal and pulse count statistics can be combined to detect the early signs of cable insulation deterioration.

[0151] After completing various feature extractions, the system compares the current fault case with the historical known fault models or the expert knowledge base to determine whether there is an identifiable precursor trend for this fault. If an abnormal trend with statistical significance is identified, the master station associates and archives this trend with the fault event and records key information such as the relevant line, equipment, time point, parameter change curve, etc., to form a structured abnormal event record.

[0152] Furthermore, based on the characteristics of the abnormal event, the system can automatically adjust the operation strategy. For example, when it is identified that the same type of harmonic distortion has occurred before multiple historical faults on a certain feeder, the system can appropriately increase the monitoring frequency, lower the abnormal waveform trigger threshold for this line, or suggest to the dispatcher to arrange maintenance in advance. Similarly, for a cable section with frequent abnormal partial discharge signals, the system can include it in key inspections or reconfigure the power supply path to reduce the load.

[0153] The above strategy updates not only apply to the handling process of this fault but also are written into the operation strategy database of the master station, becoming the knowledge basis for judging similar faults in the future. The strategies can include multiple dimensions such as dynamic threshold adjustment, warning level setting, equipment priority ranking, or linkage switch strategy changes, enhancing the warning accuracy and response efficiency of the system when facing similar working conditions in the future.

[0154] It should be emphasized that the entire process requires no manual intervention and is automatically completed by the system for data extraction, analysis, matching, and strategy update, ensuring that the warning mechanism has real-time performance, adaptability, and sustainability.

[0155] In summary, step S105 is not only a retrospective exploration of the pre-fault state but also a future-oriented adaptive risk control mechanism. By deeply analyzing historical waveforms, identifying abnormal trends, and dynamically optimizing operation strategies, the system is equipped with self-learning, self-optimizing, and self-defense capabilities, thus greatly enhancing the warning ability and self-healing level of the intelligent power distribution system.

[0156] Furthermore, the retrospective analysis of the current waveform, harmonic characteristics, or partial discharge signal before the fault occurs includes:

[0157] After a fault is detected, the main station of the distribution automation system retrieves the original sampling data cached by each relevant intelligent switch device during the period before the fault, and establishes a time - continuous waveform set covering multiple devices. Each data sequence in the time - continuous waveform set includes a timestamp, current value, voltage value, zero - sequence quantity, and the amplitudes of each harmonic. The reference time nodes are uniformly aligned to form a comparable analysis basis.

[0158] The main station of the distribution automation system conducts a comparative analysis of the characteristic parameters in the time - continuous waveform set to identify whether there is a continuous changing trend of the following three types of indicators before the fault: First, whether the amplitude of a specific harmonic increases in a continuous number of sampling periods; second, whether there is a stable offset trend in the zero - sequence component of the current; third, whether the amplitude of the voltage fluctuation repeatedly exceeds the set fluctuation threshold within a short period. If any of the above indicators meets the trend - change condition, it is determined that the fault has abnormal precursor characteristics.

[0159] The main station of the distribution automation system compares the fault event with abnormal precursor characteristics with the current response process record and automatically updates the operation strategy data, including: the weight parameter increased corresponding to the precursor characteristics, the adjustment range of the threshold sensitivity in the future criterion, or issues an early - warning configuration parameter update instruction to the relevant intelligent switch device to improve the early identification and response ability for subsequent similar events.

[0160] This embodiment also proposes a system - implementable electrical anomaly prediction mechanism based on the combination of sampling data trend judgment and strategy update. This mechanism not only focuses on the reuse of historical data after a fault occurs, but also emphasizes the coherent cooperation among the data organization structure, trend judgment method, and strategy update logic, which can significantly improve the identification ability of the distribution system for potential faults and the evolution ability of the operation strategy.

[0161] When the system detects a fault in the line, the distribution automation master station will immediately call the local cached data in each intelligent switch device related to the fault. During normal operation, each intelligent switch device continuously records the electrical quantity data of the line it is connected to at a fixed sampling period and caches it in the local storage module. When issuing a data retrieval command, the master station requests each device to upload the original sampling data within a certain number of seconds before the fault. These data include the timestamp of each sampling point, the three-phase current values, the three-phase voltage values, the zero-sequence current component, and the voltage and current amplitudes of at least the first five harmonics. For subsequent comparison and analysis, the master station performs unified time alignment processing on the data sequences reported by all devices, selects the fault trigger time as the unified reference point, and converts the time information in each data sequence into relative time with respect to this reference point to ensure the consistency and synchronization of the analysis window in time.

