Adaptive recovery method and device for low-voltage power grid self-healing control, computer equipment, readable storage medium and program product

Abstract

The application relates to an adaptive recovery method and device for low-voltage power grid self-healing control, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: monitoring the operation state of a low-voltage power distribution network in real time, and marking a suspected fault node; determining a fault influence area based on a topological structure diagram of the low-voltage power distribution network and the suspected fault node; screening out a candidate recovery node set with a health degree meeting a preset requirement through a node health degree scoring model; constructing multiple feasible recovery paths from a path starting point on the boundary of the fault influence area to a path ending point in the candidate recovery node set; calculating the comprehensive recovery cost and network disturbance influence value of the feasible recovery paths based on the multidimensional indexes of the feasible recovery paths; outputting an optimal recovery path and a switching node action sequence based on a recovery strategy optimization model, and performing corresponding recovery operations. The method can realize adaptive fault recovery of the low-voltage power distribution network.

Description

Technical Field

[0001] This application relates to the field of power system automation control technology, and in particular to an adaptive recovery method, device, computer equipment, computer-readable storage medium, and computer program product for low-voltage power grid self-healing control. Background Technology

[0002] With the emergence of new energy sources, distributed power sources, energy storage devices, and electric vehicles, the structure of modern low-voltage power distribution networks is becoming increasingly complex. In actual operation, low-voltage power grids often experience local node voltage instability, branch circuit tripping, or regional power outages due to line faults, equipment aging, severe weather, or communication anomalies.

[0003] In traditional technologies, fault location is usually determined by manual inspection and recovery by manual switching; or a self-healing control mechanism is introduced in low-voltage distribution networks to achieve rapid recovery of local faults through real-time monitoring, fault identification and power supply path switching.

[0004] However, fault recovery through manual inspection is inefficient and untimely, and can easily lead to prolonged power outages, decreased user satisfaction, and even large-scale disturbances and power supply safety risks. Existing self-healing control mechanisms are mostly based on a "one-time recovery" approach, lacking real-time evaluation and adaptive correction mechanisms for recovery effectiveness. This results in the inability to quickly retry or adjust strategies after recovery failure. Furthermore, most existing methods rely solely on static recovery path selection based on topology, failing to comprehensively consider multi-dimensional factors, leading to unstable recovery results. Most control strategies also depend on preset rule bases, unable to self-optimize based on different fault types, network states, and historical experience, lacking intelligent evolution capabilities. Summary of the Invention

[0005] Therefore, it is necessary to provide an adaptive recovery method, device, computer equipment, computer-readable storage medium, and computer program product for low-voltage power grid self-healing control that can quickly, stably, and adaptively recover from low-voltage power distribution network faults, addressing the aforementioned technical problems.

[0006] In a first aspect, this application provides an adaptive recovery method for self-healing control of low-voltage power grids, comprising:

[0007] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0008] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0009] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0010] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0011] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0012] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0013] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0014] In one embodiment, based on the optimal recovery path and the switching node action sequence, corresponding recovery operations are performed, including:

[0015] Based on the location of the fault node or fault segment, the fault branch is disconnected to isolate the fault source; the power supply path is switched according to the optimal recovery path and the action sequence of the switching node.

[0016] In one embodiment, the method further includes:

[0017] If the affected area fails to recover or the operating parameters deviate from the expected performance, a suboptimal path is selected from multiple feasible recovery paths to trigger adaptive reconfiguration, and the control strategy is iteratively corrected until recovery is successful.

[0018] In one embodiment, the branch between the last normal node and the first abnormal node is identified as the faulty segment, including:

[0019] The nodes within the affected area are sorted according to their topological connections to determine the power supply conduction relationships between the nodes;

[0020] Based on the power supply conduction relationship, identify the first abnormal node at the forefront of the continuous abnormal node sequence, and the last normal node upstream of the first abnormal node.

[0021] The physical connection branch between the last normal node and the first abnormal node is identified as the faulty segment.

[0022] In one embodiment, a node health scoring model is used to score the health of all controllable nodes in the low-voltage distribution network, and a set of candidate recovery nodes whose health meets preset requirements is selected, including:

[0023] The health score of a controllable node is obtained by weighted summation based on its electrical operation stability, communication quality, historical fault frequency, and topological location. Nodes with health scores higher than the health threshold are initially included in the candidate recovery node set.

[0024] If there are multiple controllable nodes in the candidate recovery node set whose health scores differ within a preset range, a secondary sorting and filtering process is performed based on the load capacity and backup capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

[0025] In one embodiment, the multidimensional metrics include path load-carrying capacity, communication latency, power balancing capability, and number of switching operations; based on the multidimensional metrics of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, including:

[0026] The multidimensional indicators are normalized separately, and a corresponding weight coefficient is assigned to each normalized multidimensional indicator. The multidimensional indicators are then weighted and summed according to the weight coefficients to obtain the comprehensive recovery cost of the feasible recovery path. The comprehensive recovery cost is used to quantify the economic cost and operational difficulty of the feasible recovery path.

[0027] Secondly, this application also provides an adaptive recovery device for low-voltage power grid self-healing control, comprising:

[0028] The detection and judgment module is used to monitor the operating status of each node and branch in the low-voltage power distribution network in real time. If a node is found to have abnormal voltage, abnormal current, or communication failure, the corresponding node is marked as a suspected fault node.

[0029] The fault determination module is used to determine the corresponding abnormal impact range based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, tracing back the corresponding power supply path upwards and searching the corresponding branch load downwards, and using the breadth-first traversal algorithm or the depth-first traversal algorithm; the set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameters is determined as the fault impact area.

[0030] The detection and judgment module is also used to determine the abnormal node as the fault node if there is only one abnormal node in the fault-affected area; and to determine the branch between the last normal node and the first abnormal node as the fault segment if there are multiple consecutive abnormal nodes in the fault-affected area.