[0162] Based on the time-continuous waveform set constructed above, the distribution automation master station then conducts trend analysis at the parameter level. First, the master station detects the amplitudes of each harmonic, especially whether the third, fifth, and seventh harmonics show an increasing trend in multiple consecutive sampling periods before the fault occurs. If the amplitude of a certain harmonic component continues to rise in two or more periods, it can be preliminarily determined that there may be a potential power quality disturbance. Second, the master station checks whether the zero-sequence current component slowly deviates from zero or a normal small amplitude state and continuously shows obvious asymmetric load characteristics, which is usually a precursor to low-level ground faults or insulation deterioration. Third, the master station counts the fluctuation frequency and amplitude of the voltage per unit time to determine whether it continuously exceeds the set short-period fluctuation threshold, such as more than three fluctuations exceeding ±10% of the rated value within one second. When any of the above indicators meets the trend change condition, the master station marks the fault as "having abnormal precursor characteristics", that is, not only an electrical fault has occurred, but also clear physical precursor signals can be traced from the historical data before the fault.

[0163] In order to incorporate this abnormal event into the knowledge-based operation strategy system, the master station will correlate and compare such faults with their response processes. For example: whether this fault can be quickly isolated within the specified time, whether there are phenomena of misoperation or refusal to operate, whether a specific communication path is used, etc. Combining these response records, the master station adjusts the current operation strategy library. Specifically, for the parameters whose precursor characteristics are successfully identified, the master station will automatically increase their weight parameters in future fault criteria, and improve the sensitivity of this characteristic in the local criteria of intelligent switchgear. At the same time, for events with trend anomalies that have occurred but have not been responded to in time, the master station can appropriately lower the action thresholds in some criteria to enable the device to intervene in the judgment earlier. In addition, the master station can also directly send the warning parameters to the relevant intelligent switchgear, so that once similar trends are detected during future operation, it can issue an alarm signal in advance, or preset a power-off action under permitted conditions.

[0164] Through the above-mentioned retrospective analysis and policy adaptive update mechanism, the present invention not only realizes the reverse modeling of the causes of the occurred faults, but also establishes a self-learning closed loop with "trend recognition - behavior association - policy evolution" as the core, improving the system's ability to prevent hidden faults or marginal disturbances. Different from the traditional single-point waveform threshold judgment or pure manual experience rules, this embodiment provides a cross-node, cross-time, and accumulative distribution anomaly recognition logic, providing a solid technical foundation for realizing self-healing control under hierarchical cooperation.

[0165] The second embodiment of the application provides an electronic device, and the electronic device includes:

[0166] A processor;

[0167] A memory for storing a program, which when read and executed by the processor, executes an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided in the first embodiment of the present application.

[0168] The third embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it executes an intelligent distribution hierarchical self-healing fault isolation and power supply restoration method provided in the first embodiment of the present application.

[0169] Although the present application is disclosed above with preferred embodiments, it is not used to limit the present application. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of the present application. Therefore, the protection scope of the present application should be defined by the scope defined in the claims of the present application.

Claims

1. An intelligent distribution hierarchical self-healing fault isolation and power supply restoration method, characterized in that, Including: Establish a hierarchical control architecture in the distribution system, where the hierarchical control architecture includes intelligent switch devices at the lower layer and a distribution automation master station at the upper layer; The intelligent switch devices collect the current and voltage states of the connected lines. When a signal that meets the preset fault criterion is detected, the fault section is identified locally and the corresponding line section is cut off to achieve fault isolation in the first time; The intelligent switch devices upload the fault information and key electrical data to the distribution automation master station. The distribution automation master station combines the network topology information to judge the power supply recoverable area and dynamically generates a reconstruction plan; The distribution automation master station controls a group of tie switches to execute operations according to the reconstruction plan, and switches the non-fault area to other power supply paths to achieve power supply restoration; While the power supply is being restored, perform retrospective analysis on the current waveform, harmonic characteristics or partial discharge signal before the fault occurs. If an abnormal trend is identified, record the abnormal event and update the operation strategy for early warning discrimination of future similar faults; The establishment of the hierarchical control architecture in the distribution system includes that after the distribution automation master station completes the deployment in the hierarchical structure, it performs a chain mutual recognition operation on each group of intelligent switch devices in the partition, so that adjacent switch devices complete identity authentication and communication handshake locally, and establish the self-organizing boundary recognition ability between devices to support the regional autonomous fault handling in the offline state of the master station; Among them, the chain mutual recognition operation includes: After receiving the chain mutual recognition instruction issued by the distribution automation master station, the intelligent switch device conducts mutual recognition negotiation with other intelligent switch devices physically adjacent to it through a low-latency local communication protocol, and determines the fault response priority order of each intelligent switch device in the link based on device identification, geographical location information and historical response performance indicators; After each intelligent switch device completes the mutual recognition, it periodically broadcasts its own status summary information to the adjacent intelligent switch devices. The status summary information includes the current load status, fault judgment summary and received control instruction identification, which is used to establish a local area operation status consistency judgment mechanism; In the case of communication interruption or unreachability of the distribution automation master station, the intelligent switch devices that have completed the mutual recognition independently operate the reconstruction strategy in the local area according to the pre-negotiated response order, including delayed closing control, cascaded fault isolation and communication link integrity self-check, so as to achieve autonomous self-healing operation within a limited range.

2. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 1, characterized in that, The establishment of the hierarchical control architecture in the distribution system further includes: In the initialization stage, the distribution automation master station generates a logical hierarchical structure of multi-level power supply areas based on the geographic information system (GIS) and historical operation topology data, and assigns corresponding switch device addresses, fault criterion templates and communication strategies to each level node to achieve synchronous matching of the topology structure and function configuration; Through the issued configuration information, synchronize the operation roles and action permissions of each intelligent switch device, so that it has different levels of fault response priorities, coordination waiting times and reporting window parameters according to the network level it is in, so as to achieve adaptive collaborative response among multiple devices in a concurrent fault scenario.

3. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 1, wherein, The intelligent switch device collects the current and voltage states of the connected line. When a signal that meets the preset fault criterion is detected, it identifies the fault section locally and cuts off the corresponding line section at the first time to achieve fault isolation, including: The intelligent switch device adopts a dynamic criterion adjustment mechanism. According to the topological position of the device itself, the fluctuation characteristics of the historical load curve, and the coordination parameters of other intelligent switch devices in the link, it adjusts the threshold range of local fault identification in real time during operation, making the fault discrimination more in line with the scenario characteristics and avoiding misoperation or refusal to operate; When a specified intelligent switch device detects that the electrical state of the connected line meets the preliminary fault characteristics, it does not immediately cut off the power. Instead, it synchronizes the preliminary fault judgment information to other adjacent intelligent switch devices in the link through a chain communication mechanism. Multiple devices jointly complete multi-point synchronous verification and judgment to confirm a consistent fault, and then select the device with the highest priority to execute the isolation action. Other devices enter the waiting response state to prevent repeated cutting off; After the cutting-off action is completed, a post-fault data caching mechanism is executed. The intelligent switch device that executes the isolation action records the complete electrical waveforms and criterion parameters before and after the fault, and saves the recorded data in a structured manner in the chain communication network for subsequent backtracking analysis by the master station or adjustment of the linkage strategy of other devices, realizing event-level cascade closed-loop management.

4. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 3, characterized in that The dynamic criterion adjustment mechanism of the intelligent switch device includes calculating the local short-circuit current action threshold based on the following formula 1 : ； Among them, is the maximum load current within the most recent 24 hours; is the standard deviation of the load current within the most recent 1 hour, used to characterize the current load volatility; is the average phase voltage within the current 15 minutes; is the target voltage value issued by the main station of distribution automation; represents the topological level parameter of the intelligent switching device. The value for the main feeder device is 1.0, the value for the branch line device is 0.5, and the value for the end node device is 0; is an empirical coefficient.

5. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 4, wherein The intelligent switch device uploads the fault information and key electrical data to the distribution automation master station. The distribution automation master station combines the network topology information to judge the power supply recoverable area and dynamically generates a reconstruction plan, including: Before uploading the fault information, the intelligent switch device performs time series segmentation and feature extraction processing on the current and voltage waveform data within a specified time before and after the fault. Through the principal component analysis and standard deviation filter compression algorithm, it extracts a low-dimensional feature vector including the core contents such as the fault trend, waveform disturbance amplitude, and spectrum distortion characteristics, and packs it together with the switch state and device identifier to form structured compressed data and upload it to the distribution automation master station; After receiving the structured compressed data, the distribution automation master station constructs a weighted graph model based on the topological position of the fault node, the historical power supply path record, and the current state of the tie switch. The edge weights comprehensively consider the path resistance, remaining load capacity, switch state, power supply priority, and safety margin, and perform a minimum-cost path search on this graph to determine a set of alternative power supply path sets; After generating multiple feasible power supply reconstruction plans, the master station is based on the local operation evaluation data returned by each intelligent switch device in the link, including the expected switching load, switch response delay, and voltage regulation ability. Through a multi-objective optimization algorithm, it calculates the comprehensive scores of each plan in terms of power supply coverage rate, load balance degree, and control efficiency, and selects the plan with the highest score as the final power supply reconstruction strategy, and writes it into the fault recovery operation log before execution to realize traceable management of the decision-making process.

6. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 5, wherein, The comprehensive scoring function used to evaluate the quality of the power supply reconstruction path in the multi-objective optimization algorithm is given by the following formula 2: ； Among them, represents the number of loads that the current candidate path can cover; represents the total number of loads in all current fault cut-off areas; is the remaining available capacity of the device where the node with the heaviest load in the path is located; is the theoretical maximum power supply capacity of this path; represents the average response delay returned by all intelligent switch devices participating in the connection operation in this path; is the reference maximum allowable response time defined in the system; is the comprehensive reliability score calculated by each device in the path according to its own local voltage regulation ability, historical stability score and device action accuracy rate, and the value range is from 0 to 1; is the weighting factor.

7. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 6, wherein The distribution automation master station controls a group of tie switches to perform operations according to the reconstruction plan, and switches the non-fault area to other power supply paths to achieve power supply recovery, including: Before issuing a power supply reconstruction command, the distribution automation master station pre-calculates the operation risk factor based on the current load status, voltage stability, and action success history of all equipment where the tie switches involved in the optimal path determined by the comprehensive scoring function are located, marks the nodes with operation risk greater than the preset threshold as nodes to be confirmed, delays the issuance of the execution command for the nodes to be confirmed, and broadcasts a confirmation request at the same time; After receiving the confirmation request, the intelligent switch device marked as a node to be confirmed negotiates its status with the adjacent intelligent switch devices in the link, obtains information such as voltage fluctuation at both ends, load prediction after switching, device temperature rise, or response saturation state through chain mutual recognition communication, and makes a local operability judgment; if it is confirmed that the switching condition is met, it returns an allowed switching status flag, otherwise it returns a refusal to execute and attaches the negotiation reason; After receiving the feedback from all key nodes, the distribution automation master station dynamically modifies the original path plan, and issues control instructions to all groups of tie switches in the updated path order, controls the power supply reconstruction process in a step-by-step delayed closing manner, and confirms the status of the tie nodes in real time after each action. The unmet parts enter the retry mechanism and are written into the operation log.

8. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 7, wherein When controlling the tie switch to execute an operation, the distribution automation master station generates the control sequence and control delay according to the following rules: The control sequence of the tie switch is comprehensively sorted according to the topological level where the device is located, the expected load increment, and the operation risk level, and the devices with lower levels, smaller load changes, and lower risks are preferentially controlled; An independent delay time is set for each control instruction of the tie switch, and the delay time is dynamically adjusted according to the basic safety interval and combined with the load mutation amplitude and operation risk level of the device; The distribution automation master station performs status confirmation for each closing operation. If the response is abnormal or times out, the subsequent instructions are suspended, and the abnormal device is included in the retry queue. The retry operation has a maximum number of attempts and is automatically recorded in the operation log.

9. The intelligent distribution hierarchical self-healing fault isolation and power supply restoration method according to claim 8, wherein The retrospective analysis of the current waveform, harmonic characteristics, or partial discharge signal before the fault occurrence includes: After detecting a fault, the distribution automation master station retrieves the original sampling data cached by each relevant intelligent switch device during the period before the fault, and establishes a time-continuous waveform set covering multiple devices. Each data sequence in the time-continuous waveform set includes a timestamp, current value, voltage value, zero-sequence quantity, and the amplitude of each harmonic, and uniformly aligns the reference time node to form a comparable analysis basis; The distribution automation master station compares and analyzes the characteristic parameters in the time-continuous waveform set to identify whether the following three types of indicators have a continuous change trend before the fault: First, whether the amplitude of a specific harmonic increases in consecutive several sampling periods; Second, whether there is a stable offset trend in the zero-sequence component of the current; Third, whether the voltage fluctuation amplitude repeatedly exceeds the set fluctuation threshold within a short period; if any of the above indicators meets the trend change condition, it is determined that the fault has abnormal precursor characteristics; The distribution automation master station compares the fault events with abnormal precursor characteristics with the current response process records and automatically updates the operation strategy data, including: the weight parameters increased for the corresponding precursor characteristics, the adjustment range of the threshold sensitivity in the future criterion, or issues an early warning configuration parameter update instruction to relevant intelligent switch devices to improve the early recognition and response capabilities for subsequent similar events.

Images