[0031] The evaluation and scoring module is used to score the health of all controllable nodes in the low-voltage distribution network using a node health scoring model, and to select a set of candidate recovery nodes whose health meets the preset requirements.

[0032] The modeling and evaluation module is used to determine the power supply interruption point on the boundary of the fault-affected area as the starting point of the path, select the node with power supply capability from the candidate recovery node set as the ending point of the path, construct multiple feasible recovery paths from the starting point to the ending point of the path based on the topology diagram, calculate the comprehensive recovery cost of the feasible recovery path based on the multi-dimensional index of the feasible recovery path, simulate the impact of the feasible recovery path on the low-voltage distribution network after its access, and determine the network disturbance impact value of each feasible recovery path.

[0033] The adaptive path decision module takes the current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value and historical recovery data as input to the recovery strategy optimization model, and outputs the optimal recovery path and switching node action sequence required to recover the fault-affected area.

[0034] The execution control module is used to perform corresponding recovery operations based on the optimal recovery path and the action sequence of the switching node.

[0035] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0036] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0037] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0038] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0039] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0040] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0041] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0042] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0043] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0044] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0045] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0046] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0047] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0048] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0049] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0050] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0051] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0052] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0053] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0054] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0055] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0056] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0057] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0058] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0059] The aforementioned adaptive recovery method, device, computer equipment, computer-readable storage medium, and computer program product for low-voltage power grid self-healing control, by marking suspected faulty nodes and starting from these nodes, tracing back the corresponding power supply path upwards and searching the corresponding branch loads downwards to determine the fault-affected area, can quickly and accurately locate the fault range and prevent further spread of the fault's impact. Through a node health scoring model, a set of candidate recovery nodes whose health meets preset requirements is selected, providing reliable node resources for subsequent construction of feasible recovery paths. By determining the power supply interruption point on the boundary of the fault-affected area as the path starting point, the system proceeds from the candidate nodes... Nodes with power supply capability are selected from the set of recovery nodes as path endpoints. Multiple feasible recovery paths are constructed from the path start point to the path endpoint. The comprehensive recovery cost and network disturbance impact value of each feasible recovery path are calculated, and the advantages and disadvantages of each feasible recovery path are comprehensively evaluated. By constructing a recovery strategy optimization model, the optimal recovery path and switching node action sequence required to restore the fault-affected area are output. Based on the optimal recovery path and switching node action sequence, the corresponding recovery operation is executed, ensuring the scientific nature and effectiveness of the recovery operation. It also realizes rapid self-healing recovery of low-voltage power grid faults and improves the stability and reliability of low-voltage power grid operation. Attached Figure Description

[0060] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0061] Figure 1 This is an application environment diagram of the adaptive recovery method for low-voltage power grid self-healing control in one embodiment;

[0062] Figure 2 This is a flowchart illustrating an adaptive recovery method for low-voltage power grid self-healing control in one embodiment.

[0063] Figure 3 This is a flowchart illustrating the adaptive recovery method for low-voltage power grid self-healing control in another embodiment.

[0064] Figure 4 This is a structural block diagram of an adaptive recovery device for low-voltage power grid self-healing control in one embodiment.

[0065] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0067] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0068] The adaptive recovery method for low-voltage power grid self-healing control provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located on a cloud or other network server. Specifically, terminal 102 or server 104 implements an adaptive recovery method for low-voltage power grid self-healing control, which includes:

[0069] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0070] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0071] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0072] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0073] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0074] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0075] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0076] Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0077] In one exemplary embodiment, such as Figure 2 As shown, an adaptive recovery method for self-healing control of low-voltage power grids is provided, which is then applied to... Figure 1 Taking the server in the example, the explanation includes the following steps 202 to 214. Wherein:

[0078] Step 202: Monitor the operating status of each node and branch in the low-voltage power distribution network in real time. If a node is found to have abnormal voltage, abnormal current, or communication failure, mark the corresponding node as a suspected fault node.

[0079] For example, the operating status of each node and branch in the low-voltage power distribution network is monitored in real time. If a node is found to have abnormal voltage, abnormal current, or communication failure, the node is identified as a suspected fault node.

[0080] Step 204: Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, trace back the corresponding power supply path upwards and search for the corresponding branch loads downwards. Use the breadth-first traversal algorithm or the depth-first traversal algorithm to delineate the corresponding abnormal impact range. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameters is determined as the fault impact area.

[0081] For example, a low-voltage distribution network topology diagram is dynamically generated. Starting from a suspected faulty node, the power supply parent node link of that node is traced back upwards. If the tracing back reaches a power supply node, transformer node, or circuit breaker node in normal condition, the upward tracing stops, and the tracing path is taken as the corresponding power supply path of the suspected faulty node.

[0082] For example, starting from the suspected fault node, all branch nodes are traversed downwards to search for the affected branch load. A breadth-first traversal algorithm or a depth-first traversal algorithm is used to quickly define the corresponding abnormal impact range. All successor nodes D connected to the suspected fault node are recorded. If the successor node set D is in a state of power disconnection, communication failure, or abnormal parameters, the successor node set D is determined as the fault impact area. At the same time, the state parameters of each successor node are used to determine the propagation impact.

[0083] Step 206: If there is only one abnormal node in the fault-affected area, the abnormal node is designated as the fault node; if there are multiple consecutive abnormal nodes in the fault-affected area, the branch between the last normal node and the first abnormal node is designated as the fault segment.

[0084] For example, if only one node in the fault-affected area experiences voltage, current, or communication abnormalities (i.e., only one abnormal node exists within the fault-affected area, while its parent and branch nodes are normal), it can be determined as a single-node fault, and that abnormal node is designated as the faulty node. If multiple consecutive nodes in the fault-affected area experience voltage, current, or communication abnormalities (i.e., multiple consecutive abnormal nodes exist within the fault-affected area), then the connection branch between the last normal node and the first abnormal node is identified as the faulty segment.

[0085] Step 208: Using the node health scoring model, score the health of all controllable nodes in the low-voltage distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0086] For example, a node health scoring model is constructed using node operation data. Based on electrical operation stability, communication quality, historical fault frequency, and topological location, the health of all controllable nodes in the low-voltage power distribution network is scored, and a health threshold is set to filter out a set of candidate recovery nodes whose health meets the preset requirements.

[0087] Step 210: Determine the power supply interruption point on the boundary of the fault-affected area as the starting point of the path, select a node with power supply capability from the candidate recovery node set as the ending point of the path, and construct multiple feasible recovery paths from the starting point to the ending point of the path based on the topology diagram; calculate the comprehensive recovery cost of the feasible recovery path based on the multi-dimensional index of the feasible recovery path, and simulate the impact of the feasible recovery path on the low-voltage distribution network after its access, and determine the network disturbance impact value of each feasible recovery path.

[0088] For example, a set of nodes in the affected area of ​​the fault is determined, and the power supply interruption point on the boundary of this area is taken as the starting point of the feasible recovery path, and defined as the path starting point set S. A node R with power supply capability is selected from the candidate recovery node set as the endpoint of the feasible recovery path. Based on the low-voltage distribution network topology, a constrained path search algorithm is executed from each path starting point to all candidate recovery nodes to construct multiple feasible recovery paths from the path starting point S to the path ending point R.

[0089] For example, for each feasible recovery path, the comprehensive recovery cost of the path is calculated based on the path's multi-dimensional indicators. The impact of the feasible recovery path on the voltage distribution and load balance of the original low-voltage distribution network is simulated, and the maximum voltage deviation, changes in load balance after connection, and potential inrush current or short-term instability risks are estimated using a distribution simulation engine or simplified calculation formula. This determines the network disturbance impact value of each feasible recovery path, and uses the network disturbance impact value as an additional constraint or penalty to sort multiple feasible recovery paths, generating a list of sorted feasible recovery paths.

[0090] Step 212: The current network status, the multi-dimensional indicators corresponding to each feasible recovery path, the comprehensive recovery cost, the network disturbance impact value, and the historical recovery data are used as inputs to the recovery strategy optimization model. The output is the optimal recovery path and the action sequence of switching nodes required to recover the affected area of ​​the fault.

[0091] For example, based on the current network state, the multi-dimensional indicators corresponding to each feasible recovery path, the comprehensive recovery cost, the network disturbance impact value, and historical recovery data, the low-voltage distribution network fault recovery problem is abstracted into a Markov decision process, and a reward function is constructed:

[0092]

[0093] in, , , , These are the weights of each factor; To determine whether the recovery was successful after the action was performed; The cost function for a feasible recovery path; This represents the impact value of network disturbances. The time required to perform the action.

[0094] A recovery strategy optimization model is constructed based on deep reinforcement learning. This model uses a deep Q-network as its basic structure, where the input layer of the structure is a state vector. The hidden layers are multi-layer fully connected networks or GNN structures, and the output layer is the Q-value estimate for all possible actions; at the same time, an empirical replay pool is used to save state transition samples at each step. ,in, The current network status. For the action to be performed, As a reward, To determine the next network state, a target network mechanism is introduced to estimate the Q-value. During the model training phase, the following is employed: - Greedy strategy, based on probability Choose a random action, with probability Select the action with the highest Q-value and define the maximum number of training epochs or the training reward convergence threshold. And it gradually decreases as the number of training rounds increases. After each training round, the recovery success rate is evaluated on a set of fault samples that were not included in the training. If the policy meets the preset performance requirements in both the simulated set and the unknown set, the policy network is output, including the optimal recovery path and the sequence of switching node actions required to recover the fault-affected area. The preset performance requirements can be a recovery success rate greater than 95% within the perturbation limit, and no specific restrictions are imposed here.

[0095] Step 214: Based on the optimal recovery path and the sequence of actions for switching nodes, perform the corresponding recovery operation.

[0096] For example, based on the optimal recovery path and the switching node action sequence, the operation of disconnecting the faulty branch and switching the power supply path is performed.

[0097] In the aforementioned adaptive recovery method for low-voltage power grid self-healing control, suspected faulty nodes are marked, and the corresponding power supply path is traced back upwards and the corresponding branch loads are searched downwards from the suspected faulty nodes to determine the fault-affected area. This allows for rapid and accurate location of the fault range, preventing further spread of the fault's impact. A node health scoring model is used to select a set of candidate recovery nodes whose health meets preset requirements, providing reliable node resources for constructing feasible recovery paths. By determining the power supply interruption point on the boundary of the fault-affected area as the path starting point, nodes with power supply capability are selected from the candidate recovery node set as the path endpoint, constructing multiple feasible recovery paths from the path starting point to the path endpoint. The comprehensive recovery cost and network disturbance impact value of each feasible recovery path are calculated, comprehensively evaluating the advantages and disadvantages of each feasible recovery path. A recovery strategy optimization model is constructed to output the optimal recovery path and switching node action sequence required to recover the fault-affected area. Based on the optimal recovery path and switching node action sequence, corresponding recovery operations are executed, ensuring the scientific validity and effectiveness of the recovery operations. This achieves rapid self-healing recovery of low-voltage power grid faults, improving the stability and reliability of low-voltage power grid operation.

[0098] In one embodiment, based on the optimal recovery path and the switching node action sequence, the corresponding recovery operation is performed, including: disconnecting the fault branch based on the located fault node or fault segment to isolate the fault source; and performing a power supply path switching operation according to the optimal recovery path and the switching node action sequence.

[0099] For example, after the recovery strategy optimization model outputs the optimal recovery path and the action sequence of the switching node, the residual control signal of the original fault branch is cleared based on the located fault node or fault segment, the fault branch is disconnected to isolate the fault source, and the current running status of the optimal recovery path node is queried to check whether the target node of the recovery path is in an operable state, such as normal communication and online power supply.

[0100] Then iterate through the intermediate nodes and switch device numbers in the selected optimal recovery path, for each node... circuit breakers The system executes the closing command; if remote control access fails or the remote control signal is inconsistent, the anomaly is recorded and the process jumps to the backup path selection procedure. It controls the load transfer switch connected to the downstream node of the fault area, switching the node from the original fault power supply path to the restored path power supply, and enables power outage monitoring protection to avoid instantaneous overcurrent or zero-voltage problems.

[0101] For example, when a microgrid (MG), energy storage system (ESS), or distributed generation (DG) is involved in the recovery process, a synchronization and grid connection operation is required to obtain the main grid bus voltage. ,frequency Phase angle And obtain the output voltage of DG. ,frequency Phase angle And determine whether the synchronization condition is met:

[0102]

[0103] After synchronization, perform the grid-connected switch closing operation and set soft-start current limiting logic, such as 5% incremental load connection or virtual impedance start mode, to prevent impact.

[0104] After each power supply path switching operation, the voltage, current, power, and communication delay of the nodes in the recovery area are collected and compared with the historical operating baseline and the preset expected range. If the node operating parameters are stable within a preset time period, it is preliminarily determined that the fault-affected area has been successfully restored. The preset time period can be 3 to 5 seconds.

[0105] For example, if the recovery of the affected area fails or the operating parameters deviate from the expected performance, a suboptimal path is selected from multiple feasible recovery paths to trigger adaptive reconfiguration, and the control strategy is iteratively corrected until recovery is successful.

[0106] In this embodiment, the fault source is isolated by locating the fault node or fault segment and performing a disconnection operation. The power supply path switching operation is performed according to the optimal recovery path and the switching node action sequence. When the recovery of the fault-affected area fails, the suboptimal path is selected to trigger adaptive reconfiguration. This can prevent the further spread and impact of the fault, while ensuring the orderliness and efficiency of the recovery process, and improving the success rate and reliability of low-voltage power grid fault self-healing recovery.

[0107] In one embodiment, identifying the branch between the last normal node and the first abnormal node as a fault segment includes: sorting the nodes in the fault-affected area according to the topological connection order and determining the power supply conduction relationship between the nodes; identifying the first abnormal node at the forefront of the continuous abnormal node sequence and the last normal node upstream of the first abnormal node according to the power supply conduction relationship; and identifying the physical connection branch between the last normal node and the first abnormal node as a fault segment.

[0108] For example, the nodes within the fault-affected area are sorted according to their topological connections to determine the power supply conduction relationships between the nodes. If a continuous sequence of abnormal nodes exists... Based on the power transmission relationship, identify the first abnormal node at the very beginning of the sequence. and located at the first abnormal node The last normal node upstream Then the fault segment is defined as "fault segment = "" means that the physical connection branch between the last normal node and the first abnormal node is the fault segment.

[0109] In this embodiment, by sorting the nodes in the affected area according to the topological connection order and determining the fault segment, the specific location of the fault can be accurately located, reducing the time for fault investigation and repair, and improving the efficiency and accuracy of low-voltage power grid fault handling.

[0110] In one embodiment, a node health scoring model is used to score the health of all controllable nodes in the low-voltage distribution network, and a set of candidate recovery nodes whose health meets preset requirements is selected. This includes: weighting and summing the controllable nodes based on their electrical operation stability, communication quality, historical fault frequency, and topological location to obtain the corresponding health score of the controllable nodes; selecting nodes whose health scores are higher than the health threshold and initially including them in the candidate recovery node set; if there are multiple controllable nodes in the candidate recovery node set whose health score differences are within a preset range, a secondary sorting and selection is performed based on the load capacity and reserve capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

[0111] For example, based on the electrical operational stability, communication quality, historical fault frequency, and topological location of the controllable node, the controllable node is calculated. Health Score The health score function can be expressed as:

[0112]

[0113] in, For the first Health score of each node; Score for electrical operation stability; A score for communication quality; Score based on historical failure frequency; Weighted scores are assigned based on topological location; , , , The weights of each indicator are given.

[0114] For example, based on each controllable node Health Score Build a health score list Set health threshold A preliminary set of candidate recovery nodes is selected from the score list. ,Right now:

[0115]

[0116] in, .

[0117] If multiple nodes in the initial candidate recovery node set have similar health scores (i.e., the difference in health scores is within a preset range, such as less than 0.2), then auxiliary ranking indicators are introduced, including the real-time load capacity of controllable nodes. and remaining reserve capacity The controllable nodes are then sorted and filtered a second time, namely:

[0118]

[0119] in, These are the weighting coefficients.

[0120] according to Sort the nodes from largest to smallest to generate the final set of candidate recovery nodes. Each candidate node must include a node ID, health score, load capacity, spare capacity, topology location number, and real-time status label.

[0121] In this embodiment, the health score of the controllable node is calculated based on the electrical operation stability, communication quality, historical fault frequency and topology location of the controllable node, and a preliminary set of candidate recovery nodes is selected. At the same time, the load capacity and reserve capacity parameters of the controllable node are introduced for secondary sorting and screening to obtain the final set of candidate recovery nodes. This can comprehensively and accurately evaluate the health status and recovery capability of the nodes, and ensure the reliability and effectiveness of the selected set of candidate recovery nodes.

[0122] In one embodiment, the multidimensional indicators include the path's load-carrying capacity, communication delay, power balancing capability, and number of switching operations. Based on the multidimensional indicators of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, including: normalizing the multidimensional indicators respectively, and configuring a corresponding weight coefficient for each normalized multidimensional indicator; weighting and summing the multidimensional indicators according to the weight coefficients to obtain the comprehensive recovery cost of the feasible recovery path; the comprehensive recovery cost is used to quantify the economic cost and operational difficulty of the feasible recovery path.

[0123] Among them, the multidimensional indicators of feasible recovery paths include path load-carrying capacity, communication delay, power balancing capability, and number of switching operations.

[0124] For example, a comprehensive recovery cost function for a feasible recovery path is constructed based on the normalized multidimensional indicators and their corresponding weighting coefficients. ,Right now:

[0125]

[0126] in, , , , These are the weighting coefficients corresponding to the multidimensional indicators; , , , These are index functions that are normalized to the path's energized load capacity, communication delay, power balancing capability, and number of switching operations, respectively.

[0127] For example, using a constrained multi-source shortest path search algorithm, with constraints dynamically pruning during the search process to output the set of all feasible recovery paths between each pair of breakpoints and candidate nodes, can be represented as:

[0128]

[0129] The final feasible recovery path comprehensive score is:

[0130]

[0131] in, This is a trade-off coefficient between the impact of the disturbance and the cost of the recovery path. The first in the candidate recovery path From the starting node to the... A path between candidate recovery nodes. For path The basic recovery cost function value, For path The access penalty value for disturbances caused to the original low-voltage distribution network operation.

[0132] In addition, the perturbation penalty trade-off coefficient Used to reconcile path-based costs with disturbance impacts, when When, it indicates that the overall path score prioritizes system stability; when When this is the case, it indicates that the overall path score focuses more on recovery efficiency and operational costs. Among these, generally... It can be dynamically set or optimized through learning.

[0133] In this embodiment, by normalizing the multidimensional indicators and calculating the comprehensive recovery cost of feasible recovery paths, a scientific and reasonable basis can be provided for the optimization of recovery strategies, ensuring the optimality and reliability of the selected recovery path.

[0134] like Figure 3 As shown, an adaptive recovery method for low-voltage power grid self-healing control is illustrated using a specific embodiment, including steps 302 to 310. Wherein,

[0135] Step 302: Monitor the operating status of each node and branch in the low-voltage distribution network in real time, identify suspected faulty nodes and locate the area affected by the fault.

[0136] Specifically, the system monitors the operating status of each node and branch in the low-voltage power distribution network in real time. If a node is found to have abnormal voltage, abnormal current, or communication failure, it is identified as a suspected fault node.

[0137] Based on the dynamically generated low-voltage power distribution network topology diagram, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The set of nodes that are topologically connected to the suspected fault node and are abnormal is defined as the fault-affected area, and the fault node or fault segment is determined according to the location of the abnormal node.

[0138] Step 304: Select a set of candidate recovery nodes whose health meets the preset requirements from all controllable nodes in the low-voltage power distribution network.

[0139] Specifically, a node health scoring model is constructed using node operation data. Based on the electrical operational stability, communication quality, historical fault frequency, and topological location of controllable nodes, the health score of controllable nodes is calculated. Health Score And set a health threshold. Filter out health scores Greater than the health threshold The nodes are used to obtain a preliminary set of candidate recovery nodes. .

[0140] If the initial candidate recovery node set There are multiple controllable nodes whose health scores differ within a preset range. The real-time load capacity of these controllable nodes is then considered. and remaining reserve capacity The controllable nodes are then sorted and filtered a second time to obtain the final set of candidate recovery nodes. .

[0141] Step 306: Construct multiple feasible recovery paths from the fault-affected area to the candidate recovery nodes, and calculate the comprehensive recovery cost of the feasible recovery paths.

[0142] Specifically, the set of nodes in the affected area is determined, and the power supply interruption point on the boundary of the area is taken as the starting point of the feasible recovery path; nodes with power supply capability are selected from the candidate recovery node set as the ending point of the feasible recovery path; and multiple feasible recovery paths are constructed from the starting point to the ending point of the path.

[0143] For each feasible recovery path, the comprehensive recovery cost function of the path is calculated based on the path's multidimensional indicators. The network disturbance impact value is used to generate a list of multiple sorted feasible recovery paths, and the final path comprehensive score is calculated. .

[0144] Step 308: Construct a recovery strategy optimization model and output the optimal recovery path and switching node action sequence required to recover the affected area of ​​the fault.

[0145] Specifically, a recovery strategy optimization model is constructed based on deep reinforcement learning. The current network state, the multi-dimensional indicators corresponding to each feasible recovery path, the comprehensive recovery cost, the network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model. The output is the optimal recovery path and the sequence of switching node actions required to recover the affected area of ​​the fault.

[0146] Step 310: Based on the optimal recovery path and the switching node action sequence, perform the operation of disconnecting the faulty branch and switching the power supply path.

[0147] Specifically, based on the location of the faulty node or faulty segment, the faulty branch is disconnected to isolate the fault source; and the current running status of the optimal recovery path node is queried to check whether the target node of the recovery path is in an operable state.

[0148] Then iterate through the intermediate nodes and switch device numbers in the selected optimal recovery path, for each node... circuit breakers The system executes a closing command to control the load transfer switch connected to downstream nodes in the fault area, switching the node from the original fault power supply path to the restored power supply path. If microgrids, energy storage systems, or distributed power sources are involved in the restoration, synchronous grid connection operations must be performed, and soft-start current limiting logic must be set to prevent current surges after power restoration.

[0149] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0150] Based on the same inventive concept, this application also provides an adaptive recovery device for low-voltage power grid self-healing control, used to implement the adaptive recovery method for low-voltage power grid self-healing control described above. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the adaptive recovery device for low-voltage power grid self-healing control provided below can be found in the limitations of the adaptive recovery method for low-voltage power grid self-healing control described above, and will not be repeated here.

[0151] In one exemplary embodiment, such as Figure 4 As shown, an adaptive recovery device 400 for self-healing control of a low-voltage power grid is provided, including: a detection and judgment module 402, a fault determination module 404, an evaluation and scoring module 406, a modeling and evaluation module 408, an adaptive path decision module 410, and an execution control module 412, wherein:

[0152] The detection and judgment module 402 is used to monitor the operating status of each node and branch in the low-voltage power distribution network in real time. If a node is detected to have abnormal voltage, abnormal current, or communication failure, the corresponding node is marked as a suspected fault node.

[0153] The fault determination module 404 is used to determine the corresponding abnormal impact range based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, tracing back the corresponding power supply path upwards and searching the corresponding branch load downwards, and using the breadth-first traversal algorithm or the depth-first traversal algorithm; and determining the set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure or abnormal parameters as the fault impact area.

[0154] The detection and judgment module 402 is also used to determine the abnormal node as the fault node if there is only one abnormal node in the fault-affected area; and to determine the branch between the last normal node and the first abnormal node as the fault segment if there are multiple consecutive abnormal nodes in the fault-affected area.

[0155] The evaluation and scoring module 406 is used to score the health of all controllable nodes in the low-voltage power distribution network through a node health scoring model, and to select a set of candidate recovery nodes whose health meets the preset requirements.

[0156] The modeling and evaluation module 408 is used to determine the power supply interruption point on the boundary of the fault-affected area as the starting point of the path, select the node with power supply capability from the candidate recovery node set as the ending point of the path, construct multiple feasible recovery paths from the starting point to the ending point of the path based on the topology diagram, calculate the comprehensive recovery cost of the feasible recovery path based on the multi-dimensional index of the feasible recovery path, simulate the impact of the feasible recovery path on the low-voltage distribution network after its access, and determine the network disturbance impact value of each feasible recovery path.

[0157] The adaptive path decision module 410 is used to take the current network status, the multi-dimensional indicators corresponding to each feasible recovery path, the comprehensive recovery cost, the network disturbance impact value and historical recovery data as input to the recovery strategy optimization model, and output the optimal recovery path and the action sequence of switching nodes required to recover the fault-affected area.

[0158] The execution control module 412 is used to perform corresponding recovery operations based on the optimal recovery path and the action sequence of the switching node.

[0159] In one embodiment, the execution control module is further configured to perform corresponding recovery operations based on the optimal recovery path and the switching node action sequence, including: performing a disconnection operation of the fault branch based on the located fault node or fault segment to isolate the fault source; and performing a switching operation of the power supply path according to the optimal recovery path and the switching node action sequence.

[0160] In one embodiment, the adaptive path decision module is further configured to select a suboptimal path from multiple feasible recovery paths to trigger adaptive reconstruction if the recovery of the affected area fails or the operating parameters deviate from the expected performance, and to iteratively correct the control strategy until the recovery is successful.

[0161] In one embodiment, the detection and judgment module is further configured to determine the branch between the last normal node and the first abnormal node as a fault segment, including: sorting the nodes in the fault-affected area according to the topological connection order and determining the power supply conduction relationship between the nodes; according to the power supply conduction relationship, identifying the first abnormal node at the forefront of the continuous abnormal node sequence and the last normal node upstream of the first abnormal node; and determining the physical connection branch between the last normal node and the first abnormal node as a fault segment.

[0162] In one embodiment, the evaluation and scoring module is further used to score the health of all controllable nodes in the low-voltage distribution network using a node health scoring model, and to select a set of candidate recovery nodes whose health meets preset requirements. This includes: weighting and summing the controllable nodes based on their electrical operation stability, communication quality, historical fault frequency, and topological location to obtain the corresponding health score of the controllable nodes; selecting nodes with health scores higher than the health threshold to initially include them in the candidate recovery node set; if there are multiple controllable nodes in the candidate recovery node set whose health score differences are within a preset range, then a secondary sorting and selection is performed based on the load capacity and reserve capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

[0163] In one embodiment, the modeling and evaluation module is further used to calculate the comprehensive recovery cost of a feasible recovery path based on the multi-dimensional indicators, including the path's load-carrying capacity, communication delay, power balancing capability, and number of switching operations. This includes: normalizing the multi-dimensional indicators and configuring corresponding weight coefficients for each normalized multi-dimensional indicator; weighting and summing the multi-dimensional indicators according to the weight coefficients to obtain the comprehensive recovery cost of the feasible recovery path; and using the comprehensive recovery cost to quantify the economic cost and operational difficulty of the feasible recovery path.

[0164] Each module in the aforementioned adaptive recovery device for low-voltage power grid self-healing control can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0165] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements an adaptive recovery method for self-healing control of a low-voltage power grid.

[0166] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0167] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0168] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0169] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0170] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0171] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0172] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0173] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0174] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0175] In one embodiment, when the processor executes the computer program, it further performs the following steps: based on the optimal recovery path and the switching node action sequence, it performs a corresponding recovery operation, including: based on the located fault node or fault segment, it performs a fault branch disconnection operation to isolate the fault source; and according to the optimal recovery path and the switching node action sequence, it performs a power supply path switching operation.

[0176] In one embodiment, when the processor executes the computer program, it also performs the following steps: if the recovery of the affected area fails or the operating parameters deviate from the expected performance, a suboptimal path is selected from multiple feasible recovery paths to trigger adaptive reconstruction, and the control strategy is iteratively corrected until the recovery is successful.

[0177] In one embodiment, when the processor executes the computer program, it further implements the following steps: determining the branch between the last normal node and the first abnormal node as a fault segment, including: sorting the nodes in the fault-affected area according to the topological connection order and determining the power supply conduction relationship between the nodes; identifying the first abnormal node at the forefront of the continuous abnormal node sequence and the last normal node upstream of the first abnormal node according to the power supply conduction relationship; and determining the physical connection branch between the last normal node and the first abnormal node as a fault segment.

[0178] In one embodiment, when the processor executes the computer program, it further performs the following steps: scoring the health of all controllable nodes in the low-voltage distribution network using a node health scoring model, and selecting a set of candidate recovery nodes whose health meets preset requirements, including: performing a weighted summation based on the electrical operation stability, communication quality, historical fault frequency, and topological location of the controllable nodes to obtain the corresponding health score of the controllable nodes, and initially including nodes with health scores higher than the health threshold into the candidate recovery node set; if there are multiple controllable nodes in the candidate recovery node set whose health score differences are within a preset range, then performing a secondary sorting and selection based on the load capacity and reserve capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

[0179] In one embodiment, when the processor executes the computer program, it further implements the following steps: multidimensional indicators include path load capacity, communication delay, power balancing capability, and number of switching operations; based on the multidimensional indicators of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, including: normalizing the multidimensional indicators respectively, and configuring a corresponding weight coefficient for each normalized multidimensional indicator; weighting and summing the multidimensional indicators according to the weight coefficients to obtain the comprehensive recovery cost of the feasible recovery path; the comprehensive recovery cost is used to quantify the economic cost and operational difficulty of the feasible recovery path.

[0180] The implementation principle and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.

[0181] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0182] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0183] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0184] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0185] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0186] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0187] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0188] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0189] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: based on the optimal recovery path and the switching node action sequence, it performs corresponding recovery operations, including: performing a disconnection operation of the fault branch based on the located fault node or fault segment to isolate the fault source; and performing a switching operation of the power supply path according to the optimal recovery path and the switching node action sequence.

[0190] In one embodiment, when the computer program is executed by the processor, it also performs the following steps: if the recovery of the affected area fails or the operating parameters deviate from the expected performance, a suboptimal path is selected from multiple feasible recovery paths to trigger adaptive reconstruction, and the control strategy is iteratively corrected until the recovery is successful.

[0191] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: identifying the branch between the last normal node and the first abnormal node as a fault segment, including: sorting the nodes in the fault-affected area according to the topological connection order and determining the power supply conduction relationship between the nodes; identifying the first abnormal node at the forefront of the continuous abnormal node sequence and the last normal node upstream of the first abnormal node according to the power supply conduction relationship; and identifying the physical connection branch between the last normal node and the first abnormal node as a fault segment.

[0192] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: scoring the health of all controllable nodes in the low-voltage distribution network using a node health scoring model, and selecting a set of candidate recovery nodes whose health meets preset requirements, including: performing a weighted summation based on the electrical operation stability, communication quality, historical fault frequency, and topological location of the controllable nodes to obtain the corresponding health score of the controllable nodes, and initially including nodes with health scores higher than the health threshold into the candidate recovery node set; if there are multiple controllable nodes in the candidate recovery node set whose health score differences are within a preset range, then performing a secondary sorting and selection based on the load capacity and reserve capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

[0193] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: multidimensional indicators include path load capacity, communication delay, power balancing capability, and number of switching operations; based on the multidimensional indicators of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, including: normalizing the multidimensional indicators respectively, and configuring a corresponding weight coefficient for each normalized multidimensional indicator; weighting and summing the multidimensional indicators according to the weight coefficients to obtain the comprehensive recovery cost of the feasible recovery path; the comprehensive recovery cost is used to quantify the economic cost and operational difficulty of the feasible recovery path.

[0194] The implementation principle and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.

[0195] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0196] Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node.

[0197] Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using the breadth-first traversal algorithm or the depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameter state is determined as the fault impact area.

[0198] If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is regarded as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is determined as the fault segment.

[0199] The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements.

[0200] The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. The node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Multiple feasible recovery paths from the starting point to the ending point are constructed based on the topology diagram. Based on the multi-dimensional index of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, and the impact of the feasible recovery path on the low-voltage distribution network after its access is simulated. The network disturbance impact value of each feasible recovery path is determined.

[0201] The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to restore the fault-affected area.

[0202] Based on the optimal recovery path and the sequence of actions for switching nodes, the corresponding recovery operations are executed.

[0203] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: based on the optimal recovery path and the switching node action sequence, it performs corresponding recovery operations, including: performing a disconnection operation of the fault branch based on the located fault node or fault segment to isolate the fault source; and performing a switching operation of the power supply path according to the optimal recovery path and the switching node action sequence.

[0204] In one embodiment, when the computer program is executed by the processor, it also performs the following steps: if the recovery of the affected area fails or the operating parameters deviate from the expected performance, a suboptimal path is selected from multiple feasible recovery paths to trigger adaptive reconstruction, and the control strategy is iteratively corrected until the recovery is successful.

[0205] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: identifying the branch between the last normal node and the first abnormal node as a fault segment, including: sorting the nodes in the fault-affected area according to the topological connection order and determining the power supply conduction relationship between the nodes; identifying the first abnormal node at the forefront of the continuous abnormal node sequence and the last normal node upstream of the first abnormal node according to the power supply conduction relationship; and identifying the physical connection branch between the last normal node and the first abnormal node as a fault segment.

[0206] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: scoring the health of all controllable nodes in the low-voltage distribution network using a node health scoring model, and selecting a set of candidate recovery nodes whose health meets preset requirements, including: performing a weighted summation based on the electrical operation stability, communication quality, historical fault frequency, and topological location of the controllable nodes to obtain the corresponding health score of the controllable nodes, and initially including nodes with health scores higher than the health threshold into the candidate recovery node set; if there are multiple controllable nodes in the candidate recovery node set whose health score differences are within a preset range, then performing a secondary sorting and selection based on the load capacity and reserve capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

[0207] In one embodiment, when the computer program is executed by the processor, it further implements the following steps: multidimensional indicators include path load capacity, communication delay, power balancing capability, and number of switching operations; based on the multidimensional indicators of the feasible recovery path, the comprehensive recovery cost of the feasible recovery path is calculated, including: normalizing the multidimensional indicators respectively, and configuring a corresponding weight coefficient for each normalized multidimensional indicator; weighting and summing the multidimensional indicators according to the weight coefficients to obtain the comprehensive recovery cost of the feasible recovery path; the comprehensive recovery cost is used to quantify the economic cost and operational difficulty of the feasible recovery path.

[0208] The implementation principle and technical effects of the above embodiments are similar to those of the above method embodiments, and will not be repeated here.

[0209] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0210] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0211] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0212] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. An adaptive recovery method for self-healing control of low-voltage power grids, characterized in that, The method includes: Real-time monitoring of the operating status of each node and branch in the low-voltage power distribution network; if abnormal voltage, abnormal current or communication is detected at a node, the corresponding node is marked as a suspected fault node. Based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, the corresponding power supply path is traced back upwards and the corresponding branch load is searched downwards. The corresponding abnormal impact range is delineated using a breadth-first traversal algorithm or a depth-first traversal algorithm. The set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameters is determined as the fault impact area. If there is only one abnormal node in the affected area of ​​the fault, the abnormal node is designated as the fault node; if there are multiple consecutive abnormal nodes in the affected area of ​​the fault, the branch between the last normal node and the first abnormal node is designated as the fault segment. The node health scoring model is used to score the health of all controllable nodes in the low-voltage power distribution network and select a set of candidate recovery nodes whose health meets the preset requirements. The power supply interruption point on the boundary of the fault-affected area is determined as the starting point of the path. A node with power supply capability is selected from the candidate recovery node set as the ending point of the path. Based on the topology diagram, multiple feasible recovery paths are constructed from the starting point to the ending point of the path. Based on the multi-dimensional indicators of the feasible recovery paths, the comprehensive recovery cost of the feasible recovery paths is calculated, and the impact of the feasible recovery paths on the low-voltage distribution network after access is simulated to determine the network disturbance impact value of each feasible recovery path. The current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value, and historical recovery data are used as inputs to the recovery strategy optimization model, and the output is the optimal recovery path and switching node action sequence required to recover the fault-affected area. Based on the optimal recovery path and the switching node action sequence, the corresponding recovery operation is performed.

2. The method according to claim 1, characterized in that, The step of performing corresponding recovery operations based on the optimal recovery path and the switching node action sequence includes: Based on the location of the fault node or fault segment, the fault branch is disconnected to isolate the fault source; according to the optimal recovery path and the switching node action sequence, the power supply path is switched.

3. The method according to claim 2, characterized in that, The method further includes: If the affected area fails to recover or the operating parameters deviate from the expected performance, a suboptimal path is selected from multiple feasible recovery paths to trigger adaptive reconstruction, and the control strategy is iteratively corrected until recovery is successful.

4. The method according to claim 1, characterized in that, The step of identifying the branch between the last normal node and the first abnormal node as the faulty segment includes: The nodes within the affected area of ​​the fault are sorted according to their topological connection order to determine the power supply conduction relationship between the nodes; Based on the power supply conduction relationship, identify the first abnormal node at the forefront of the continuous abnormal node sequence, and the last normal node upstream of the first abnormal node; The physical connection branch between the last normal node and the first abnormal node is identified as the faulty segment.

5. The method according to claim 1, characterized in that, The node health scoring model is used to score the health of all controllable nodes in the low-voltage distribution network, and a set of candidate recovery nodes whose health meets preset requirements is selected, including: The health score of a controllable node is obtained by weighted summation based on its electrical operation stability, communication quality, historical fault frequency, and topological location. Nodes with health scores higher than the health threshold are initially included in the candidate recovery node set. If there are multiple controllable nodes in the candidate recovery node set whose health score differences are within a preset range, then a secondary sorting and filtering is performed based on the load capacity and backup capacity parameters of the controllable nodes to obtain the final candidate recovery node set.

6. The method according to claim 1, characterized in that, The multidimensional indicators include path load carrying capacity, communication delay, power balancing capability, and number of switching operations; The calculation of the comprehensive recovery cost of the feasible recovery path based on the multidimensional indicators of the feasible recovery path includes: The multidimensional indicators are normalized respectively, and a corresponding weight coefficient is assigned to each normalized multidimensional indicator. The multidimensional indicators are weighted and summed according to the weighting coefficients to obtain the comprehensive recovery cost of the feasible recovery path; the comprehensive recovery cost is used to quantify the economic cost and operational difficulty of the feasible recovery path.

7. An adaptive recovery device for self-healing control of a low-voltage power grid, characterized in that, The device includes: The detection and judgment module is used to monitor the operating status of each node and branch in the low-voltage power distribution network in real time. If a node is found to have abnormal voltage, abnormal current, or communication failure, the corresponding node is marked as a suspected fault node. The fault determination module is used to determine the corresponding abnormal impact range based on the topology diagram of the low-voltage power distribution network, starting from the suspected fault node, tracing back the corresponding power supply path upwards and searching the corresponding branch loads downwards, and using a breadth-first traversal algorithm or a depth-first traversal algorithm; the set of nodes that are topologically connected to the suspected fault node and are in a state of power disconnection, communication failure, or abnormal parameters is determined as the fault impact area. The detection and judgment module is further configured to: if there is only one abnormal node in the fault-affected area, then regard the abnormal node as a fault node; if there are multiple consecutive abnormal nodes in the fault-affected area, then determine the branch between the last normal node and the first abnormal node as a fault segment. The evaluation and scoring module is used to score the health of all controllable nodes in the low-voltage distribution network using a node health scoring model, and to select a set of candidate recovery nodes whose health meets the preset requirements. The modeling and evaluation module is used to determine the power supply interruption point on the boundary of the fault-affected area as the starting point of the path, select a node with power supply capability from the candidate recovery node set as the ending point of the path, construct multiple feasible recovery paths from the starting point to the ending point of the path based on the topology diagram, calculate the comprehensive recovery cost of the feasible recovery path based on the multi-dimensional index of the feasible recovery path, simulate the impact of the feasible recovery path on the low-voltage distribution network after its access, and determine the network disturbance impact value of each feasible recovery path. The adaptive path decision module is used to take the current network status, multi-dimensional indicators corresponding to each feasible recovery path, comprehensive recovery cost, network disturbance impact value and historical recovery data as input to the recovery strategy optimization model, and output the optimal recovery path and switching node action sequence required to recover the fault-affected area. The execution control module is used to perform corresponding recovery operations based on the optimal recovery path and the action sequence of the switching node.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images